Ghana - Accra District Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 13287 PERFORMANCE AUDIT REPORT GHANA ACCRA DISTRICT REHABILITATION PROJECT (CREDIT 1564-GH) JUNE 29, 1994 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. CURRENCY EQUIVALENTS (annual averages) Currency Unit = Cedi (C) EXCHANGE RATE (CEDI TO US$) 1985 (SAR) US$1.00 = 0.0184 1986 US$1.00 = 0.0112 1987 US$1.00 = 0.0061 1988 US$1.00 = 0.0049 1989 US$1.00 = 0.0037 1990 US$1.00 = 0.0031 1991 US$1.00 = 0.0028 1992 US$1.00 = 0.0025 ABBREVIATIONS AND ACRONYMS ACC Accra City Council ADRP Accra District Rehabilitation Project ADRU Accra District Roads Unit AMA Accra Metropolitan Authority AMRD Accra Metropolitan Roads Department DUR Department of Urban Roads ERP Economic Recovery Program GHA Ghana Highway Authority GOG Government of Ghana IDA International Development Association LVB Land Valuation Board LVD Land Valuation Department M&E Monitoring and Evaluation MIS Management Information System MLG Ministry of Local Government MOF Ministry of Finance MRH Ministry of Roads and Highways MWH Ministry of Works and Housing NGO Non-governmental Organization OED Operations Evaluation Department PAR Performance Audit Report PCR Project Completion Report PWP Priority Works Project SAR Staff Appraisal Report TA Technical Assistance TSC Technical Services Center VIP Ventilated Improved Pit FISCAL YEAR January I - December 31 FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A Office of Director-General Operations Evaluation June 29, 1994 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on Ghana Accra District Rehabilitation Project (Credit 1564-GH) Attached is the "Performance Audit Report on Ghana - Accra District Rehabilitation Project (Credit 1564-GH)" prepared by the Operations Evaluation Department. The audited project was closely linked with the Economic Recovery Program of the Government of Ghana, launched in 1983, to remove transport bottlenecks, upgrade basic infrastructure, mobilize revenues, and assist in municipal capacity building in Accra. The project was rather complex, and with its far reaching institutional goals, was a new challenge both for the borrower and the Bank as it was the first urban development project ever in Ghana funded by the Bank and was implemented under difficult country conditions. The project achieved most of its physical targets and managed to establish property valuation and tax collection systems in Accra and Tema. The project, however, fell short in making the systems fully operational thus leaving revenue mobilization efforts incomplete. This, and some institutional weaknesses negate the sustained maintenance of municipal services and thus full realization of project benefits. The project formed the basis for preparation of two follow-on operations, one of which was recently completed; the other is under implementation. On balance, the Audit rates the project's outcome as satisfactory, its institutional development as modest, and the sustainability of benefits as uncertain. These audit ratings confirm the prior PCR assessments. Robert Picciotto by H. Eberhard K6pp Attachment This document has a restricted distribution and my be used by recipients only in the performance of their official duties. Its contents my not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY PERFORMANCE AUDIT REPORT GHANA ACCRA DISTRICT REHABILITATION PROJECT (Credit 1564-GII) TABLE OF CONTENTS Page No. PREFACE ........................................... i BASIC DATA S EET ...... ............................. ii EVALUATION SUMMARY................................ vii 1. BACKGROUND ................................... 1 A. Ghana's Macroeconomic Setting .......................... 1 B. Urban Sector Setting ................................. 1 C. Bank's Roles ...................................... 2 2. THE PROJECT AND ITS IMPLEMENTATION EXPERIENCE ...... 3 A. Objectives and Description of the Project .................... . 3 B. Project Processing ................................... 4 C. Implementation Experience.............................. 4 3. PROJECT RESULTS .................................. 5 A. Physical Results .................................... 5 B. Project Costs ...................................... 7 C. Institutional Results .................................. 8 D. Studies .......................................... 9 4. BORROWER AND BANK PERFORMANCE .................. 10 A. Borrower Performance ................................ 10 B. Bank Performance ................................... 10 C. Compliance with Covenants and Directives ................... . 11 This report was prepared by Tauno Skytta (Principal Evaluation Officer) who audited the project in October 1993, with secretarial assistance from Ms. E. Tweddle. 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. Table of Contents (Cont.) Page No. 5. PERFORMANCE ASSESSMENT AND ISSUES .. 11 A. Project Performance Assessment ...... 11 B. Issues .. ........................................... 11 6. CONCLUSIONS AND LESSONS LEARNED ................. 12 A. Conclusions ....................................... 12 B. Lessons of Experience ................................ 12 i PERFORMANCE AUDIT REPORT GHANA ACCRA DISTRICT REHABILITATION PROJECT (CREDIT 1564-GII) PREFACE 1. This is a Performance Audit Report (PAR) on the Accra District Rehabilitation Project, approved by the Board of Directors on March 3, 1985 and financed by Credit 1564-GH in the amount of SDR 22.5 million (US$22.0 million equivalent). Over 95% of the credit was disbursed by the closing date of June 30, 1992, and the remaining balance of SDR 0.9 million (US$1.2 million equivalent) was canceled. 2. The PAR consists of an Evaluation Summary and a Performance Audit Report prepared by the Operations Evaluation Department (OED). A Project Completion Report (PCR), prepared by the Infrastructure Operations Division of the Western Africa Department, Africa Regional Office, was previously submitted to the Board as Report No. 12225, dated July 31, 1993. 3. The PCR provides an interesting discussion on the local capacity building process and related implementation problems. The PAR elaborates on particular aspects such as institution and operational capacity building, and related TA activities. The PAR is based on the PCR, the President's Report, project legal documents, the Staff Appraisal Report (SAR), review of project files, other relevant background material and studies, and discussions with Bank operational staff. The PAR also draws on field observations and on interviews with officials of the Government of Ghana (GOG) in Accra during the audit mission in October 1993. The courtesy and cooperation of these officials is gratefully acknowledged. 4. The Audit endorses PCR findings in most respects and expands on the lessons of experience. It puts particular emphasis on the practical arrangements for running effective TA support. The Audit also highlights the importance of a careful retrospective review of capacity building efforts after a series of operations. 5. Following standard OED procedures, the draft PAR was sent to the Borrower for comments; however no comments were received.  ii PERFORMANCE AUDIT REPORT GHANA ACCRA DISTRICT REHABILTATION PROJECT (Credit 1564-GII) BASIC DATA SHEET KEY PROJECT DATA (US$ million) Actual as % Appraisal of Appraisal Itwm Estimate Actual Estimate Total Project Cost 26.33 28.50 108.00 Loan Amount 22.00 27.71 126.00 Economic Rate of Return % - Road Rehabilitation component 83.0 84.0 Institutional Performance: Modest CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS Appraisal Estimate (US$M) 22.00 Actual (US$M) 27.71 Actual as % of Appraisal 126.00 Date of Final Disbursement: October 31, 1992 ii PRQJ DATES Item Date Date Planned Actual Identification 08/80 Preparation 12/83 Appraisal 01/78 05/84 Negotiations 11/84 02/85 Board Approval 02/85 03/21/85 Signature 03/85 03/29/85 Effectiveness 07/85 09/04/85 Closing Date 06/30/91 06/30/92 Completion Date 12/31/90 12/31/92 STAFF INPUTS (staff-weeks) EIME2 EYMI EM EM Elf EY2 EFY1 FY12 E Y - Y21 . FY93 TItal Preappraisal 23.2 3.4 34.2 - - - - - - - - - 60.7 Appraisal - - 29.6 46.4 - - - - - - - - 75.9 Negotiation - - - 10.2 - - - - - - - - 10.2 Supervision - - - 9.3 25.0 14.5 6.9 15.6 11.3 10.6 11.1 4.3 108.5 Other 1.2 .3 1.2 9.3 .9 .1 - - - - - - 13.5 1 This is the date of identification of the revised project on which Credit 1564-GH is based. iv MISSION DAT Date No. of No. of Month/Year _Day_ Persons Identification 12/75 45 3 Preparation 05/76 14 2 Preappraisal 06/77 90 6 Appraisal 10/77 105 6 Reconnaisance 02/80 2 2 Preparation I 07/80 5 2 Preparation II 08/80 5 2 Preparation III 02/81 5 1 Preparation IV 05/81 10 2 Preparation V 09/81 10 2 Preappraisal 10/81 30 2 Preparation VI 03/84 2.5 1 Appraisal 06/84 75 5 Post Negotiation 03/85 2.5 1 Supervision I 12/84 2.5 1 Supervision II 06/85 10 2 Supervision III 09/85 37.5 3 Supervision IV 10/85 20 2 Supervision V 11/85 2.5 1 Supervision VI 03/86 15 3 Supervision VII 09/86 20 2 Supervision VIII 06/87 5 1 Supervision IX 09/87 2.5 1 Supervision X 11/87 10 2 Supervision XI 02/88 10 2 Supervision XII 07/88 15 2 Supervision XIII 11/88 15 2 Supervision XIV 03/89 15 2 Supervision XV 07/89 2.5 1 Supervision XVI 07/90 15 2 Supervision XVII 11/90 15 3 Supervision XVIII 01/91 15 3 Supervision XIX 05/91 10 2 Supervision XX 10/91 5 1 Supervision XXI 02/92 5 1 Supervision XXII 02/92 10 2 Supervision XXIII 06/92 2.5 2 Total 661 V OTHER PROJECT DATA Borrower: Republic of Ghana Executing Agencies: Ministry of Works and Housing, Ministry of Local Government, Accra City Council. Vi MISSIONS State of Project Cycle Monthl go. of WYe in cialsaion Pert. I Of year Staff Field srsted Status Or1S. Proj. Identification 12/75 3 9.0 UP, COp Preparation 05/76 2 2.0 up. EG Preappraisal 06/77 6 18.0 FA,C,PAS,MO,CoM Appraisal 10/77 6 21.0 FA,EC,PS,EM,CON Subtotal 50.0 Revised Proj. Recounaiseance 02/80 1 0.3 Eo Preparation 1 02/80 2 1.0 EsG,FA Preparation 2 06/60 2 1.0 EG,FA Preparation 3 02/81 1 1.0 7A. Preparation 4 05/81 3 7.5 YA,WV,COM Preparation 5 09/91 2 2.0 EsG,FA Preappraisal 10/81 2 6.0 Ex,FA Preparation 6 03/84 1 0.5 UP Appraisal 06/84 5 15.0 EGF,A,IC,C,VAL Post Negotiation 03/85 1 0.5 EC Subtotal 34.6 Supervision 1 12/84 1 0.5 ING IA Supervision 2 06/85 2 2.0 E0,PA 1.2 Supervision 3 09/85 3 7.5 UP,IA,CO 1,2,1,1 Supervision 4 10/85 2 4.0 UP,A 1,2,1,1 Supervision 5 11/65 1 0.5 UP,C=g 1,2,1,1 Supervision 6 05/84 3 3.0 UP,C= 1,2,1,1 Supervision 7 09/66 2 4.0 UP,Cg 1,2,1,1 Supervision S 06/87 1 1.0 UP 2,1.1,1 Supervision 9 09/67 1 0.5 up Supervision 10 11/87 2 2.0 Wrzx VIA Supervision 11 0.1s8 2 2.0 UP,CCg 2,2,1,2 Supervision 12 07/68 2 3.0 UP,CO 2,2,1,2 Supervision 13 11/8 2 3.0 UP 1,2,1,1 Supervision 14 03/9 2 3.0 UP,FA 2,1,,2,1 2,1,1,1,2,2 Supervision 15 05/69 1 0.5 UP 1,1,1,2,1 1,1,1,1,1,1 Supervision 16 07/90 2 3.0 UP,FA 1,1 1 1,1,1,1,1,1 Supervision 17 11/90 3 3.0 =G,FA.Cag 1,1, 1 1,1,1,1,1,1 Supervision 16 01/91 3 3.0 wG,IA,C 1,1 1,1 1,1,1,1,1,1 Supervision 19 05/91 2 2.0 =.sA 1,1,1,1,1 1,1.1,11 Supervision 20 10/91 I 1.0 EG,IrA 1.1,1,2,1 1,1,1,1,1,1 Supervision 21 02/92 1 1 3C Supervision 22 02/92 2 2 OG,sA Supervision 23 06/92 2 0.5 IA Total 52.0 Key: CON - Consultant PHS - Public Health Specialist EC - Economist UP - Urban Planner ENH - Engineer VAL - Valuer FA - Financial Analyst N/A - Not available  vii PERFORMANCE AUDIT REPORT GHANA ACCRA DISTRICT REHABILITATION PROJECT (CREDIT 1564-GH) EVALUATION SUMMARY Background 1. The economic downturn of the 1970s slowed down the movement of import/export and early 1980s hit Ghana hard, leading to items, thereby affecting the Government's a decline in the standard of living for the economic recovery efforts. The tax base vast majority of Ghanaians. In 1983, the had eroded and collection declined. Government of Ghana (GOG) launched an Furthermore, the lack of trained manpower Economic Recovery Program (ERP) which was an acute problem in all ministries and the Bank supported through several IDA agencies involved in the urban sector (paras. credits. In 1987, the Government, together 1.03-1.06). with the donor community, developed a program (Second Phase of ERP) to mitigate 3. The Bank supported the GOG in its the social costs of adjustment caused by the efforts to solve the infrastructure problems ERP. A set of short-term interventions of Accra through the Accra District were identified, among them a 2-year Rehabilitation Project (ADRP), the subject program of priority labor-intensive urban of this evaluation. As part of the ERP, the works (paras. 1.01-1.02). Bank continued its support by appraising another operation in 1987, the Priority 2. Accra's population, which stood at Works Program (Cr. 1874-GH), which was 950,000 in 1984, was estimated to double completed recently. To consolidate its before the year 2000. To better meet the support to the sector further, the next Bank challenges of this projected urban growth, intervention, the Urban H Project (Cr. 2157- the Government set out to reorganize urban GH), was appraised in 1990, and is now institutions through a decentralization under implementation (para. 1.07). program designed to be more responsive to local needs. Bank Group assistance was The Project and Its Implementation sought to support this decentralization effort. Experience Also, the infrastructure of Accra, where more than 50% of population lived in areas 4. The ADRP aimed at (i) removing with poor or no urban services, had severely critical transport bottlenecks in Accra, (ii) deteriorated over the years because of mobilizing municipal revenues, (iii) inadequate maintenance. Traffic bottlenecks demonstrating cost effective basic viii infrastructure improvements, and (iv) 7. Costs of some main components rose strengthening capacities to maintain from 14% for road rehabilitation and municipal services. The project involved a maintenance, to over 40% for basic multitude of organizations at the central and infrastructure upgrading and municipal local municipal levels and, as such, required revenue mobilization, as compared to careful coordination during implementation appraisal estimates. This was mainly due to (paras. 2.01-2.05). the increased scope of these components. Because of a significant contingency 5. In general, project preparation was provision in the appraisal estimates, the closely linked with the ERP and was based overall project costs went up by only 8% in on earlier work related to other ERP US dollar terms. At the closing of the activities. Components were of adequate credit in June 1992, a commitment of technical quality. Due to the large number US$4.2 million for additional vehicles and of entities involved, the borrower ownership equipment was outstanding but it was was perhaps not sufficiently promoted disbursed within six months (paras. 3.02- during the preparation process. The 3.03). capacity building efforts were supported through a number of twinning arrangements 8. The coordination and overall and relevant on-the-job and overseas administration of this multi-faceted project training. As the introduced concept of was entrusted to the TSC which was urban management was new to local established within the MWH (in 1983) for administrations, several key senior officials specific development purposes in the urban were selected to participate in overseas sector. This proved to be an effective training (paras. 2.06-2.09). arrangement at least in the short term. The capacity of the AMRD, and DUR was Project Results developed to carry out O&M of essential infrastructure, and planning and 6. The project achieved most of its implementation of major capital works in physical objectives. The main component Accra. Especially AMRD is still weak and was the construction of a 14 km segment of requires both financial and manpower the southern carriageway of the Accra outer support from the central level. The upkeep ring road. The scope of this critical of vehicles and equipment is still a chronic component was increased from appraisal problem and causes some concern as far as estimate and, as a result, rehabilitation of sustained O&M activities are concerned key trunk roads was dropped. The drainage awas. 3.04-3.06). rehabilitation program was completed as scheduled, including procurement of 9. Some 120 operational staff of LVB vehicles and equipment for continued received on-the-job training in property maintenance of roads and drains. Basic valuation technics. Revaluation of over infrastructure of East Maamobi was 80,000 properties in Accra was completed upgraded as per the program stipulated in and was expanded to Tema, covering about the SAR. The slightly expanded scope was 30,000 properties. Staff of AMA was financed with savings from changes in the trained in municipal finance. Both activities value of the Cedi (para. 3. 01). are now well established but are falling ix behind again due to problems in Institutional achievements were promising as transportation to carry out field work, and far as the establishment of the property some computer hardware constraints. valuation system and improvements to the Although an improved property tax system tax collection are concerned. The project and collection initially increased AMA's failed, however, to make these new systems revenues significantly, it is currently facing fully operational, thus leaving revenue major problem with defaulters. Private mobilization efforts incomplete. This casts lawyers have now been hired to process the some doubt on AMA's capacity to handle outstanding claims (paras. 3.07-3.11). the maintenance of road and drainage networks in the long-run, as well as on 10. The project produced traffic studies sustained operation of other municipal covering road network improvements and services. On balance, the Audit rates the traffic management not only in Accra but project's outcome as satisfactory, its also in Secondi-Takoradi and Kumasi. institutional development as modest, and the Results of the studies were used in the sustainability of its benefits as uncertain. preparation of two follow-on projects, one The PCR-based ratings are identical (para. of which is now complete and the other S.01). which is under implementation (para. 3.12). 13. In the area of sanitation, the project Borrower and Bank Performance completed its targets but the applied solution did not prove to be the "best practice 11. The performance of the borrower solution" in a densely populated area such as was reasonable despite a delay of one year East Maamobi. Drawing on experience in the completion of the project. The from a sanitation program in Kumasi, a shortage of counterpart funding was an more suitable and affordable solution could occasional problem. The Bank provided be developed. Drainage in many area of exceptionally frequent supervision support, Accra was improved but its maintenance especially in engineering and finance. The seems to fail, thus negating the intended Bank failed, however, to provide support in environmental benefits. The project's the training aspects of the project although institutional achievements were noteworthy this need was specifically stipulated in the but some concern has been raised as to their SAR. The project was in compliance with long-term sustainabiity. This is mainly due all key covenants and did not compromise to perhaps too much emphasis on the any critical OMSs and OPNs (paras. 4.01- strengthening of TSC at the cost of other 4.04). entities who ultimately are responsible for O&M of the municipal services (paras. Performance Assessment and Issues 5.02-5.04). 12. The Audit finds that the ADRP Conclusions and Lessons Learned achieved most of its physical objectives with some minor exceptions only. In addition, 14. The ADRP had ambitious objectives the scope of upgrading the East Maamobi and hence, it became a complex operation. infrastructure was increased, utilizing the Furthermore, it was implemented during a savings realized as the Cedi was devalued, very difficult period in Ghana, but with x appropriate arrangements and dedicated c) In a multi-faceted operation like the work on the part of the implementing ADRP, effective coordination agencies the project was successfully between, and clearly defined roles completed. The Bank invested the required among the various entities are the manpower and expertise in the supervision key to success; of the project. As a result, the development of the urban sector in Ghana was steered in d) The involvement of beneficiaries in the right direction. The lessons from this infrastructure upgrading as carried project and the follow-on operations that out i East Maamobi is important. should be evaluated in due course, will be The project should also include an important in designing future urban effective M&E to fully realize the activities. Key lessons learned from the benefits of lessons learned and ADRP include (paras. 6.01-6.02): enable a correction of the course of action during the implementation as a) It is essential to involve all the appropriate; and participating entities in the project preparation process to the fullest e) After this first project, the Bank has extent, in order to promote sound supported several urban operations in ownership among the project Accra (and other towns in Ghana) stakeholders. This should which followed one another only a particularly apply to those who are few years apart. All of them have ultimately responsible for O&M of (rightly) focused on capacity building the new facilities; of the sector. It would be essential to carry out a careful retrospective b) To make the TA support fully review of the results achieved so far effective, it is of utmost importance before any new operation is to involve the recipient entities in the launched. selection of the candidates for TA personnel, and, together with their twinning partners, define the nature of the required assistance program in full detail; PERFORMANCE AUDIT REPORT GHANA ACCRA DISTRICT REHABILTATION PROJECT (CREDIT 1564-GH) 1. BACKGROUND A. Ghana's Macro-economic Setting' 1.01 Compared to most West African nations, Ghana had traditionally enjoyed a fairly high standard of living, but the economic decline of the 1970s and early 1980s hit Ghana hard. Causes were: a severe drought (1975-77); falling prices of cocoa on the world market with subsequent decline in foreign earnings; large public deficits; a difficult balance of payments situation; and declining investments; to mention a few. As a result, for the vast majority of Ghanaians, real incomes fell substantially. The per capita income was further eroded by high population growth estimated at 2.6% between 1970 and 1984 with subsequent decline in the standard of living. Ghana's large reservoir of skilled manpower was severely affected by a "brain drain" to neighboring countries and overseas. In 1983, the Government of Ghana (GOG) launched an Economic Recovery Program (ERP) which was supported, inter alia, by the Bank through several IDA credits. 1.02 Under the second phase of the Government's ERP (1987-89), some 45,000 persons in the public sector and 20,000 in the public enterprise sector were released from their jobs. In 1987, an interagency workshop was arranged with the Government of Ghana to develop a program to mitigate the social costs of adjustments caused by the ERP. A set of short-term interventions were identified, among them a 2-year program of priority labor-intensive urban works. Through these projects, job opportunities were to be provided for the rural and urban unemployed and underemployed. B. Urban Sector Setting2 1.03 The intercensal urban growth rate (from 1970 to 1984) in Ghana stood at 3.2%. In 1984, the urban population was about 30% of the country's total population, of which one-third was concentrated in the Greater Accra Region, expanding at a rate of about 5% per year. At these This information is summarized from the country background sections of the SARs of Credits 1564-OH, 1874- GH, and 2157-OH. 2 This information is summarized from the urban sector descriptions in the SARs of Credits 1564-OH, 1874-OH, and 2157-OH. 2 rates, Accra's population, which was about 950,000 in 1984, was estimated to double before the year 2000. 1.04 In order to improve the effectiveness of urban sector institutions to better meet the challenges of the projected urban growth, the Government begun an institutional reorganization. Also, a decentralization program, designed to make local governments more autonomous, technically more capable, and more responsive to local needs was launched. To improve the progress of decentralization, the Government requested Bank Group assistance in the planning and implementation of a training program for local authority personnel. 1.05 The condition of the once highly-developed urban infrastructure in Accra had severely deteriorated over the years. The lack of maintenance had taken its toll, especially on the heavily-traveled paved roads. Traffic bottlenecks countrywide seriously affected the implementation of the Government's economic program by slowing down movement of import/export items. Therefore, eliminating the traffic bottlenecks became one of the main objectives of the Government's budgetary applications and the international aid efforts. In Accra, the capacity of national and city governments to maintain existing infrastructure had been exceeded. More than 50% of the city's population lived in areas that suffered from the poor condition (or lack) of roads, drainage, sanitation, water supply, electricity and refuse disposal systems. 1.06 In spite of the Government's efforts to strengthen some sector institutions, all ministries and agencies active in the urban sector suffered from problems related to the lack of trained staff or otherwise inadequate technical capability. In the Accra City Council (ACC), the shortage of key technical and financial personnel was acute, and there was an almost total lack of machinery and vehicles for maintenance and service activities. The situation had been brought about by the general economic decline and ACC's poor financial condition. The tax base in Accra had eroded and collection had declined. To reverse this decline, considerable support to introduce improved accounting, budgeting and control procedures was required. Also, the tax base and fee and rate levels had to be updated and collection performance improved. C. Bank's Role 1.07 The Bank began its support of the efforts of the GOG in addressing some of the basic institutional, financial, and infrastructure problems in Accra through the Accra District Rehabilitation Project (ADRP), the subject of this evaluation (see para. 2.01 below). It was the first Bank urban operation ever in Ghana. As part of the ERP, the Bank continued its support by appraising another operation in 1987, namely the Priority Works Program (Cr. 1874-GH). This project, completed recently, was among the short term projects identified for the second phase of the ERP, see para. 1.02 above. To further consolidate the Bank support to the sector, the next project, Urban II Project (Cr. 2157-GH), was appraised in 1990. It is now under implementation. 3 2. THE PROJECT AND ITS IMPLEMENTATION EXPERIENCE A. Objectives and Description of the Project 2.01 The objectives and components of the ADRP are presented in the PCR in detail. The general nature of the project is as follows: Objectives: the project was designed to; (i) remove transport bottlenecks in Accra which were constraining general economic recovery, (ii) mobilize revenues at the municipal level in Accra, (iii) demonstrate cost-effective improvements to basic infrastructure in a small low-income area of Accra, and (iv) strengthen the capacity of the institutions responsible for infrastructure maintenance in Accra. Components included (i) road and drainage rehabilitation and maintenance, (ii) basic infrastructure upgrading in East Maamobi (about 19,000 people), including roads, drainage, water supply, electricity, sanitation, and garbage disposal, (iii) property revaluation, and (iv) municipal revenue mobilization. 2.02 The ADRP was prepared over a period of three and a half years; nine missions' in all were undertaken from identification to appraisal. The ADRP was an element of the first phase of the ERP and its preparation involved thorough policy discussions with the GOG. The preparation activities engaged a number of organizations at the central and local municipality levels. Details are discussed in paras. 2.03 to 2.05 below. 2.03 The ADRP was rather complex considering its relatively small size, and involved several organizations. Due to this situation, the Ministry of Works and Housing (MWH) established a Technical Services Center (TSC) in 1983 for developing new and innovative approaches to urban problems. It eventually became the responsible entity for coordinating and administering the implementation of the ADRP. TSC was also directly responsible for the implementation of the basic infrastructure upgrading in East Maamobi. 2.04 During project implementation, the ACC was renamed Accra Metropolitan Authority (AMA). In the early years of the project, the Accra District Roads Unit (ADRU) was operating under the control of the Ghana Highway Authority (GHA) but was moved back to AMA in 1989 and renamed Accra Metropolitan Roads Department (AMRD). Both the GHA and the Department of Urban Roads (DUR) under the Ministry of Roads and Highways (MRH) played an important role in the implementation of the project. GHA was responsible for the This also included the Technical Services Center (TSC) which was established to coordinate and administer the project, see para. 10. 4 This does not count the four missions (50 SWs in the field) undertaken prior to the changing of the project concept in 1980. 4 construction of the outer ring road, and DUR provided the training and support in the strengthening of the AMRD; DUR is still in charge of planning and construction of major capital works within the metropolitan area. 2.05 The identification and revaluation of properties in the Accra-Tema area was carried out by the Land Valuation Department (LVD) of the Ministry of Local Government (MLG). At the end of the project, these activities were transferred to the newly established Land Evaluation Board (LVB) under the Ministry of Finance (MOF). B. Project Processing 2.06 The ADRP focused on urban issues that were identified in the government's ERP as needing urgent attention. The preparation of the components was based on earlier work such as sanitation under the German bilateral program, and an IDA-supported port component of the Export Rehabilitation Project (Cr. 1029-GH). Among the priority goals of the project were the capacity building aspects of operational efficiency and resource mobilization, both of which were in agreement with the overall goals of the ERP. In general terms, the Audit finds that the preparation was carried out appropriately and it produced components of adequate technical quality. 2.07 Borrower ownership was perhaps not sufficiently promoted during the preparation process, as the Bank teams had to deal with a large number of entities. Furthermore, these entities changed during the implementation, which made it even more difficult to establish a common ground for understanding the project's goals. The SAR (para. 4.21) specified that project monitoring was to be the responsibility of the Chief Technical Director of the MWH but defines this function only in terms of progress reporting rather than as a dynamic M&E process. On the institutional side it was required that the progress made by the ARDU (later AMRD) be monitored on an annual basis. Overall, it seems that the monitoring process as defined in the SAR served mainly the purpose of the Bank's supervision rather than providing lessons and feedback for a dynamic learning process to support the preparation of follow-on operations. C. Implementation Experience 2.08 As the institutions involved in the ARDP were weak at the time of project appraisal, a program of TA support was designed and included in the project. This covered two main areas, namely (i) property referencing and revaluation, and (ii) municipal revenue mobilization. Two UK twinning partners were engaged to provide the necessary training and assistance in carrying out these activities. Although both activities were carried out reasonably successfully, AMA informed the Audit that the quality of TA (twinning) personnel was less than expected. AMA felt that they should have been given a role in the selection of the candidates, as some were below the requirements of the posts in question. Particularly, the financial advisor did not meet the required qualities and results were much less than expected. 5 2.09 The TA under the project included a substantial provision for training, especially in the property referencing/revaluation and municipal revenue areas. Local staff training was effectively executed through on-the-job exercises and also through strengthening of the Local Government Training School. As the field of urban management was new, many senior managers, including the City Treasurer, were selected for overseas study and training tours. This included arrangements with the twinning partners who in advance carefully assessed the training needs of AMA and others. However, the overseas study tour for four valuators was dropped, and the City Engineer's study tour to the UK was canceled. 3. PROJECT RESULTS A. Physical Results 3.01 The ADRP achieved most of its physical objectives. The following are item-wise observations: * The main component was the completion of a key segment of the outer ring road of Accra (motorway). Work had started earlier but was left unfinished after the completion of basic land clearing and earthwork. The cross section was constructed for a dual carriageway road, but the project completed only the southern carriageway. The GHA informed the Audit that the total length of the completed segment of the motorway is about 14 km compared to 8.5 km as indicated in the SAR. The rehabilitation of four segments of key trunk roads in Accra which were included under this project were dropped. * The drainage rehabilitation program was completed as scheduled. A large number of vehicles and equipment were provided under the project to support the maintenance of the roads and drainage network. They were all procured as per the SAR estimates except for minor changes between various items (see Table 1 below). The Audit notes, however, that the maintenance of the vehicles and equipment seems to be seriously lacking, as a substantial number of the items are no longer roadworthy. Institutional aspects of these operations are discussed in paras. 3.05 and 3.06 below. * Basic infrastructure upgrading in the Nima-Maamobi area, with a population of about 19,000 people, was completed more or less as per the quantities given in the SAR. The work included nearly 4 km of roads with lighting and drains, 4.8 km of water distribution pipelines, 8 public toilets using KVIP technology, and 12 refuse containers. An NGO was engaged to handle community mobilization of the sanitation component. Sanitation activities and their costs are further discussed in para. 5.02 below. Land A ventilated improved pit latrine developed in Kumasi, see "Household Demand for Improved Sanitation Services: A Cane Study of Kumasi, Ohana", UNDP-World Bank Water and Sanitation Program. 6 compensation issues delayed the completion of the Nima-Maamobi component somewhat. Additional rehabilitation/cleaning of drains and installation of latrines were carried out, involving small community based construction groups and funded with savings from changes in the value of the Cedi. * Vehicles and equipment were provided for all the main areas of operation. At the closing of the credit, there was a commitment for procuring additional vehicles and equipment; the total value was recorded as US$4.2 million (see Table 2 below). The PCR does not give the specifics of this commitment and the actual purpose for which the items were procured. The Audit was also unable to gather further clarification from project records on this issue. Table 1. ADRP: Maintenance Vehicles and Equipment Item SAR Estimates Actual Procured Tipper Trucks 11 Water Tankers 1 Pick-up Trucks 18 Total Trucks 36 30 Compactors/Rollers 11 18 Graders/Loaders/Dozers 9 10 Mixers/Spreaders 7 8 Mobile Cranes 1 1 Cars 2 2 Motorcycles 31 33 7 B. Project Costs 3.02 Comparison of appraisal estimates and actual cost of the ADRP are presented in Table 2 below. Table 2. ADRP: Summary Project Costs (US$ Million) Component SAR % Actual % Change Etimates Base Costs Base % Costs Costs Road Rehabilitation 15.5 74% 17.7 73% + 14 and Maintenance E. Maamobi Upgrading 2.6 12% 3.7 15% + 42 Property Revaluation 1.8 9% 1.2 5% - 33 Municipal Revenue 1.2 6% 1.7 7% + 42 Mobilization Total Base Cost 21.1 100% 24.3 100% -Contingencies 5.2 4.26 Total Project Costs 26.3 28.5 + 8 3.03 Although the overall costs of the project went up only by 8%, the costs of many components increased substantially as follows: * road and drainage rehabilitation (including the construction of the two missing segments of the outer ringroad) was 14% more costly than estimated at appraisal due to increase in its scope. At the same time, however, rehabilitation of trunk roads was dropped; * the costs of the East Maamobi upgrading went up by 42% mainly because of rehabilitation of additional drains and increased number of latrines; and * assistance to the municipal revenue mobilization exercise was increased by 42% as the referencing of properties required more work than estimated at appraisal. On the other hand, the costs of the property revaluation exercise turned out to be about 33% lower than estimated at appraisal. The total contingency of US$5.2 million was adequate to cover the increases of some of the components. 6 The PCR states that this is a total for commitments outstanding as of June 30, 1992, largely for vehicles and equipment. The disbursement of these commitments was by December 31, 1992. 8 C. Institutional Results 3.04 The TSC was entrusted with the overall coordination and administration of the project. This proved to be an effective arrangement, at least in the short term. There are some doubts, however, about the longer-term prospects. Some concern was voiced e.g. by the AMA, as they felt that the TSC may become too dominant and it may claim too much of the resources -- financial, training, and vehicles -- at the expense of the needs of other organizations which are supposed to carry out e.g. O&M in the long run. 3.05 AMRD' was established under AMA to take over the sole responsibility for O&M activities of the Accra road and drainage network. Still, today AMRD receives most of its funding from the GOG through DUR. AMRD has a rather small staff, comprising four civil engineers, one quantity surveyor, three technical operators, three accountants, and some 315 laborers (skilled and semi-skilled). Although no specific indicators can be cited, the AMRD does not seem to have adequate capacity to fully carry out its responsibilities. This is evident in many neighborhoods in Accra where potholes and badly maintained drains are a common sight. The capacity problem becomes specially eminent at the level of senior engineering and managerial staff because their number is low and their experience is less than adequate. This staffing shortage is further compounded by the fact that its two managers are seconded from (and paid by) DUR. The upkeep of vehicles and maintenance equipment is a chronic problem. In order to help AMRD coordinate its work program, DUR is developing an MIS for the purpose. 3.06 The AMRD's maintenance schedule includes a routine maintenance program run on a biweekly basis to deal with potholes, traffic signs, and desilting of drains. The targets of such maintenance are determined through weekly inspection reports and complaints received from the public. A periodic preventive maintenance program includes larger scale works such as resealing of roads, and repairs of whole sections of roads. In addition to the regular maintenance of streets and drains, the AMRD carries out some capital works as well. It usually contracts out construction of new drains or improvements to existing ones and resurfacing of sections of streets. At the time of the Audit, 17 such contracts were under implementation. In 1993, the total budget of the AMRD was C2,900 million, about US$4.1 million equivalent (C2,300 million for contracts and C600 million for force-account), up by 26% from 1992. All planning and larger contracts are still today handled by DUR, as well as more complicated repairs such as repair of traffic lights. 3.07 The training in property valuation techniques was initially given to selected staff of the LVD/MLG involving some 120 staff in total. It was provided by a UK valuation team of three valuers who stayed in Accra for three years, including during the actual valuation process itself. The property referencing and revaluation was a joint effort by the City Engineer's Department of the AMA, LVD, and the UK team. The process covered all the properties in Accra (about 80,000 properties at the time), as most did not have property reference numbers. The process This was to replace and expand the role of the then existing Accra District Roads Unit (ADRU). 9 was completed by the end of 1987 for Accra, but the reference numbering turned out to be complex and required one of the UK valuers to stay over for one extra year to complete the property records with the AMA. The process continued in Tema (about 30,000 properties) by LVB, the successor of LVD. The Tema valuation was completed in June 1990. 3.08 During the initial valuation process, all properties under construction were left out and have now, after their completion, been included in a supplementary exercise. The process is of course a continuous exercise and the LVB has the capacity to carry out updating of the records to include new properties after which they are rated by the AMA. The updating process is, however, falling behind again due to shortage of transportation facilities for valuators. The system, by necessity, is computerized but the AMA and LVB face some hardware problems as there are no adequate funds available to replace outdated computers and printers. 3.09 Another twinning team from a UK county council provided assistance to the AMA and training to its staff on revenue mobilization, specifically in accounting, budgeting, revenue collection and general financial control. In addition, this assistance included developing systems for labor programming and contract management. This activity was effectively launched and immediate results were visible. A financial advisor worked with AMA for two and a half years to help develop a proper budgeting process. As a result, a computerized billing and collection system was introduced in January 1989. The Audit was informed by AMA that the advisor, although well experienced in public utility accounting, was not adequately familiar with city treasury functions, and was therefore unable to fully assist in making the plans operational. Certain problems still remain unsolved, due to non-existence of bylaws and regulations as well as practical manuals. 3.10 The new property tax system coupled with improved collection increased AMA's revenues more than threefold since 1984. Some time after the departure of the advisor, the arrears situation deteriorated; the Audit was informed that the collection ratio is now in the order of 75 to 80%. The identification of additional properties is reported to be the main problem, but the number of defaulters has also remained at an unacceptably high level. AMA is currently engaging some private lawyers to help in correcting the situation. At the time of the Audit, AMA was also in the process of establishing a special team which, with the help of LVB, was to update the property identification by March 1994. 3.11 The lack of adequate computer skills has impaired AMA's ability to fully implement the new management initiatives. AMA's payroll, property tax rates, license fees, and refuse collection levies were already computerized at the time of the Audit. General MIS, billing, and accounting systems were still under implementation. A local consultant was being engaged to install the system and provide sufficient training of staff over a longer period of time. D. Studies 3.12 The traffic study initially covered the road network and traffic management needs of the central business district in Accra. The study was expected to recommend some minor works for 10 immediate improvements and pilot traffic management measures but, in the end, did not come up with any such recommendations. The funds earmarked for the minor works and pilot measures were instead utilized to expand the study to cover Secondi-Takoradi and Kumasi. The results of the study were useful for the preparation of the follow-on project (Credit 2157-GH). Some aspects of the study also helped to prepare a quick disbursing "Priority Works Project"' (PWP, Credit 1874-GH) which was implemented in parallel with the ADRP. 4. BORROWER AND BANK PERFORMANCE A. Borrower Performance 4.01 Despite the delay of one year in the completion of the project, the borrower's performance was reasonable. Especially the start of the project was effective, thanks to assigning the TSC to be in charge of project coordination and administration. Some doubts, however, about the relevance of this approach in the long-run were presented to the Audit (see para. 5.04 below). The shortage of counterpart funding was occasionally a problem but did not cause critical delays. In the case of purchase of vehicles and equipment, the IDA funds were made available to compensate for the shortfalls. B. Bank Performance 4.02 The Bank's "technical assistance" through frequent supervision was much appreciated in general by the implementing agencies. As the project components all included training, both on- the-job and specialized programs, the SAR stipulated (para. 4.20) that, in addition to the regular engineering and financial expertise, the supervision missions should also provide a training specialist's support. The project records, however, show that this training specialist input was never provided. SAR also suggested that the Bank was to provide support for property valuation through its supervision, but no such specialist participated in Bank missions after project appraisal. 4.03 Bank missions provided good staff continuity. The total number of missions was 23 over a period of seven and a half years, thus giving an average frequency of just below four months. Especially after the credit effectiveness, project entities received careful coaching by frequent Bank missions. The longest interval between missions was 12 months, from July 1989 to July 1990. At this time, the disbursements also began to level off. The Audit considers that this high level of supervision was essential to deal effectively with issues that came up from time to time in this complex first urban project in Ghana, which in addition was implemented under difficult country conditions. Coordination between many project entities, and two projects running in parallel (ADRP and PWP) was a demanding task for the Bank staff. An emergency program of priority and labor-intensive urban works to mitigate the adverse effects of Ghana's adjustment program. 11 C. Compliance with Covenants and Directives 4.04 The compliance with key covenants was not a problem. The borrower failed, however, to make adequate foreign exchange available for purchase of equipment and vehicles (by ACC/AMA) as required in Section 4.03 of the Credit Agreement. IDA agreed to increase its share of this local funding. Fortunately, this was possible as the dollar value of the credit, nominated in SDRs, increased. The Audit also finds that the project did not compromise any critical OMSs and OPNs. 5. PERFORMANCE ASSESSMENT AND ISSUES A. Project Performance Assessment 5.01 The Audit finds that the ADRP achieved most of its physical objectives. Exceptions were some trunk road rehabilitation works that were dropped due to the increased scope of the ring road construction. In addition, the scope of upgrading the East Maamobi infrastructure was increased utilizing the savings realized due to the devaluation of the Cedi. Institutional achievements were promising as far as the establishment of the property valuation system and improvements to the tax collection are concerned. The project failed, however, to make these new systems fully operational, thus leaving revenue mobilization efforts incomplete. This casts some doubt on AMA's capacity to handle the maintenance of road and drainage networks in the long-run, as well as on sustained operation of other municipal services. On balance, the Audit rates the project's outcome satisfactory, its institutional development as modest, and the sustainability of its benefits as uncertain. The PCR-based ratings are identical. B. Issues 5.02 As discussed above (para. 3.01) the construction of community latrines in East Maamobi was successfully carried out. During the field trip to the area, the Audit observed, and was later provided with more specific information, that the latrines are not used by the residents as willingly as anticipated at the design stage of the project (and as initially observed). It has been difficult to organize resident committees for the upkeep and operation of the latrines to keep them sufficiently hygienic. Also the collection of user fees to cover the costs needs to be improved. Emptying of the latrines seems to be the main problem. The cost of an 8- compartment community latrine was in the order of US$1,300 (in 1993 prices), and the cost of a domestic (one family) latrine"e may be nearly US$200. Domestic latrines have been Aqua-privies with a larger septic tank have been constructed in the nearby Nima area. The solution is more costly but the latrines can be easily emptied with vacuum-tank truck, with emptying intervals of about six months. 10 This is based on cost information from a recent program which constructed some 100 domestic latrines. 12 constructed at the cost of US$100 to 150 (single-pit and double-pit respectively) elsewhere. Drawing on a wealth of information on these issues from an ongoing program in Kumasi where these activities have generally been privatized, a more suitable solution could be developed that would be better affordable to low-income beneficiaries. 5.03 The project improved drainage and refuse collection within the East Maamobi area. The Audit observed, however, that refuse collection does not seem to function adequately any longer; only two years after the project completion. Drains are being used for garbage dumping to the extent that the maintenance of drains can't keep up. The intended environmental benefits are thus negated. 5.04 The project placed a heavy load on TSC as it was to act as a catalyst and coordinator of the capacity building initiative. Therefore, a concentrated effort was made to strengthen TSC itself in order for it to successfully shoulder this responsibility. Over time, some questions have, however, arisen if this approach might compromise the effective capacity building of other relevant urban organizations, and indeed the objective of decentralization. As this same approach is continued under the PWP (recently completed) and Second Urban Project (Credits 1874-GH and 2157-GH respectively), the institutional development aspects of AMA and the urban sector as a whole should be revisited after completion of the Second Urban Project. 6. CONCLUSIONS AND LESSONS A. Conclusions 6.01 The ADRP had ambitious objectives and as such became a complex operation. It was also implemented during a very difficult period when Ghana faced many serious economic problems. With appropriate arrangements and dedicated work on the part of the implementing agencies, the project was brought to a successful completion with a rather small time overrun, and approximately within the stipulated costs. The Bank invested the required manpower and expertise in the supervision of the project. As a result, the development of the urban sector in Ghana moved in the right direction. The lessons from this project and the follow-on operations that should be evaluated in due course, will be important in designing future urban activities. B. Lessons of Experience 6.02 Key lessons learned from the ADRP include: a) It is essential to involve all the participating entities in the project preparation process to the fullest extent, in order to promote sound ownership among the project stakeholders. This should particularly apply to those who are ultimately responsible for O&M of the new facilities; 13 b) To make the TA support fully effective, it is of utmost importance to involve the recipient entities in the selection of the candidates for TA personnel, and together with their twinning partners define the nature of the required assistance program in full detail; c) In a multi-faceted operation like the ADRP, effective coordination between, and clearly defined roles among the various entities are the key to success; d) The involvement of beneficiaries in infrastructure upgrading as carried out in East Maamobi is important. The project should also include an effective M&E to fully realize the benefits of the lessons learned and enable a correction of the course of action during the implementation as appropriate; and e) After this first project, the Bank has supported several urban operations in Accra (and other towns in Ghana) which followed one another only a few years apart. All of them have (rightly) focused on capacity building of the sector. It would be essential to carry out a careful retrospective review of the results achieved so far before any new operation be launched.