Ghana - Local Government Development Project

Document of The World Bank Report No: 27064 IMPLEMENTATION COMPLETION REPORT (IDA-25680) ON A CREDIT IN THE AMOUNT OF SDR 27.7 MILLION (US$ 38.5 MILLION equivalent) TO THE REPUBLIC OF GHANA FOR THE LOCAL GOVERNMENT DEVELOPMENT PROJECT October 28, 2003 Water & Urban II Africa Region  CURRENCY EQUIVALENTS (Exchange Rate Effective September 9, 2003) Currency Unit = Cedis 8,690 = US$ 1.00 US$ 1.00 = 8,690 FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS ADRP Accra District Rehabilitation Project APL Adaptable Program Lending CAGD Controller and Accountant General's Department DA District Assembly DACF District Assembly Common Fund DCE District Chief Executive ECOWAS Economic Community of West African States EOP End of Project GOG Government of Ghana GPRTU Ghana Private Road Transport Union HIPC Highly Indepted Poor Countries ICR Implementation Completion Report IDA International Development Association KfW Kreditanstalt fur Wiederaufbau KVIP Kumasi Ventilated Improved Pit LGDP Local Government Development Project LGPSU Local Government Project Support Unit LI Legislative Instrument LVB Land Valuation Board MA Municipal Assembly MFEP Ministry of Finance and Economic Planning MLGRD Ministry of Local Government and Rural Development MoU Memorandum of Understanding MWH Ministry of Works and Housing NDPC National Development Planning Commission NGO Non-Governmental Organizations O&M Operation and Maintenance PNDC Provisional National Defense Council PSCS Personal Services Contract Staff PWP Priority Works Project QAG Quality Assurance Group RCC Regional Coordinating Council RIAP Revenue Improvement Action Plan SAR Staff Appraisal Report SD Survey Department TCPD Town and Country Planning Department TSC Technical Services Center UESP Urban Environmental Sanitation Project Vice President: Callisto Madavo Country Director: Mats Karlsson Sector Manager: Inger Andersen Task Team Leader: Charles K. Boakye  GHANA LOCAL GOVERNMENT DEVELOPMENT PROJECT CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 4 5. Major Factors Affecting Implementation and Outcome 12 6. Sustainability 14 7. Bank and Borrower Performance 15 8. Lessons Learned 17 9. Partner Comments 19 10. Additional Information 23 Annex 1. Key Performance Indicators/Log Frame Matrix 24 Annex 2. Project Costs and Financing 28 Annex 3. Economic Costs and Benefits 30 Annex 4. Bank Inputs 34 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 36 Annex 6. Ratings of Bank and Borrower Performance 37 Annex 7. List of Supporting Documents 38  Project ID: P000936 Project Name: LOCAL GOVT DEV. Team Leader: Charles K. Boakye TL Unit: AFTU2 ICR Type: Core ICR Report Date: October 29, 2003 1. Project Data Name: LOCAL GOVT DEV. L/C/TF Number: IDA-25680 Country/Department: GHANA Region: Africa Regional Office Sector/subsector: General water/sanitation/flood protection sector (38%); General transportation sector (38%); Sub-national government administration (12%); Central government administration (12%) Theme: Access to urban services for the poor (P); Municipal finance (P); Municipal governance and institution building (P); Decentralization (P); Pollution management and environmental health (S) KEY DATES Original Revised/Actual PCD: 12/11/1989 Effective: 12/30/1994 12/30/1994 Appraisal: 06/10/1993 MTR: 02/01/1997 06/23/1997 Approval: 02/17/1994 Closing: 12/31/2001 03/31/2003 Borrower/Implementing Agency: GOVERNMENT OF GHANA/Local Government Project Support Unit Other Partners: Kreditanstalt fur Wiederaufbau (KfW) STAFF Current At Appraisal Vice President: Callisto Madavo Edward V. K. Jaycox Country Director: Mats Karlsson Edwin Lim Sector Manager: Inger Andersen James O. Wright Team Leader at ICR: Charles K. Boakye Jagdish K. Bahal ICR Primary Author: Charles K. Boakye and Gerhard Tschannerl; assisted by Ephrem Asebe 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome:S Sustainability:L Institutional Development Impact:M Bank Performance:S Borrower Performance:S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No  3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The objectives of the project were to: a) Improve basic infrastructure and urban services in secondary cities, especially services benefiting lower-income people. b) Promote the sustainability and expansion of urban services by strengthening the District Assemblies' (DAs) financial, technical and managerial capacities. c) Support the Government's decentralization program to promote accountability and efficiency in the provision of infrastructure and services. The project objectives were in accordance with the Government's priorities of infrastructure provision to urban areas with decentralized management of urban services. The project objectives were in line with IDA's Country Assistance Strategy (CAS) for Ghana that supported decentralized development and local ownership of the development process. The Project was preceded by Sector Study: Ghana, Strengthening Local initiative and Building Local Capacity, (Report No. 11369)-GH, Washington, D.C.: February 1993. It addressed the process and scope of devolution of functions, and the local/central fiscal relationships, including revenue sharing, and reviewed the capacity of DAs to plan, finance, manage and maintain local infrastructure and service delivery. A previous urban study - Ghana, Reviving the Urban Sector (July 1988) - had also proposed a comprehensive agenda on decentralization, urban management and institutional development, and key elements were incorporated in the preceding Urban II project, which also remained relevant for the follow-up Local Government Development Project (LGDP). While Part (a) of the objective was clear, Parts (b) and (c) contained cause and effect statements, which made it difficult to choose a consistent set of Performance Indicators (PI) for impact (see Annex 1). Nevertheless, the intention of the objective was clear. The PIs, developed during the Mid-Term review of July 1997 offered the opportunity for DAs to review their understanding of project objectives and components. 3.2 Revised Objective: The project objectives were not changed. 3.3 Original Components: a) Rehabilitation and Upgrading of Infrastructure and Urban Services Component (US$35.69 million). This included the rehabilitation and upgrading of three types of basic infrastructure and urban services: i) Roads and Storm Water Drainage consisted of the rehabilitation of 73.9 km of existing key town center roads and links with the trunk road network; associated storm water drains; and basic traffic management measures, including intersection improvements and traffic signs. ii) Markets and Lorry Parks involved rehabilitation and upgrading of existing markets and lorry parks, including the provision of paving, drainage, water standpipes, sanitation and washing facilities, security lighting, and structures. iii) Waste Management included a) Solid Waste, covering the establishment of environmentally engineered landfill sites in each of the 11 towns, fencing of refuse collection stations, and provision of containers, tractors, and other equipment; and b) Liquid Waste, involving the establishment of liquid waste disposal sites, renovation of public toilets, construction of new pubic toilets, extension of water and lighting to public toilets, provision of cesspit emptiers, construction of underground holding tanks, and funding of matching grants to households for latrine construction. b) Institutional Strengthening Component (US$7.74 million): i) Improving the financial, managerial, and technical capacity of District Assemblies to perform functions consistent with Ghana's overall economic reform program; and ii) Strengthening the capacity of selected central government agencies to support the District - 2 -  Assemblies. The central Government agencies involved in this component included the Land Valuation Board (LVB) for the revaluation of properties; Survey Department (SD) for the preparation of maps; Town and Country Planning Department (TCPD) for the preparation of simple land-use plans; and Technical Services Center (TSC) for project management and coordination for Ministry of Local Government and Rural Development (MLGRD) and the DAs. c) Operation and Maintenance Component (US$5.21 million) included establishing and maintaining a fund for operation and maintenance of the facilities provided under the project, and recurrent cost of DA's key professional staff. d) Contingencies (physical and price) (US$6.88 million) 3.4 Revised Components: The project objectives were not changed. 3.5 Quality at Entry: The quality at entry is rated satisfactory. The project was designed in the context of the Government of Ghana's evolving decentralization policy, the basic premise of which was that the DAs would eventually have discretionary responsibility for policy formulation, planning, implementation and maintenance. It focused on improving the living conditions of the population of 11 secondary towns, namely Bawku, Bolgatanga, Wa, Techiman, Sunyani, Ho, Koforidua, Agona-Swedru, Keta/Anloga, Cape Coast and Elmina. Six of the towns are regional capitals while the remaining five are district capitals. The lessons learned from previous projects were taken into account during project preparation. The project evolved after the success of the IDA-financed Accra District Rehabilitation Project (ADRP) and the Urban II projects first in Accra and Tema, then in the five Metropolitan/Municipal cities. The Government felt the need to replicate the lessons for the development of 11 District headquarters that were next in the order of size and economic importance. The project objectives were consistent with Government's priority for the sector. The Government's ambitious decentralization program started in 1988 with the passage of the Local Government Law (PNDCL 207) that gave authority for the establishment of 45 new local Governments bringing the total to 110. This was followed by the 1992 constitution of the Republic of Ghana after 11 years of military rule, and Chapter 20 of the new constitution was devoted entirely to "Decentralization and Local Government." The Local Government Law assigned to the DAs the responsibility for performing the functions previously carried out at the local level by the 22 central government departments, making the DAs directly responsible for health, construction, rehabilitation, maintenance of buildings, roads and drains, markets and lorry parks and community development. The project therefore focused on building capacity in the DAs, while other services were managed by the central government. The project was consistent with the results of sector work and the CAS objectives. Two World Bank sector studies that were carried out about the same time stressed the need for the Bank to support the sector to deepen the decentralization program (see Section 3.1). The project had local government ownership from the start, but limited community participation. The subcomponents selected for financing in each district were prepared with the involvement of the Assembly members, and all the draft designs were presented to the General Assembly in the District before the designs were finalized. The consultations took place with the local political establishment, although not directly with the intended beneficiaries. The potential environmental and social impact was adequately studied. Most of the works were small and had a low potential for negative environmental impact, with the exception of the final waste disposal sites, which were located away from settlements. Environment Impact Assessments (EIAs) were carried - 3 -  out on all the sites. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The overall outcome of the project is rated satisfactory, and the impact on poverty reduction is modest. More than 100,000 low-income people who did not have access to proper sanitation benefited from household toilets in their homes. Traders and other commuters also benefited immensely from public toilets and transport terminals that were renovated or improved under the project. A number of skilled and unskilled laborers got temporary employment from over 60 civil works contracts that were awarded. While a gender objective was not explicitly stated, many of the infrastructure works benefit women more than men, including, markets, lorry parks (used largely for moving goods to and from the market), and urban roads and drains (facilitating trading and making cleaning easier). They also benefit from the improved working environment at lorry parks and health conditions in the markets. A. The improvement in basic infrastructure and urban services in the project towns is satisfactory. a) Most DAs have met the target of no additional refuse accumulating outside designated areas and at least 90 percent of the refuse collection equipment in operating condition (see Annex 1). Most of the refuse that has accumulated in parts of the cities over several years has been removed and unauthorized dumping of fresh refuse in the town has ceased, which will improve the health status of the people. Some accumulated refuse remains in the towns' and it is noteworthy that some DAs are gradually removing them on their own. b) There is greater access to sanitation and better maintenance of latrines through the use of more cesspit emptiers, which in some cases have served other towns in the region in addition to project towns. c) The construction of urban roads and drains has made an impact. Road construction has eased traffic congestion (aided by the introduction of one-way systems), improved accessibility to businesses and residences, improved road safety through the introduction of traffic signals, and reduced dust. d) In most DAs drainage has eliminated flooding on improved or rehabilitated roads adjacent to, as well as in markets and transport terminals (Annex 1). In three towns drainage has helped to reclaim seriously eroded land. e) The improvement of 14 markets combined with transport terminals was successful after long delays. To date, some markets have not yet been put to use. Only three towns erected all the market sheds and two erected a part of them with their own funds as originally planned. f) All the lorry parks are in operation, managed by GPRTU. Some have already encountered major maintenance problems. In Bolgatanga, as a result of substandard work performed by the contractor, the servicing and washing of vehicles, as well as general use of unauthorized heavy duty articulated trucks on the lorry park, the condition of the pavement in the Bolgatanga lorry park deteriorated rapidly. Subsequently, it was rehabilitated again through the project just before the credit came to a close. B. The outcome of promoting the sustainability and expansion of urban services by strengthening the DAs' financial, technical and managerial capacity is satisfactory. The objective was to implement and sustain project benefits through effective integration of the staff of previously centralized departments assigned to the DA level. The project was to support this objective by assisting the 11 DAs to reduce the number of district-level departments and establish an effective internal structure within departments. The project elements focused on integrating staff of the centralized departments into District administration, adopting efficiently-oriented approaches, where feasible to District Assembly employment (greater use of fixed term or contract employment), and demonstrating effective commitment to adequate recurrent financing of O&M requirements of infrastructure investments financed under the project, including the involvement of the private sector. The project demonstrated that strong institutional capacity in the fields of planning, management, finance and operational and maintenance are essential to providing adequate services, and assuring the - 4 -  sustainability of services. The Zonal Officers and District Contract Staff (see Section 4.2 B) helped to strengthen the District Works and Finance Departments. The personnel assisted in the preparation of annual implementation action plans for all the DAs, assisted DAs in identifying on-the-job training needs for staff of the works department, and created awareness in the reorganization of the districts in conformity with the Local Government Law Act 462. They did not only work on the project, but also several other projects, including education, health and agriculture projects implemented by the DAs and financed by other donors and GoG. The staff also privatized the operation of the public toilets that improved the revenue generated from this source by about 700 percent in some DAs. In addition to the revenues generated, the DAs further saved the costs involved in the running of the toilets, such as disinfectants and the wages of the attendants and conservancy laborers. However, the additional Legislative Instruments (LI) under Act 462, which were expected to be implemented to enable the decentralized departments work together, were not put in place by the Government. Of the 22 decentralized departments of the district, only the Department of Agriculture was integrated into the DA as envisaged under Local Government Act 462. All other constituent departments acted independently of each other as in PNDC Law 207. As the working arrangement for the decentralized departments was not in place, the Zonal Officers could not achieve the required results. A new Local Government Act was enacted in August 2003 that is designed to place the DA staff under a separate working arrangement from the Civil Service. While the contract staff were effective in producing short-term outputs of the project, they had only a limited long-term impact on the capacity of the DAs to carry out the tasks (see Section 4.2 B). Performance of DAs Revenue Collection All 11 DAs exceeded the target of increasing the internal revenue collection by at least 10 percent in real terms (adjusted for inflation) in the period of 1997-2002. However only 7 DAs met the target of increasing market revenue collection by at least 30 percent in real terms (see PIs in Annex 1). The reason for the shortfall on market revenues was that the rehabilitation or upgrading of markets was completed only towards the end of the project and could therefore not have had a significant impact on the market rates collected. The combined internal revenues for all DAs increased by 169 percent (Annex 1, Table 1). The collection of property rates for all DAs combined increased by 887 percent over the period (Annex 1, Table 2) and the collection of market revenues by 107 percent (Annex 1, Table 3). As the property re-valuation was completed around 1999, an increase in the property rates collection could be expected in the period of 2000 to 2002, which was indeed the case: the average annual collection for all DAs combined increased in real terms by 68 percent from the period of 1997-1999 to the period of 2000-2002 (Annex 1, Table 2). In 2003, a pilot computerized budgeting and accounting system was established in four selected DAs that had the capacity to ensure proper management, and there are indications that the system will improve DAs financial management system. There was considerable variability in the revenue collection performance of the different DAs, which was particularly high for property rates. By and large, the performance followed the same pattern for the different kinds of revenues and in most cases can be ascribed to the prevailing social, economic, and political conditions in the respective DA. Bolgatanga had the biggest overall increase, followed by Koforidua and Cape Coast. Bawku, Keta and Elmina had the lowest and showed a decline in market revenues. Keta has suffered economic decline due to the devastating advance of the sea, and Bawku due to periodic communal strife. Bolgatanga showed an unusual trend, where the property rates increased 63-fold, while the market revenues declined over the period. As a result of the direct intervention made on property revaluation, property rates now form the major share of locally generated revenues in most DAs. Figures recorded show that average DA revenues from property revaluation have increased from a range of 3 and 20 percent of internal revenues at the time of appraisal to 5 and 58 percent of internal revenues in 2002. (The rates formed 58 percent of revenues from Koforidua, but only 5 percent from Keta). The share of property rates as a proportion of total DA internal revenues collected increased over the period from 7 percent to 27 percent, while that of market revenues rose from 22 percent to 30 percent from 1997 to 2000 and then declined to 17 percent by 2002 (Annex 1, Tables 2 and 3). Projections of total revenues were made at the time of appraisal up - 5 -  to the year 2000. The actual performance surpassed the projections for all DAs, which can be attributed more to an underestimation of the inflation rate at appraisal than to a consistent overachievement on revenue collection by the DAs (Annex 1, Table 4). Most DAs have contracted out the public toilets to private management through a competitive bidding process, and substantial revenue has been collected from these toilets in Agona Swedru, Koforidua and Sunyani District Assemblies. Keta, Bolgatanga, Bawku and Wa have not been able to involve the private sector in the management of these toilets, and it was learned that the DAs actually spend considerable amount of money in maintaining them. Privatization of public toilets was successful in DAs that put in place competitive procedures to the award of contracts, resulting in increased revenue from toilets. A survey was undertaken in 1999 to sample public opinion on the privatization of public toilets in Sunyani and the response was positive. Ninety-three percent of those sampled were satisfied with the privatization exercise and with the level of services provided by the operators. C. The outcome of the support to the Government's decentralization program for better efficiency and accountability in infrastructure provision is satisfactory. The project was not intended to support a comprehensive program on decentralization, but to make some specific interventions to advance decentralization. Most of the support was in the form of technical assistance within the confines of the prevailing state of decentralization, in the areas of reform as well as capacity building at the District level. Based on recommendations of the sector study, the project was designed to provide Government with technical assistance to review the revenue mobilization and transfer system, and generate proposals for giving districts more autonomy and flexibility in revenue raising and making transfers more predictable. During implementation, the Government decided to undertake this assignment on its own due to the political issues involved, and actually moved ahead of the project in that respect. Subsequently, based on the provisions in the Local Government law that established the District Assembly Common Fund (DACF) grants, the Government established a formula for allocating grants to the Districts. The allocation is approved by Parliament every year. Since 1995, central government has transferred nearly all of the targeted 5 percent of revenues to the DAs on a quarterly basis. The transfers were irregular in the early years and fell short of the targeted 5 percent, but the situation has improved in the past two years. All the central government agencies, i.e., Survey Department, TCPD and LVB contributed to the satisfactory performance of the institutional strengthening components. As a result of the capacity built, the agencies are carrying out similar assignments for follow-up Bank-financed projects, as well as GoG projects. Prior to the year 2000, the property rates billing and collection system was manual. However, after the preparation of a new valuation roll by the LVB, the system has been computerized, and staff has been trained to operate the system. The new Chief Executives engage LVB staff regularly on contract to resolve valuation-roll updates and other nagging issues that come up from time to time. 4.2 Outputs by components: See Annex 1 for the output indicators. A. Rehabilitation and Upgrading of Infrastructure and Urban Services Component (US$34.6 million SAR; US$38.0 million ICR). The output of the component is satisfactory. - 6 -  Road improvement and stormwater drainage (US$15.5 million SAR; US$16.6 million ICR). The output of this component is satisfactory. A total of 57.6km was constructed out of a targeted 73.9km. The shortfall resulted from increase in cost of implementing road contracts and changes in DAs priorities. Some roads were dropped because they were planned or had in the meantime been constructed by the Ghana Highway Authority. Some savings from the road works were used to finance the construction of more storm drains and street lights. About 4.2km of storm drains were constructed in 4 towns, as against a target of 3.2km, and 1,196 street lights installed, more than twice the original target of 503. The inclusion of major electrical works in some civil work contracts was ill advised as some of the contractors did not have the necessary expertise in electrical works, with the result that some of the lights installed were sub-standard. Market and lorry park infrastructure (US$6.6 million SAR; US$9.6 million ICR). The output of this component is satisfactory. A total area of 42 hectares compared to appraisal target of 28 hectares was covered. KfW contributed the bulk of the funds for the market and lorry parks infrastructure at the original cost of DM16 million (US$ 8.7 million) while GoG contributed the rest. KfW also financed the construction of 85 market sheds in Wa, Bolgatanga and Bawku, and 16 sheds in Anloga. KfW assistance included the services of a consultant who was the lead supervisor. An additional amount of DM1 million was later transferred to the project from the Promotion of District Capitals project to finance the extra cost of the market sheds in Bolgatanga, Bawku, Wa and Anloga, bringing the total KfW contribution to $9.14 million. IDA also contributed about $0.6 million for the construction of 8 market sheds in Techiman, lockable stores in Elmina and 10 market sheds, loading bay and drive area in Ho. Waste Management(US$12.5 million SAR; US$11.8 million ICR). The output of the overall component is satisfactory. Ten of the planned 11 final disposal sites, for both liquid (i.e. sludge) and solid waste, were constructed. An additional solid waste disposal site was rehabilitated in Koforidua, which was already in place and well engineered. All final disposal sites are now managed by the DAs and nearly all of them have had problems with their operation due to one or more of the following reasons: poor site selection, poor designs, and inadequate operational management. The liquid waste facilities in some of the towns were constructed as waste stabilization ponds and the rest were open pits. Some of the solid waste disposal sites were open spaces, but most of them were controlled tipping cells in a walled enclosure. Of the targeted 125 refuse collection points that served as transfer sites of solid waste, the project constructed 193. The following waste management equipment was delivered to the DAs: 11 bulldozers, 23 skip loaders, 23 cesspit emptiers, 11 tractors, 15 Roll-on Roll-off trucks, 64 trailers, 241 Open containers, 16 covered containers, 22 night soil containers, and 36 Roll-on Roll-off containers. Some assemblies converted one of the two septic tank emptiers that were supplied to them to a water tanker, and the bulldozer that was provided for the efficient filling of final solid waste disposal sites was frequently hired out to contractors for construction work. Of the 277 public toilets targeted for rehabilitation or construction, about 206 were completed. Fewer suitable sites could be identified than had been planned, and many of the toilets that were meant for construction only needed rehabilitation. This shortfall was amply made up by the construction of 5,968 household toilets, more than twice the targeted 2,700. They benefited an estimated 135,800 people instead of the originally planned 70,000, and consisted of several popular types: 70 percent were water closets and the rest were KVIPs, Aqua privies, and other types. Based on preliminary estimates, the total cost of a toilet facility was put at $300 to $400 and the project paid up to 50 percent ($200) of the cost into a DA sanitation fund for each household. An audit report carried out in 2002 showed anomalies in the disbursement of funds in some DAs, and recommended that in future, the component should be implemented by NGOs. B. Institutional strengthening (US$7.7 million SAR; US$11.9 million ICR). The overall outcome is satisfactory. - 7 -  Technical Assistance to DAs (US$2.7 million SAR; US$2.0 million ICR). The overall outcome is satisfactory. Technical assistance was provided through 12 national consultants in three key fields of municipal engineering, financial management and district planning and management over a three-year period. The 11 participating towns were grouped into four zones. Each zone was served by a Zonal Office, and comprised a Civil Engineer, a Finance Officer and a Planning Officer, to advise DAs in their respective fields for three years. The most senior of the three served as Project Coordinator. Adequate resources, including equipment and operational support were provided to enable the professionals provide the needed assistance for capacity building. Although there were outstanding tasks, the contracts of the zonal officers were not extended after the scheduled three years. Factors that affected their performance included conflicts with DAs personnel, lack of acceptability by DAs staff, the inability of the Assemblies to recruit their requisite chartered accountants as counterpart staff, and the fact that the various departments of the Assemblies had not yet been decentralized. Furthermore, the zonal officers were supposed to assist the DAs in the adaptation and introduction of improved budgeting and accounting systems, as well as software and manuals, which had been developed under the Urban II project. MLGRD apparently was not committed to carrying out these reforms. It never made these studies available to the zonal officers and the DAs and did not procure the necessary computer networking equipment required for their implementation. The major infrastructure works from which revenues could be collected, that is markets, lorry parks and public toilets, were not fully completed when the zonal officers were in place, and billing for property rates could not start since the property revaluation had not been completed. Key outputs are as follows: The Planning Officers in the zonal offices facilitated the preparation of 5-year medium term development plans, prepared for the Assemblies by consultants, and introduced the costing element to the plan. The limited success was due to the lack of an institutional link between the zonal planning unit and the National Development Planning Commission (NDPC.) The latter body has statutory responsibilities for the administration of planning studies throughout the country, and the support provided by the zonal planning office conformed to the technical requirements of the NDPC. The zonal planning officers could also influence the perceptions of the NDPC through the field experiences submitted to them. The Engineers organized briefing sessions for senior staff of DAs on relevant aspects of civil works contracts. Although the engineers were supposed to be advisors to the DA staff, they played a major role in both LGDP and non-LGDP civil works contracts, especially in towns where DA staff was not available. They helped to put in place arrangements for the maintenance of the waste management equipment and established modalities for the privatization of public toilets. The Finance Officers, who were chartered accountants, helped to introduce activity based budgeting in some DAs, which better reflects reality than incremental budgeting. The existing accounting system in most DAs reported only cash transactions, and the officers helped to record bills according to the accrual system of accounting. As a result of difficulties encountered in the recruitment of chartered accountants as counterpart staff in the DAs, (see below), the zonal Finance Officers not only advised, but also implemented the finance component in the Districts. Their effectiveness was severely constrained by the institutional fragmentation of financial management at the District level. Budgeting is controlled by the Budget section, headed by the Budget Officer who was a staff of Ministry of Finance and Economic Planning (MFEP), whereas accounting is controlled by the District Finance Officer, a staff of the Controller and Accountant General's Department (CAGD), who concentrates on expenditure controls in the assembly, leaving the revenue generation to the revenue section, controlled by Revenue staff who are not under any Ministry, but are staff of the Assembly. This arrangement, which still prevails in the Assemblies, creates a low incentive for revenue generation, and is responsible for a lack of coordination among the different sections. Strengthening the capacity of selected central government agencies to support the DAs Town and Country Planning Department (TCPD) for Land-Use Planning (US$0.33 million SAR; US$0.17 million ICR). The outcome of the component is satisfactory. Land-Use Planning was completed in 2003 with the delivery of electronic and hard copies of structural plans for both developed and undeveloped areas of two urban towns. The scheme will guide public and - 8 -  private investments in infrastructure, revenue generation and urban planning, ensure efficient urban growth and reduce environmental problems. It was originally meant for 9 towns, but due to insufficient cooperation among the TCPD, the consultants, and the project implementation unit, it was reduced to two towns, Sunyani and Techiman, with the understanding that Planning Officers in the other towns will replicate lessons in their respective towns at the end of the assignment. The replication has not yet materialized. TCPD staff in the head office and regional offices have, however, been trained in various aspects of development planning. All six regional offices were supplied with 6 double cabin pick-ups, 6 motorcycles and computers. Land Valuation Board (LVB) for Property Revaluation (US$0.66 million SAR; US$1.16 million ICR). The outcome of the component is satisfactory. A total of 125,043 properties were revalued in 12 towns, as against the targeted 74,300. The increase was due to the addition of 40,000 newly constructed properties. LVB staff carried out the revaluation and completed the exercise in 8 towns at the cost of $680,187 while three private sector valuation firms revalued properties in the remaining 3 towns, which amounted to 39 percent of properties at the cost of $481,361. LVB staff also carried out post-valuation exercises, involving outreach with DAs, updating of valuation rolls and a computerization program. Two LVB staff members were exposed to valuation techniques in the United Kingdom's Valuation Office and other staff members were trained in various aspects of valuation. The project also supplied 5 double-cabin pickups, 33 motorcycles and 2 computers, printers and accessories to the Department. Survey Department (SD) for Mapping (US$1.6 million SAR; US$1.8 million ICR). The outcome of the component is satisfactory. Of the targeted 300 sq. km, about 700 sq. km of large-scale maps of scale 1:2500 were produced for 11 project towns. The additional 400 sq. km covered the undeveloped areas of the urban towns to facilitate future planning. The maps, which are available in both digital and hard copy forms covering the project towns, will also be used for infrastructure programming and revenue mobilization services. The maps are essential to enable land users, planners, engineers, land valuers, and providers of utility services to carry out their work properly. The component was implemented by private sector contractors under the overall supervision of the Survey Department (SD), which is also responsible for the maintenance of the maps. The SD had undertaken a similar assignment under Urban II and other GoG projects. Their staff received overseas training in digital mapping techniques. Technical Services Center (TSC) for Project management (US$0.7 million SAR; US$1.3 million ICR). The overall outcome of the component is satisfactory. Under a Memorandum of Understanding between the Ministry of Works and Housing (MWH) and the Ministry of Local Government and Rural development, the TSC provided overall implementation and project management support to the other agencies and DAs. TSC had overall responsibility, on behalf of MLGRD, for accounting, procurement, monitoring, coordinating and reporting functions. The MLGRD was responsible for the institutional strengthening component, especially with respect to technical assistance for improving local revenue collection performance, financial management systems, and DAs organizational structure and staffing. It was resourced and benefited from an accounting system that enables the production of various reports. Both TSC and MLGRD performed satisfactorily under the project and by 1999, the MLGRD wanted to improve project coordination to enhance overall implementation of the follow-up bank financed UESP. The MoU was abrogated and most of the professional staffs were co-opted in the LGPSU that was established in the MLGRD with the responsibility for implementing all Bank-financed projects. The LGPSU carried out the functions previously undertaken by TSC well, but could not adequately handle the technical assistance support to the DAs. C. Incremental Recurrent costs and Operation and Maintenance (US$5.2 million SAR; US$2.3 million ICR). The overall output of this component is unsatisfactory. Operation and Maintenance (US$3.3 million SAR; US$1.4 million ICR). The output of this component is satisfactory. - 9 -  By the end of 1998, all DAs had established an O&M fund, and both IDA and the DAs made contributions to this fund up to December 2001, after which IDA financing stopped according to the declining disbursement percentage for this category. IDA financing into this fund was 50 percent for the first 3 years and 30 percent for subsequent years. The reported cumulative contribution of the DAs is $828,129 equivalent and that of IDA $538,330. These funds were used mainly for the servicing and fueling of waste management trucks, street cleaning, and drain cleaning for the DAs and were not necessarily limited to the infrastructure and equipment provided through the project. A technical and financial audit of the O&M funds, completed in August 2003, revealed irregularities in the disbursement of the funds. Some DA staff did not follow the guidelines for the disbursement of funds and LGPSU did not put in place adequate procedures for monitoring. The report recommended that future funds should target specific maintenance activities, and not all maintenance works. A follow-up on the recommendations contained in the report has been impaired by the fact that most of the DA staff who supervised the component had been transferred at the time of the audit. Personal Service Contracts Staff (PSCS) (US$1.9 million SAR; US$0.9 million ICR). The output of this component is unsatisfactory. In parallel with the establishment of Zonal Offices, a pilot scheme to test a "labor market approach" in local government employment was initiated to recruit 33 key staff in higher-level positions in the DAs on contract basis, and as counterparts to the Zonal Officers. The aim was to improve the competency of key staff at the District level, which could in general not be maintained due to the unattractive conditions of employment in the civil service, and was to ensure that all DAs had at least one qualified Engineer, one Planning Officer, and one Finance Officer. At the end of the recruitment drive, only 22 positions were filled, including 11 Planners, 10 Engineers and 1 Finance Officer. The Finance Officers were particularly difficult to find, as they were required to be chartered accountants, and they were not satisfied with the $1,000 (equivalent in cedis) remuneration offered for the position. The financing of the District Contract Staff was initially to be shared by the Government and DAs, and was based on the premise that as DAs' self-generated revenues increased as the Revenue Improvement Action Plans (RIAPs) were implemented, the DAs would increase their share of the financing over time. The terms and conditions of their contracts were negotiated with the MLGRD and the District Chief Executives (DCEs), and a uniform rate was agreed on for all the staff. The assignments started in mid-1997, and there was an initial lack of cooperation from DAs, who saw the recruitment as top-down. The wide salary gap between the PSCS and the DA civil servants also made matters worse. PSCS were not fully embraced by the DAs, and they did not involved them in their work. Subsequently, their contracts were not extended after the initial four years. The Government also financed 100 percent of the recurrent cost throughout the entire contract period of 4 years although it was meant to be shared with the DAs. These difficulties notwithstanding, the District staff contributed to improving the technical capacities of the DAs. After their contracts ended, the Government filled these positions with civil service staff. Other Cross-Sectoral outcome Training. A wide range of training activities was undertaken as part of improving capacity of the DAs and central government agencies. More than 50 courses were provided, benefiting professional staff, heads of departments and general Assembly staff. They included computer skills, computerized budget and accounting, monitoring and evaluation, staff development, performance evaluation, engineering, and planning. The training had only a limited impact as few of the senior DA staff trained were still in their posts at End of Project (EOP). They were either transferred to other DAs as part of a routine exercise by the PSC, or left to join the private sector. There was also a widespread duplication in training, as other District-based projects, supported by the Bank or other donors, included a similar training program at the DA level, and often for the same personnel. The widespread duplication has created an attitude towards training as a form of salary supplementation. This calls for better coordination and inter-sectoral harmonization between the departments and agencies promoting urban development. Private Sector Participation (PSP). Modest gains were made in involving the private sector in the management of infrastructure services. The civil works contracts were executed by private sector contractors, and public toilets and lorry parks have been put under private management. However, the private sector could not be involved in the management of the operation and maintenance of markets - 10 -  and waste collection services. After a delay of several years, a pilot study for 4 urban towns was completed for the establishment of a waste management system through PSP. The consultant designed ways of implementing a basic waste management system that takes into account financial, legal, institutional, social, technical and environmental aspects and details of the potential role of the private sector. Technical assistance with the implementation of these recommendations was provided to the four Districts near the end of the project, cutting short the completion of their implementation. The waste management equipment supplied under the project had not been leased to the private sector by EOP and implementation of this is being planned for within the next phase of the ongoing Urban 5 program. 4.3 Net Present Value/Economic rate of return: The economic viability of the project was measured in the SAR by calculating the EIRR of 14 road rehabilitation activities, 3 market interventions and two each of the solid waste and liquid waste management activities. These constituted about 20 percent of the project cost. An ex-post evaluation was carried out for the 14 road rehabilitation activities using actual contract values and updated traffic data. The data used for the analysis, the assumptions made, and the results of the exercise are presented in Annex 3, which shows an average weighted EIRR of the projects of 68 percent, well above the 15 percent required for approval of projects in the sector. These results also compare favorably with the corresponding figures of 84 percent overall weighted EIRR for the roads and drainage component, as given in the SAR. 4.4 Financial rate of return: No financial rate of return calculation is shown in the SAR. 4.5 Institutional development impact: The institutional development impact is rated as modest. 1. At the end of the project, few of the senior DA staff trained was still in place. They were transferred to other districts every three or four years, but some of them left for employment in the private sector, depriving the project DAs of the full benefits from the training that had been provided. The project assemblies have also benefited from senior professionals from other DAs. 2. The O&M account that was established in every District was not actually used for the upkeep of the entire municipal infrastructure services as intended. It has nevertheless increased the awareness of most Districts that better O&M of the equipment can be achieved through planning and budgeting (Section 4.2 C). 3. An attempt was made to improve the management, financial and technical capacity of the DAs through the recruitment of key staff on better terms than the civil service conditions. The rationale was that the additional revenue generated and the increased budgeting and accounting efficiency would far exceed the additional salary cost of these employees. This idea was successful insofar as it demonstrated that strong institutional capacity in the fields of planning, management, finance and operational and maintenance can provide improved municipal services. It failed, however, as a long-term measure because not enough chartered accountants could be attracted to these jobs, and the contract staff were largely considered "outsiders" (Section 4.2 B). 4. A pilot system for improved budgeting and financial management was introduced in 4 DAs that had a relatively good management capacity, and there are indications that this will improve financial management and budgeting (Section 4.1 B). 5. The TSC under the Ministry of Works and Housing was virtually dismembered in the course of project implementation and replaced with LGPSU. MLGRD found the cooperation with TSC across institutional lines increasingly difficult and in 2000 abrogated the MoU and moved the core of the TSC project implementation staff to MLGRD to form its own project implementation unit. While this may have improved the implementation of several World Bank-assisted projects, it did not result in a greater role for the regular MLGRD staff. - 11 -  6. The project has improved the capacity of the LVB and SD to better carry out their mandate. As a result, the DAs now use a permanent database for property revaluation and a computerized property rates billing and collection system. LVB monitors the use of the system in the DAs and provide them with needed technical assistance. LVB and SD are implementing similar services under the Urban 5 project. The improvement of land use planning in the DAs was less successful. 7. The local construction and consulting industry was strengthened through the awarding of contracts. All civil works construction, including two ICB contracts, was carried out by national contractors, and all design and supervision services were likewise done by national consultants. The absence of any form of performance monitoring in the construction industry affected the quality and efficiency of construction services. 8. Most DAs have contracted out the management of public toilets to private operators through a competitive process and are collecting substantial revenues (Section 4.2 C). 9. The waste management equipment supplied under the project had not been leased to the private sector by EOP, as intended. A study on PSP in waste management was completed only 2 years prior to the end of the project and could be introduced only as a pilot in 4 DAs, none of which showed significant improvements in their waste management by EOP, which would require more time and a systematic follow-up (Section 4.2 B). 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: 1. Absence of additional Legislation Instruments (LI) under Act 462, which were expected to be implemented to enable the decentralized departments work together. 2. Inflexibility. Strict interpretation of project covenants often led to reduction in scope of the civil works, so as to keep within budget. However, in almost all contracts, civil works were completed with excess funds available, necessitating going through another round of the bidding procedures, and above all, paying additional supervision fees. 5.2 Factors generally subject to government control: 1. High inflation and depreciation of the local currency adversely affected the implementation of the project. The average annual inflation was about 25 percent, and peaked at 75 percent in 1995 . Internal revenue generated under the RIAPs program depreciated as a result of inflation, and contractors and consultants persistently complained of depreciation of the value of the contracts. This compelled the project unit to introduce price adjustment clauses in almost all contracts. 2. The Government lacked the commitment to proceed decisively with decentralization. Issues relating to leadership, accountability and full devolution of power in DAs are almost absent in the local Government administration. This is largely due to the nature of the local Government law that currently exists. Generally, political decentralization has not, so far, been accompanied by a commensurate decentralization of authority: transfer of significant decision-making authority has not accompanied creation of fully locally elected legislative bodies at district level. 3. Nearly all the District Chief Executives were changed in 1996 and 1999 as part of the local government reshuffle administration, and subsequently most of their senior staff was also changed. The new Chief Executives did not understand the project and attempted to change the priorities that had been agreed with the predecessors. This caused considerable disruption to the project. Following the change in government after the elections of December 2000, the Chief Executives were again changed, but by this time nearly all the components were completed. They did, however, give an impetus to the revenue generation components. 4. Counterpart Funds: The Government reneged on their counterpart fund contribution and this - 12 -  affected the implementation of civil works components in the first few years. The implementing agency was left with no option but delay the payment of withholding tax deductions due the Internal Revenue Service (IRS), thus creating a kind of vicious cycle, where there was no money for counterpart fund contribution as a result of poor performance in tax collection, contributed in part by the inability of project officers to pay tax deductions promptly. This arrangement, apparently with the approval of MFEP and MLGRD, helped improve the cashflow and eased the counterpart fund crises. 5. More than half the DAs obtained adequate evidence of land rights for the solid waste disposal sites before the project became effective, which allowed a quick start of the project. The other DAs obtained such evidence soon after Effectiveness, with the exception of Ho, where issues arose after the contractor moved onto the site; after 3 years the problem in Ho was resolved. 6. Delays in completing civil works. Out of 60 contracts awarded under the project, none was completed on schedule, with some of them lasting beyond 18 months, instead of the 6 months stipulated in the contract, and 4 were terminated. The quality of work by some contractors was unsatisfactory. The reasons for this are nationwide, but much worse for contracts administered by MLGRD and DAs than the Ministry of Roads and Transport. There is little incentive for contractors--as well as supervising consultants--to complete the construction on time, as the contractual provisions (such as liquidated damages) are not enforced, delays are the norm, and payments to contractors are often seriously delayed. The Government and the industry failed to put an effective monitoring system in place to regulate the performance of contractors and consultants. 5.3 Factors generally subject to implementing agency control: 1. Staff turnover in LGPSU was stable throughout, and the experience gained in managing previous projects contributed to the satisfactory civil works implementation performance achieved. 2. Rising inflation compelled LGPSU to introduce price adjustment formulae in civil works contracts, reducing the risks to contractors. 3. The presence of a Financial Management Specialist in LGPSU, even on a short-term basis, could have improved the performance of the RIAPs program. Both MLGRD and LGPSU did not find the need to recruit one. 4. Inadequate outreach to Zonal Officers and District Assembly staff by MLGRD and project unit contributed to the modest gains on the institutional building component. After the abrogation of the MoU (Section 4.2), the unit was mandated to manage the technical assistance component, but it was still preoccupied with civil works, and did not adequately provide the needed advise on institutional issues. 5. Inadequate coordination between the accounting and engineering staff in the project unit, and lack of financial forecast of outstanding project funds and required contingency sums, resulted in identifying excess funds belatedly at the end of each contract. This contributed to the extension of the credit and the cancellation of more than half a million US dollars of uncommitted funds after the closing of the Credit. 6. As two new urban projects (Urban Environmental and Sanitation and Urban V) were under preparation, attention shifted from implementation of LGDP to the preparation of the new projects, and this had adverse effects on some LGDP components. 5.4 Costs and financing: Total project cost, as estimated in the SAR, was US$55.51 million, comprising US$26.02 million (47 percent) in local cost and US$29.49 million in foreign costs (53 percent). Financing was to be provided as follows: IDA: US$38.51 million (69.3 percent of total project cost); KfW: US$8.29 million (15 percent); District Assemblies: US$3.16 million (5.7 percent); and GoG: US$5.55 million (10 percent). Final financing was US$52.28 million, comprising US$26 million in local cost and US$24.63 million in foreign cost. Final disbursement were IDA: US$36.88 million, KfW US$9.14 million, GoG US$3.08, DAs - 13 -  US$1.77 million and also beneficiaries contribution of US$1.41 million for household toilet, which was not included at appraisal. Out of total SAR counterpart forecast of US$5.55 million, US$3.08 million was actually received by the project unit as counterpart fund contribution, representing about 55 percent of forecast. A withholding tax amount of about US$0.8 million that should have been paid to GoG was used to meet a part of the counterpart fund shortfall. An unspecified amount was also waived in taxes and custom duties (import, ECOWAS tax, and destination inspection) during the importation of vehicles, waste management equipment and computers. It was not possible to assess the exact Government contribution associated with the granting of exemptions for taxes and duties, however, this could be substantial. On the whole, Government total financial contribution could be much closer to SAR's forecast. The total disbursed amount was reduced to US$52.28 million due to the following factors: i) US dollar appreciation against the SDR decreased the total IDA contribution from US$38.5 million to US$37.5 million; ii) undisbursed amounts of about US$0.6 million earmarked for cancellation; and iii) large foreign exchange ratio of payments that were allowed in civil works contracts, which increased IDA proportion for disbursements and reduced GoG proportions. The District Assemblies and beneficiaries were expected to contribute US$3.16 million being their share of the costs associated with the Personal Service Contracts, and Operation and Maintenance Fund. The total amount contributed was US$1.77 million due to the late start and early completion of the PSCS and O&M assignments. The total counterpart funds from Government stood at US$3.08m, the share from the District Assemblies was US$0.83 million, being their contribution to the Operation and Maintenance Matching Funds, which was what was affordable to them. Generally, actual cost was in line with SAR forecast, with the exception of consulting services and training. The cost of the former category of expenditures was higher as a result of more than 100 percent increase in the cost of consultant's supervision services. The cost of the latter was much lower than anticipated. The local currency, the Cedi, depreciated by about 1100 percent during the life of the project at an average annual rate of about 25 percent. The annual high depreciation of the local currency compelled the project unit to introduce price adjustment formulae in civil works contracts and this reduced the risk to contractors and consultants. 6. Sustainability 6.1 Rationale for sustainability rating: The sustainability of the project is rated as likely. 1. Since the project lasted for slightly more than 8 years, many of the activities provided under the project were completed long before the closing date. This has provided an opportunity to observe that most of them have performed satisfactorily. For example, since 2001 DAs have on their own employed the services of LVB to update their valuation rolls for the collection of property rates. Lorry parks are under the management of a private transport body, the Ghana Private Road Transport Union (GPRTU). Some facilities, such as markets, lorry parks, public toilets, generate their own maintenance revenues. Road maintenance is covered by the Road Fund. 2. The capacity of the DAs to prepare investment plans and implement projects has by now been well established. The investment and maintenance funds available to the Assemblies have substantially increased in recent years. Nearly all capital expenditures are financed through grants, which have mainly consisted of the DACF, donor-assisted project funds, and recently, HIPC funds. In 2002,