Ghana - Economic Management Support Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 18794 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA ECONOMIC MANAGEMENT SUPPORT PROJECT (CREDIT 2224-GH) January 13, 1999 Macroeconomics 4 Africa Region 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.  BOG Bank of Ghana BIWG Budget Improvement Working Group CAG Office of the Controller-Accountant General CEPS Customs, Excise and Preventive Services CPI Consumer Price Index CSPIP Civil Service Performance Improvement Programme CWIQ Core Welfare Indicators Questionnaire DFID Department for Intemational Development (UK) EMS Economic Management Support Project ERP Economic Recovery Program FIMTAP Financial Management Technical Assistance Project GDP Gross Domestic Product GLSS Ghana Living Standards Survey GSS Ghana Statistical Service IDA International Development Association IDF Institutional Development Fund IPPD Integrated Personnel Payroll Database IRS Internal Revenue Service MOF Ministry of Finance MOP Memorandum and Recommendation of the President MTEF Medium Termn Expenditure Framework NIRP National Institutional Renewal Programme NRS National Revenue Service ODA Office of Development Assistance (UK) OHCS Office of the Head of the Civil Service PAD Policy Analysis Division, Ministry of Finance and Economic Planning PUFMARP Public Financial Management Reform Programme SAC III Private Investment and Sustained Development Promotion Credit SAIS Structural Adjustment Institutional Support Project VAT Value Added Tax CURRENCY EQUIVALENTS: Currency Unit = Cedi IJS1.00 = 2,323.00 (June 30, 1998) Metric System FISCAL YEAR OF BORROWER: January I - December 31 Vice President: :Jean-Louis Sarbib Country Director: :Peter Harrold Sector Manager: :Charles Hurmphreys Cluster Leader: :Rocio Castro Task Manager: :Douglas Addison  FOR OFFICIAL USE ONLY TABLE OF CONTENTS Page Preface Evaluation Summary ..........................................................i PART I - Project Implementation Assessment .......................................................... I A. Project Objectives .I B. Achievement of Project Objectives .I C. Implementation Record and Major Factors Affecting the Project 7 D. Project Sustainability .9 E. Bank Performance .10 F. Borrower Perforance .12 0. Assessment of Outcome .13 H. Future Operations .13 I. Key Lessons Learned .14 PART II Statistical Annex ......................................................... 16 Table 1: Summary of Assessments .16 Table 2: Related Bank Loans/Credits .17 Table 3: Project Timetable .17 Table 4: Loan/Credit Disbursements .18 Table 5: Key Indicators for Project Implementation .18 Table 6: Key Indicators for Project Operation .20 Table 7: Studies Included in Project .22 Table 8A: Project Costs .23 Table 8B: Project Financing .23 Table 9: Economic Costs and Benefits .23 Table 10: Status of Legal Covenants .24 Table 11: Compliance with Operational Manual Statements .24 Table 12: Bank Resources: Staff Inputs .24 Table 13: Bank Resources: Missions .25 APPENDICES: Appendix A. Borrower's Assessment of Implementation Appendix B. Map 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.  IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA ECONOMIC AND FINANCIAL MANAGEMENT PROJECT (CREDIT 2224-GH) PREFACE 1. The Economic Management Support Project in Ghana (Credit 2224-GH) in the amount of SDR 10.7 million (US$15 million equivalent) was approved on March 28, 1991, made effective on July 19, 1991, and closed on June 30, 1998. The credit was extended three times from June 1996 to December 1996, then to December 1997 and again to June 30, 1998. An undisbursed balance of SDR 31,006.02 was canceled. 2. Preparation of this completion report was undertaken by the Resident Mission staff. Accordingly, there is no aide-memoire associated with a mission mounted from Washington. The report is based on materials in the project files, on discussions with Government officials associated with the project, and on information provided by the project coordinators. Additionally, excerpts from the Borrower's own Implementation Completion Report are included as Appendix A. 3. This Implementation Completion Report was prepared in November and December 1998 by Camille Lampart (AFMGH), and reviewed on December 22, 1998 in a meeting chaired by Peter Harrold, Country Director (AFC 10).  IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA ECONOMIC AND FINANCIAL MANAGEMENT PROJECT (CREDIT 2224-GH) EVALUATION SUMMARY 1. Introduction. The Government of Ghana embarked on a deep program of structural adjustment in 1983. In 1986 the need to strengthen the institutional capacity was recognized and the Bank initiated the Structural Adjustment Institutional Support Project (SAIS). By 1990, most of the first generation reforrns were complete, and the government was about to embark on more complex second generation adjustments such as rationalization of the public sector and civil service, and financial sector reform. The Economic Management Support Project (EMS) was designed primarily as a continuation of these efforts and specifically to support the reforms of the third Structural Adjustment Credit (SAC III)'. 2. Project Objectives. The project was expected to build analytical and administrative capacity in four substantive areas: economic policy analysis, fiscal management, domestic resource mobilization, and civil service reform. 3. Achievement of Objectives. The project, itself, was not successful in achieving improved analytical capacity in core economic management agencies. It did support the restoration of some administrative finctions, such as tax administration, accounting and audit. On fiscal management only very rudimentary steps were taken and the civil service reform remains largely uncompleted. Therefore the project cannot be said to have achieved its intended development goals. It did maintain the impetus, if not the capacity for later reform. 4. Implementation Record and Major Factors Affecting the Project. Spending on most components was completed on schedule by June 1996. The credit was extended three times from June 1996 to December 1996, then to December 1997 and again to June 30,1998-mainly to finance the Ghana Living Standards Survey (GLSS4). Most of the project outputs were eventually attained but the project did not achieve its developmental objectives. This disconnect can be attributed mainly to the complexity of the project (evidence by the wide breadth and lack of depth of its components); weak ownership of these developmental objectives; poor preparation and supervision; the implementing ministry's lack of capacity to manage such a complex project; and to the changing political and economic climate during the life of the project. 5. Sustainability. The government's analytical capacity should continue to improve in the medium-term because of continued reform efforts and external support. The long- term sustainability of any achievements will depend on the ability to retain skilled staff SAC III was formally named Private Investment and Sustained Development Promotion Credit (No. 2682-GH).  ii which, in turn, depends on the successful implementation of public service pay and grading reforms. 6. Bank and Borrower Performance. The overall performance of the Bank was unsatisfactory with the exception of project identification. Given the maturity of the structural adjustment program, and the fact that capacity building and institutional development are relatively slow processes, a more appropriate project sequencing would have aimed at improving implementation capacities even earlier2. Yet, the pressure for rapid preparation was significant and resulted in the project becoming a collection of somewhat unintegrated components, most of which were not comprehensive enough to achieve the stated outcomes. There was no project appraisal. In retrospect a full appraisal was necessary in light of the fact that there was not much experience with this type of project and that several of the reforms depended heavily on continued strong political commitment. The project was insufficiently supervised with a new task manager and one mission each year on average and an inadequate mix of skills. The Borrower's performance is rated as satisfactory mainly because preparation and compliance with legal covenants was adequate. Implementation was unsatisfactory mainly because of weak project management and flagging government commitment to project objectives. Between 1991 and 1998 the structural adjustment program was interrupted by two major lapses in fiscal discipline which severely undermined both the civil service reform and fiscal management components. 7. Assessment of Outcome. The overall assessment is rated as unsatisfactory because the project did not achieve the intended developmental impact. Although the government is now able to perform some functions better than at the onset (namely accounting, auditing, records management and tax collection), the major objectives of civil service reform and improvements in financial management were not achieved by the EMS project. The project did, however, safeguard the impetus for future and more comprehensive reform efforts. The challenge will be to maintain the momentum and political commitment for the difficult changes. 8. Lessons. Institutional development and capacity-building projects should be underpinned by comprehensive long-term strategies and implementation plans. These are slow processes that need to be closely supervised by the appropriate professionals. They require commitment and ownership not only of the project activities, but of the intended outcomes, and not only at the policy making level, but down the line in the executing agencies. Up-front assessments of political commitment and full stakeholder analyses are strongly recommended. 9. The project was also too broad, with insufficient depth in individual components. This had dual effects. On the one hand, the project was a complex management task that exceeded the capacity of the implementing agency and the Bank's supervision teams. On the other hand, the lack of depth made it difficult to fully address and resolve the fundamental issues of capacity building in a difficult environment characterized by structural adjustment and fluctuating political support. 2Ghana Country Assistance Review, 1996, p.50.  iii 10. Some of the predominantly policy issues in this EMS, namely civil service reform, public financial management and expenditure control, may have been more appropriately addressed through policy reform instruments than technical assistance. Although the EMS did have strong and deliberate links with the SAC III, a policy instrument, necessary reforms and conditionalities were not sufficiently defined up-front. When the structural adjustment program went off-track it was precisely this lack of definition that allowed needed reforms to be discounted and postponed further.  IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA ECONOMIC AND FINANCIAL MANAGEMENT PROJECT (CREDIT 2224-GH) PART I: - PROJECT IMPLEMENTATION ASSESSMENT A. PROJECT OBJECTIVES 1. The broad objectives of the Economic Management Support Project (EMS) project were to assist the Government of Ghana in strengthening its capacity to implement economic refonns and sectoral programs and to promote a more efficient use of public resources. This was to be accomplished by improving the analytical and administrative capacity in core economic management agencies. 2. The broad objectives of the project were appropriate and timely. The government had embarked on a deep Economic Recovery Program (ERP) in 1983. By 1990, the majority of the first generation structural adjustment reforms had been completed. The ERP was about to shift to more complicated second generation reforms such as the rationalization of state-owned enterprises, revision of investment regulations, civil service and tax reforms. These activities would require greater public sector capacity for economic analysis and administration. In addition, the 1990 Country Strategy Paper identified the need to improve public sector management and implementation capacity as a major priority. The EMS was therefore designed to support the third structural adjustment credit (SAC III)3 and to continue the technical assistance started under the previous Structural Adjustment Institutional Support Project (SAIS). 3. The specific objectives of the EMS were to build administrative and analytical capacity in four key economic management areas: (a) economic policy analysis; (b) fiscal management; (c) domestic resource mobilization and (d) civil service reform. A schedule of key project activities and expected outputs was attached to the Memorandum and Recommendation of the President (MOP). The total cost of the project was estimated at US $21 million equivalent with US$15 million financing from International Development Agency (IDA) and US$4 million from the Office of Development Assistance (ODA). B. ACHIEVEMENT OF PROJECT OBJECTIVES 1. Broad Objectives 4. The project, itself, was not successful in achieving improved analytical capacity in core economic management agencies. It did, however, support the development of subsequent initiatives and operations which appear to be making some progress. The Research Division of the Ministry of Finance (MOF), responsible for economic policy analysis, has only recently received strong political ownership. There are still SAC III was formally named Private Investment and Sustained Development Promotion Credit (No. 2682-GH).  2 weaknesses in the capacity for effective aid and debt management and deficiencies in the analytical capacities of Ghana Statistical Service (GSS). There were strong and continued efforts to improve the capacity for analyzing human resource needs, but these were specific to the now concluded redeployment efforts and were concentrated at a central agency, rather than being spread more widely across the public service. The civil service reform program remains largely uncompleted. 5. The project did successfully restore and improve some administrative functions, viz tax administration, accounting and audit through the clearance of backlogs, very basic computerization, and the resumption of regular training programs. On fiscal management, i.e. budget preparation, expenditure control and cash management, only very rudimentary steps were taken through the EMS. More recent efforts through the Public Financial Management Reform Programme (PUFMARP), which was partially funded in its design stage under EMS, have made progress in this area. 6. Overall, the project cannot be said to have achieved its intended development goals. However, it did enable several agencies to rebuild their capacity for day-to-day functioning by financing the acquisition of necessary inputs. It also maintained the impetus, if not the capacity for later reform. 2. Specific Objectives 7. Economic Policy Analysis. The objective of this component was to strengthen the Policy Analysis Division (PAD) in the Ministry of Finance; to improve the quality, timeliness and dissemination of economic statistics; and to improve the capacity for aid and debt analysis. The project was also to provide continued support to poverty monitoring through the development of the capability for carrying out studies on the social dimensions of adjustment; the extension of the Ghana Living Standards Survey (GLSS) for four years; and by strengthening the design and follow-up of poverty reduction programs and projects. 8. Policy analysis. Most of the recommendations of the initial study on strengthening the Policy Analysis Division of the Ministry of Finance (now renamed the Research Division) were not implemented. There were long delays in the appointment of the director and then frequent turnover of personnel. Computers were purchased, but the commensurate training in even basic spreadsheet skills was not acquired. The unit still lacks a strong base in macroeconomic analysis even though a number of staff members were given post-graduate training both locally and overseas. Attrition among trained staff remains a grave problem. 9. Since 1997, there have been several improvements. The Division has expanded its work to involvement in the policy dialogue and negotiations with donors, including the IMF and the Bank, and to monitoring the implementation and impact of reforms. A Macroeconomic Coordinating Group with representatives from the revenue services, the Bank of Ghana (BOG) and the National Development Planning Commission meets regularly to guide the work of the macroeconomic modeling unit. There are plans to produce a new macroeconomic model in April 1999. The Agriculture Unit is  3 coordinating the formulation of the government's new cocoa sector strategy. In addition, a computer skills training program is underway. 10. Quality of economic statistics. A social accounting matrix was designed with ODA support in 1993 and has been used to rebase the national accounts and Consumer Price Index (CPI). The GSS is quite advanced in its preparation of a producer price index. Data on private economic activities is still rudimentary. Although the GSS continues to conduct enterprise surveys, the poor response by firms (endemic in Africa) has rendered the results non-representational and they have not been published since the early 1 990s. 11. Timeliness. The GSS produces only two regular economic statistics publications - the Quarterly Digest of Statistics and a monthly newsletter on the CPI. The publication lag for the CPI monthly newsletter was reduced from 4 to between 2 and 3 weeks. Provisional national accounts estimates are available within 9 months and official accounts within 21 months. The latter compares with the IMF target of 10-14 months. National accounts for 1996 were published initially in the July 1997 Quarterly Digest of Statistics produced in late September 19984. Although the printing equipment purchased by the project in earlier years did accelerate the publication of data, the technology is now outmoded and has been replaced by newer methods. Publication of other periodic economic reports, e.g. quarterly and annual external trade statistics, is at most sporadic and reflects some skepticism about their demand. 12. Dissemination. The dissemination of reports is done by mass mailing, radio announcements and by a number of periodic seminars around the country. The National Committee of Producers and Users of Statistics was not revived as mentioned in the MOP, mainly because it was not provided for in the budget, but the GSS regularly holds subsector meetings for government users. 13. Poverty monitoring. There were long delays in the production of social and poverty monitoring statistics. The Bank postponed the planned 1994 implementation of the GLSS4 because it was felt that not enough had been done to disseminate and analyze the results of the first three rounds. In 1995, a number of studies financed by ODA were published to accomplish this. The GLSS4 was further delayed by a decision to combine it with a labour force study being designed under another Bank project and the need to seek commitments for additional funding. To fill the long lapse in social statistics, a Core Welfare Indicators Questionnaire was implemented in Ghana between July and September 1997 and a report published and disseminated. The GLSS4 is now being implemented. 14. Debt and aid management. The planned extension of the debt management system to aid management was partially achieved. The project financed the extension of the Debt Recording and Management System to the Bank of Ghana and the Office of the Controller-Accountant General (CAG). However, the system was never networked and broke down at the CAG in November 1997 and at the Bank of Ghana in May 1997. In The current Quarterly Digest of Statistics takes the reference date of the latest industrial production data rather than the date of publication. If other series that are included in the publication have more recently available data than industrial production series, these data are not published and reserved for later issues.  4 May 1998, they were still not functioning. The Aid Information Management System was only installed in 1998, but at the time of writing, data was still to be entered and it was not fully operational because of insufficient staffing. For external debt service payments, the coordinating mechanisms are excellent. For debt analysis and strategy formulation, inter-agency coordination and capacity are still weak. 15. Fiscal Management. The fiscal management component of the project was aimed at supporting improvements in budget preparation, expenditure monitoring and cash management and in the coordination of manpower and financial resource management in the civil service. 16. Budget preparation. The EMS project established the Budget Improvement Working Group (BIWG) and provided assistance to several units in the Ministry of Finance responsible for budget preparation. The main objective was the production of an integrated budget, which would include aid flows in order to encourage better programming of counterpart funds, and development expenditures and their recurrent costs implications. 17. A "broad-based" budget summary was published annually between 1992-1995 as a substitute for an integrated budget. It contained independent but not integrated summaries of recurrent expenditures, development expenditures, aid flows, debt flows and manpower ceilings for each ministry. The project clearly underestimated the requirements, both in capacity and time, for integrating the budget components. Very little was accomplished by the project on the computerization of the budget. At the time of writing, the budget was still being stored on stand-alone computers. 18. Expenditure monitoring and cash management. The project greatly improved the flow of information on government expenditures and the operations of the CAG and the Audit Service. A long backlog of audited government accounts was cleared and regular training programs for the two offices re-established. Since 1993, the government has been publishing annual public expenditure reviews, although these are not widely available except to the donor community. On cash management, very little has been accomplished. CAG utilizes a custom-built computer system to obtain total government cash position and has been assisted by the recent transfer of all government bank accounts to the Bank of Ghana. Improvements in cash management will depend partly on current efforts to strengthen expenditure control. 19. Coordination of manpower and financial budgeting. More effort and less progress has been achieved on the coordination of manpower and financial budgets. This component was to have been supported by -BIWG, the Office of the Head of the Civil Service (OHCS) and the installation of an integrated personnel payroll database (IPPD). 20. Until the 1999 budget exercise, personnel ceilings were established at manpower hearings at OHCS, while financial clearance for personnel changes was required from the Ministry of Finance. Although Ministry of Finance officials were required to participate in the manpower hearings, it is widely acknowledged that there was little connection between the establishment numbers approved and the financial resources provided for personnel expenses.  5 21. The new reform programs, Civil Service Performance Improvement Programme (CSPIP) and PUFMARP, appear to recommend the devolution of more authority on personnel and establishment to line managers while retaining control over personnel expenses. This eliminates one of the inconsistent constraints, but does not provide a mechanism for linking the remaining expenditure to operational targets. Little has been done to implement this change beyond abolishing the manpower hearings. The new expenditure control rules will allow virements into, but not out of, the line item for personnel expenses. Given the problem with controlling recruitment5 in the past, the management of this expenditure item will depend heavily on aligning personnel expenditures with operational targets and establishing effective performance sanctions for line managers. 22. The installation of IPPD was to have assisted in the coordination of manpower and financial budgets. It contains all civil service and Ghana Educational Service personnel records with the exception of the Fire Service, Highways Department and pension payments and has been utilized since 1995. The system has yet to achieve adequate operating standards. As a result, very few line ministries and agencies utilize it for other than record keeping purposes. There is a widespread impression that, even with the controls on data input and payroll output from the system, the ghost-worker problem has not been completely eliminated. 23. Domestic Resource Mobilization. The objective of this component was to increase tax revenues and lower costs of collection by strengthening tax administration and supporting the continued implementation of tax reform. These reforms were to be coordinated by the National Revenue Service (NRS) which had oversight of the two revenue collection agencies: Customs, Excise and Preventive Service (CEPS) and the Intemal Revenue Service (IRS). Specific outputs included the extension of the customs data computer system, ASYCUDA, to various ports and border points, computerization of IRS, training and public education associated with the reforms to the sales tax system and a program to enhance in-country training through the NRS. 24. Tax reform. Ghana had already undertaken extensive tax reforms under the ERP, which resulted in an increase in tax revenue from 5 percent of the Gross Domestic Product (GDP) in 1983 to 13 percent in 1991. Tax revenues increased from an average of 12 percent GDP in 1989-92 to 16 percent of GDP over the period 1993-98. The majority of this component was spent on studies for NRS, some of which were implemented as tax reforms, and for the initial preparation of the Value Added Tax (VAT) between 1993 and 1995. A study on the taxation on investment resulted in a The problem of recruitment was two fold. Firstly, the budget line for personnel expenses was based on current personnel numbers and rarely included provisions for recruitment to new posts. (This has now changed with the MTEF). Secondly, it is reported that many line agencies may simply fill posts allowed for in their establishment ceilings and engage the staff. Usually, the Ministry of Finance would eventually authorize payment of an employee who had already worked for a reasonable period, albeit without initial financial clearance. It is also reported that a policy of automatic engagement of graduates from several public training institutions often superseded, albeit after the fact, the voted budget allocations for personnel emoluments. It is not clear whether the second problem has been corrected with new expenditure and personnel rules.  6 reduction of dividend tax from 15 tolO percent and the establishment of a capital gains tax of 5 percent in 1992. 25. The implementation of the VAT began with a pilot credit system for the sales tax on paper, which was initiated in March 1991. By mid-1992, the Government committed itself to implementing a full-fledged VAT and much of the supervision efforts of the Bank revolved around an International Development Fund (IDF) grant to support these efforts. The later withdrawal of the tax was mainly due to three factors: (a) weak government capacity to implement such a complex reform, (b) a very tight implementation schedule and (c) poor timing of the launch of the tax. The capacity needs were clearly underestimated by the government, the Bank and the IMF. 26. Modernization of tax administration. The ASYCUDA system was extended to the Takoradi and Tema ports in 1993. Plans to install the system at Aflao and Elubo border points were shelved because of delays in procuring electricity generators for those sites. An untenable situation exists with regard to the data provided by the system. Although CEPS supervises the data entry staff at the ports, it neither relies on nor reports data from the ASYCUDA system. Instead it re-tabulates the hard data collated by hand from hard copies of customs documentation. Not surprisingly, there are large and unsystematic discrepancies between the data reported by CEPS and the ASYCUDA data used by the Ghana Statistical Service. 27. The proposed training strategy was implemented at the IRS and the CEPS between 1991-1993. As a result both agencies have been able to reactivate their traditional training programs for entering staff. The feasibility study for the IRS computerization was completed in early 1993 and revealed that the project had severely under-budgeted this component. Some efforts were made to secure additional financing, but eventually this component was suspended. 28. Civil Service Reform. The main objective of this component was to support the Government's Civil Service Reform Program in continuing the civil service redeployment (retrenchment) program, reforming the pay and grading structure, strengthening personnel management practices, and improving staff development and training programs. In reality, the EMS extensively supported the downsizing program within the core ministries but left largely unfinished the more critical objectives of improved manpower control and pay and grading reforms. 29. Redeployment and manpower control. The EMS supported the redeployment program through job inspections, management reviews, improved monitoring of line agencies' establishment and better records management. However, lax control of recruitment across most of the government service has reduced the overall effect, leaving public service, including subvented agencies, with even greater manpower numbers than in the past. 30. Human resource management. The project underwrote many activities to strengthen personnel management practices, including the establishment of merit pay and performance appraisals systems, which were required under the accompanying adjustment credit. These efforts did not yield any effective changes because of an  7 inconsistency between increasing manpower control which centralized key human resource authority in the OHCS, and improving human resource management in the line agencies which would have required delegating powers to line ministries to institute new practices. CSPIP will attempt to introduce performance contracting and devolve more human resource authorities to the line ministries and agencies. However, it still lacks appropriate instruments for linking personnel resources to operational targets. While there is a substantial capacity-building component in the CSPIP, the test will be in the enforcement of contracts and expenditure rules. 31. Pay and grading reforms. The government has yet to implement the major pay and grading reform that has been pending since before the EMS project. The original plan was to execute the redeployment program aimed mainly at reducing the staff numbers in lower levels of civil service and then to decompress the salary structure. The Redeployment Program was closed in 1998. However almost 70% of the retrenchments were completed before the EMS. Interim steps were taken to decompress the civil service salary structure from 1:4 to approximately 1:14 and to reconcile the grading systems between the Ghana Education Service and the Civil Service. A comprehensive reclassification exercise was canceled in 1994 because of the cost implications. In 1996, the government commissioned the latest study on establishing a uniform pay and grading structure for the entire public service. The grading exercise is now underway and the incomes policy is currently under consideration by the National Institutional Renewal Programme (NIRP). 32. Staff development and training. There were several achievements on the training front including the merger and refurbishing of several institutions. Alongside the training given to staff in the economic analysis, audit and accounting divisions, and the revenue services the extensive training of human resource professionals throughout the service, constitutes a major achievement. In 1995, a comprehensive training policy for the civil service was completed. It is not clear whether this has been implemented. C. IMPLEMENTATION RECORD AND MAJOR FACTORS AFFECTING THE PROJECT 33. There are a number of factors which contributed to the unsuccessful outcome of this project. They are listed below in order of importance. 34. The project had too many components. The project suffered from too broad an array of components, none of which were sufficiently deep to effectively address their respective problems. In addition, there was no coherent strategy linking the various components together, even though some of their outputs were interdependent, e.g. civil service reform and fiscal management. Other components were basically independent from the core objectives, e.g. improvement in economic statistics, and contributed to diluting core efforts. 35. Ownership of project objectives was weak. Although some components were overseen by high level coordinating committees (viz civil service reform and budget improvement) the approval of detailed annual work programs was limited simply to approval by the Deputy Minister of Finance and by the Bank. In turn, this made the  8 project outcomes very vulnerable to changes in the political climate. Indeed, there were significant delays in the accompanying SAC III, and the government did not complete the civil service and public expenditure reforms that were included both as conditionalities in the credit and as activities in the EMS project. 36. The project was inadequately prepared. The pressure for rapid preparation was significant to the point that the project was deemed sufficiently appraised at the pre- appraisal review meeting. This was not the case. Given the poor implementation experience of the SAIS, more attention should have been given to the design of implementation arrangements, the breadth and depth of the various components and the risks of declining political commitment6. Indeed, several of the issues raised during the pre-appraisal review meeting were to later plague the project: (i) the ineffectiveness of the governnent procurement agency; (ii) the need for information technology skills in the supervision team; and (iii) a concern about the tight implementation schedule. The project would have also benefited from the institutional, stakeholder and leadership assessments that are currently recommended. (This is further discussed in Section E.) 37. Although the project linked its technical assistance to the conditionalities of the accompanying SAC III, most of these reforms were not sufficiently defined up-front. When the structural adjustment program went off-track, it was precisely this lack of definition that allowed the more loosely defined reforms of the civil service and public financial management to be discounted. 38. Bank supervision was inadequate. The resources spent on supervision were insufficient. There were only 9 supervision missions and there were too many changes in task managers over the life of the project (on average one per year). There were also inadequate skills for managing the many different components. As a result, supervision concentrated more on procurement and supplier performance issues, and less on assessing the progress toward improved capacity. (This is further discussed in Section E.) 39. Implementation capacity was weak. Because of the poor project management of the SAIS, there were significant changes in the implementation arrangements for the EMS. The original project management unit was dissolved and the responsibility placed under the World Bank Desk in the Ministry of Finance. Although this brought remarkable improvements in the administration of project records and accounts, there was not sufficient capacity to supervise such a wide range of activities. The latter would have required significant institutional and capacity building efforts for this unit. (This is further discussed in Section F.) 40. Neither the Government's nor the Bank's project management teams had the broad range of skills to manage this project with its myriad components. In particular they lacked the skills to effectively assess the work programs of individual agencies and to manage the many computerization programs in the project. Much of the 6 One of the key lessons cited by the SAIS completion report is that the environment for institutional development is more important than the choice of instruments applied in technical assistance operations and that governance- related constraints to institutional development are fundamental and binding.  9 computerization was done in a piecemeal fashion with the results that the majority of these systems are today underutilized or only partially operational. 41. Economic and political factors. At several periods in the project, economic and political imperatives ran contrary to several of the objectives of the project. In particular, the lapses in fiscal discipline in 1992 and 1995 were partly the result of large wage increases given to the civil service in the run up to national elections. These increases along with the surrounding political climate severely undermined both the civil service reform and retrenchment programs and were indicative of the government's fluctuating commitment to expenditure control. D. PROJECT SUSTAINABILITY 42. Improvements in Analytical Capacity. The capacity to analyze data in GSS, OHCS and the Ministry of Finance should continue to grow in the medium term because of stronger reform efforts (NIRP, CSPIP, PUFMARP) and continued external support. However, the sustainability of any achievements will depend on two key conditions which are far from secure - the ability to train and retain staff, and the political and institutional demand for objective analysis. 43. Recent developments at the Research Division of the Ministry of Finance augur well for continued improvements in analytical capacity. The unit is now the direct responsibility of the Deputy Minister of Finance and is supported by PUFMARP. Training is underway in econometric analysis and macroeconomic modeling. CSPIP continues to build capacity in human resource management through OHCS, but the impact and sustainability will depend on the resolution of the manpower and financial controls, and establishing the mechanisms for linking human resources with line agencies' operational goals. GSS is the recipient of a number of externally-funded projects aimed at raising its analytical capacity to a level at least commensurate with its capacity for data collection. The recently formed Debt Strategy Working Group hosted an intensive training workshop in October 1998 to formulate a comprehensive debt reduction strategy. 44. Through PUFMARP, the 1999 budget was prepared using a 3-year Medium Term Expenditure Framework (MTEF) and has fully integrated recurrent and development expenditures and includes external project flows. The MTEF exercise has improved budget planning methods across the public service. However, without adequate expenditure management and control, the budget will revert to its usual status as an indicative target rather than a binding target. 45. The reform of expenditure control has not yet been tackled and current delays in the implementation of a computer system for budget and public expenditure management are not encouraging. The fact that effective expenditure control has eluded the government for so long is an indication of the complex political economy of this reform and supports the earlier-stated notion that it may require a policy instrument (to create more demand for good fiscal management) in addition to technical assistance. It is also important to acknowledge that the reforms taken to date are partly due to pressure from external donors.  10 46. Improvements in Administrative Capacity. The maintenance of the various skills bases and training programs established through the project will depend on the ability to retain skilled staff. The implementation of pay and grading reforms are necessary but not sufficient to secure this and the capacity to manage these programs in the face of budgetary constraints and performance criteria remains to be seen7. 47. All of the major computer systems installed or extended by the project are dependent on continued external support for their effective operation. Although there are sufficient programmers for the IPPD, none of these persons have received formal training on the software. They were all hired after the training from the supplier had been completed, without any handover from their predecessors, and have leamed to manage the system through trial and error. There are current problems in accessing technical support from the supplier, with delays of up to three months for in-country service. The Department for International Development (DFID) is currently supporting the necessary improvement to the system through a follow-on project. The present situation concerning the ASYCUDA system at CEPS and GSS is untenable because only a handful of people have effective technical understanding of the system. CEPS still does not have a dedicated IT unit, even though it is planning to embark on a new computerization program. E. BANK PERFORMANCE 48. The overall performance of the Bank was unsatisfactory with the exception of project identification. 49. Identification. The Bank's role in project identification was satisfactory. In 1990, improvements in public sector management and implementation capacity were priority objectives for Ghana. The choice to target the implementation capacity for adjustment programs rather than investment projects8 was appropriate given the primacy of adjustment programs in the portfolio. The need to reinvigorate civil service reform, to make human resource management more effective, to strengthen expenditure management and to improve tax collection were all identified as relevant activities. 50. The breadth of the EMS was guided by a need to make headway on multiple fronts to support the reforms in the accompanying structural adjustment credit and to complete existing technical assistance activities. Later insights about the complex process and slow pace of building capacity would have suggested a smaller project with fewer components and more depth in each of the components. Indeed, any one of the four components could have been a project in itself. 7The following anecdotal evidence is not promising. The beneficiary which spent the largest share of its disbursements on training recently commented that while project resources allowed them to pay for the retraining of candidates who had not successfully completed their training courses, one of the difficulties that they are facing, now that the project is over, is the need to rescind this policy. 8The latter would have entailed more activities to build project management skills and improve procurement systems.  11 51. Preparation. The Bank's performance in project preparation was unsatisfactory. The pressure for rapid preparation was significant. The result was that many of the components were not sufficiently far-reaching. The project became a collection of often unintegrated components most of which were not comprehensive enough to achieve the stated outcomes. Although, the project objectives were correctly stated as the capacity 'to do what"9 for many of them priorities and implementation details were unclear. 52. The choice of annual work program reviews as an implementation arrangement allowed a lot of flexibility to reformulate project activities. This design relied heavily on high level advocacy of the BIWG, the oversight committee for civil service reform, and the NRS to select and secure outputs, making the project vulnerable to shifts in commitment and ownership. In retrospect, it would have benefited from the institutional, stakeholder and leadership assessments that are currently recommended by the Bank. In addition, the SAIS project unit was wound up and responsibility for the EMS transferred to the World Bank Desk where capacity and experience in managing the implementation of a complex project was limited. After the poor implementation experience of the SAIS, more attention to the design of these arrangements was warranted. 53. There was no project appraisal. This was because the country team felt that the results of the pre-appraisal sufficed. However, the identical issues raised during the pre- appraisal review meeting were to later plague the project: (i) the ineffectiveness of the government procurement agency; (ii) the need for information technology skills in the supervision team; and (iii) a concern about the tight implementation schedule. In retrospect, a full appraisal was necessary in light of the fact that there was not much experience with this type of project in Ghana, and that several of the objectives (viz, civil service pay and grading changes, tax reform, and the strengthening of expenditure controls), depended heavily on strong continued political commitment. More ownership and involvement of the line staff in the beneficiary agencies may have mitigated the risk of changing political priorities. 54. Supervision. The Bank's performance in project supervision was insufficient and unsatisfactory. There were seven task managers over the life of the project, on average one per year. There were only 10 supervision missions over the life of the project and most of these were shared with other projects and preparation of adjustment credits. At the time, the choice to share supervision of the EMS with that of the adjustment program was justified. In retrospect however, the skills were not appropriate for a capacity building project with such a broad array of components. In addition, supervision was concentrated heavily on the binding legal issues of procurement and supplier performance, and less on assessing the progress toward improved capacity. Even so, procurement problems persisted throughout the life of the project and were only addressed in a constructive manner towards the end, with the by-passing of Ghana Supply Commission for UN procurement arrangements. These problems had persisted through the SAIS and were identified during the preparation of the project. However, because of 9The current wisdom on capacity building and institutional development is that, rather than concentrate on the creation of necessary conditions and generic inputs like information technologies, revision of procedures, and the establishment of new institutional arrangements, projects should define their objectives as the capacity "to do what".  12 a tight preparation schedule, the Ghana Supply Commission was not included as a beneficiary in anticipation that it could be added later. This was not done. It is also noted that the supervision missions' assessments of performance in the Form 590s did not accurately reflect project performance of development objectives - hence the disconnect between this ICR and the ratings in table 13. F. BORROWER PERFORMANCE 55. Borrower performance is rated as satisfactory. Preparation and compliance with legal covenants were satisfactory. Although implementation was unsatisfactory, the reasons for this have to do mainly with the design of the project, implementation arrangements and the borrower's lack of capacity and experience with a project of this type and complexity. 56. Preparation. The borrower's performance in project preparation was satisfactory. The majority of the project components were related to ongoing efforts in which the government was actively involved and had support from other donors. A number of action plans, e.g. for tax reform, for improving dissemination of government statistics, and for civil service reform had already been prepared and formed the basis of the borrower's implementation plans. 57. Implementation and covenant compliance. The major weaknesses in the borrower's implementation concern project management and fluctuations in commitment to the project objectives. The World Bank Desk made remarkable improvements in accounting for project expenditures and administration of project activities, as well as strong efforts to pass on this learning to beneficiary agencies. As a result, covenant compliance was satisfactory. 58. The management of the project was limited to an annual presentation of work programs and procurement plans to the World Bank Desk, which obtained direct approval from the Minister of Finance, the Chief Director of the Ministry and the Bank. The weakness in this arrangement was that the supervision of a project with myriad but connected components required stronger evaluation mechanisms. This single line of command did not allow for broader involvement by the relevant professionals in overseeing the progress toward the specific project developmental objectives. This created the impression at many beneficiary agencies that the project was a financial support facility secured and managed by the Ministry of Finance, rather than as a program of activities aimed at specified outcomes and objectives. 59. Beneficiary agencies eventually achieved the capacity to access the project resources, without being held accountable for the achievement of project objectives. One result was that apart from the required share of certain categories of expenditure, it is not possible to quantify the govermnent's in-kind contributions to the project because these were never targeted nor accounted for in a formal way. Another result can be seen in the use of consultants. While the shift to local consultants was a welcome change from previous projects, they were used too often in a shortsighted fashion to perform functions that should have been the responsibility of permanent staff. As a result, the expertise gained was quickly lost to the public service. Delays in the recruitment of staff  13 for the IPPD led to a loss of valuable training and information when there was no handover period. The almost continuous attrition of trained staff from a number of agencies was not addressed, even with the renewed emphasis on civil service reform. Very few attempts were made to bond staff who received overseas training. 60. Wavering political commitment Between 1991 and 1998 the structural adjustment program was interrupted by two major lapses in fiscal discipline. The government's commitment to civil service reform flagged during the run-up to the two elections. As a result, very little progress was made on the pay and grading reforms and on the control of manpower expenditures. There were also a number of changes in the locus of economic decision making which resulted in the long deferral in the strengthening of the economic policy analysis functions. This led to the dissipation of resources on a consultant who produced four macroeconomic models that have been subsequently lost. In addition, the downgrading of the NRS over the life of the project represented a lack of consensus or commitment within the government to the concept of strong independent revenue management. This severely affected the uptake of resources for the component on domestic resource mobilization. G. ASSESSMENT OF OUTCOME 61. The overall assessment is rated as unsatisfactory because the project did not achieve the intended developmental impact. Although the government is now able to perform some functions better than at the onset (viz, accounting, auditing, records management and tax collection), the major objectives of civil service reform and improvements in financial management were not achieved by the EMS project. 62. The project did, however, safeguard the impetus for future and more comprehensive reform efforts. Institutional development is now at the forefront of public sector reform in various manifestations: the government's "Vision 2020" strategy, NIRP, CSPIP and PUFMARP. The lesson about the need for high-level policy oversight seems to have been internalized, even though coordination problems remain. There is clearly improvement in the capacity to formulate policies. The challenge will be to maintain the momentum and political commitment for the difficult changes. H. FUTURE OPERATIONS 63. There are two subsequent operations in public sector reform in Ghana. The Financial Management Technical Assistance Project (FIMTAP) aims to complete many of the objectives of the SAIS and EMS, by supporting PUFMARP. It emphasizes the need for longer term budget planning and has articulated in more detail the necessary elements of successful institutional change in public finance. However the project had been suffering from a limited capacity to manage the large number of components and from implementation delays. The Bank and government are moving to address this. 64. The Government and the Bank are currently preparing a Public Sector Management Reform Program to support the NIRP. The main objective is to extend many of the civil service reform objectives to the wider public service. It will support a comprehensive overhaul of the government's central administrative systems and  14 procedures and a complete restructuring of the public sector. An initial adaptable program loan will target the Ministry of Finance and core government functions as a pilot for the restructuring exercise. The preparation of this program has involved important actions to resolve coordination mechanisms up front. I. KEY LESSONS LEARNED Breadth versus Depth 65. The project was too broad with insufficient depth in any one component. This had dual effects. On the one hand, the project was a complex management task that exceeded the capacity of the implementing agency and the Bank's supervision teams. On the other hand, the lack of depth made it difficult to fully address and resolve the fundamental issues of capacity building in a difficult environment characterized by structural adjustment and fluctuating political support. Ownership 66. Commitment and ownership of the project objectives, and not just project activities, are essential for effective technical assistance. This commitment and ownership should exist, not only at the policy-making level, but also down the line to staff in executing and beneficiary agencies. Their involvement, especially at the preparation stage, can deepen and secure the stimulus for change. In addition, their involvement and ownership may mitigate the risk of wavering political commitment during project implementation. 67. A formal procedure for accounting for the government's own in-kind and financial contributions to project objectives can help to build ownership. Without explicit arrangements for this task, the project may be perceived as synonymous with the credit, and may be utilized only as a financial facility for the purchase of inputs, rather than as a program of integrated and purposeful activities. Preparation 68. The possibility that some issues might be more appropriately addressed through policy reform instruments (e.g. changes in institutional incentives) rather than technical assistance should be explored in the project design stage. Such a possibility was especially evident in the civil service reform activities supported by the EMS. For example, training in improved human resource management practices was not effective without an accompanying policy decision to give trainees the commensurate authority and appropriate rules to manage their human resources. Likewise, the improved training programs have not created a sustainable increase in public sector efficiency because the public sector incomes policy still does not facilitate the retention of skilled staff. Finally, the design and preparation of performance contracts, while a useful capacity-building exercise, is likely to be ineffective without the policy decision to enforce performance sanctions.  15 Supervision 69. Capacity building is a slow complex process and requires close and in-depth supervision by the appropriately skilled professionals. Neither the government nor the Bank recognized and planned for this. 70. Implementation progress in capacity building projects should be judged against outcomes or incremental capacity targets, e.g. the ability to complete specified tasks, rather than against schedules for the provision of outputs/inputs, like civil works, completion of training programs or procurement. Capacity 71. A formal institutional assessment should be required when a follow-on project makes significant changes to the implementation arrangements from a predecessor project. This assessment is necessary to verify that the new arrangements are adequate. In the EMS, it was a mistake to assume that the World Bank Desk had the capacity to manage and coordinate an elaborate project with multiple objectives and beneficiaries because it had previously demonstrated the ability to monitor and evaluate structural adjustment programs. 72. Information technology skills should be a required component of government and Bank project management or supervision teams when computerization is involved. It is important to note, however, that the provision of information technology infrastructure cannot substitute for well-designed, explicit and effective rules and the political commitment to implement them. Political and Economic Climate 73. Fluctuating government commitment to key reforms, in the face of difficult economic conditions, adversely affected project performance-notably in the civil service component. These risks were not well assessed during project design. Future projects should assess such risks, build in compensating design elements and include exit strategies.  16 IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA ECONOMIC AND FINANCIAL MANAGEMENT PROJECT (CREDIT 2224-GH) Table 1: Summary of Assessments A Achievement of Objectives & bslanfial PaorJi Negligiblek NotApplikl Macroeconomic policies' | | _ | X Sector policies2 X Financial objectives X Institutional development X Physical objectives X Poverty reduction X Gender issues X Environmental objectives X Public sector management X Private sector development _ X 1. Macroeconomic policies include fiscal management. domestic resource mobilization. 2. Sector policies include civil service reform and economic policy analysis. B. Project Sustanability Likev nluikely Uncertain x C. Bank Performance High(y Satisfactory Saftsfactory DeflieNt Summary Assessment X Identification X Preparation assistance X Appraisal r x Supervision _ _ X D. Borrower Performance Highly Satisfacdory Deficient _______________________ Satisfadory Summary Assessment X Preparation _ X Implementation X Covenant compliance X E Assessment of Outcome High44 I I Ul d I .u y AssessmenSatf " SXWWor U sF adoy U&W&dM Summary Assessment X  17 Table 2: Related Bank Loans/Credits Preceding operations 1. Second Phase of 2nd phase of structural adjustment program: 1989 closed Structural Adjustment trade, exchange rate and tax reforms, public Prograrn expenditure management, statc enterprise and civil service reforms. 2. Structural Adjustment Capacity building to support structural 1987 closed Institutional Support adjustment reforms. Project - . Following operations 1. Private Investment and Improve regulatory framework for private 1991 closed Sustained Development investment, state enterprise reform, tax Promotion (SAC 111) reform, public expenditure. 2. Private Sector Accelerated privatization program, 1995 closed Adjustment Credit improved public expenditure management. 3. Public Enterprise and Divestiture of public enterprises and 1996 ongoing Privatization Technical rationalization of subvented organizations. Assistance Project _ 4. Financial Management Improvements in public financial 1997 ongoing Technical Assistance management. Project 5. Public Sector Institutional development n.a. appraisal Management Reform Improved efficiency and effectiveness of Program - APL public services. Table 3: Project Timetable tes In projet cycle Dafe,p9awted Dae 491Dactc* ____ ____ ____ Latest esdsite Identification (Executive Project Summary) February 5, 1990 May 28,1990 Preparation October 1989 Appraisal' June 1990 June 5,1990/a Negotiations January 1991 January 21,1991 Board presentation March 1991 March 28, 1991 Signing June 20, 1991 Effectiveness July 1991 July 19, 1991 Project completion June 1996 June 30, 1998 Loan closing June 30, 1998 a. There was no formal appraisal. The decision, taken at the pre-appraisal review meeting, was that the project was sufficiently appraised during the pre-appraisal mission.  18 Table 4: Cumulative Credit Disbursements (USS millions) r __ __ __ _ FY92 FY93 F794 FY95 FY96 FY97 FY98 Appraisal estimate 4.1 8.0 11.2 13.7 15.0 as percent of total 27 53 75 91 100 Actual 2.0 3.7 5.1 9.0 12.3 13.9 15.3' as percent of total 13 24 33 59 80 91 100 Actual as percent of estimate 49 46 46 66 82 I. Increase due to changes in SDR/US$ exchange rate. Table 5: Key Indicators for Project Implementation Acll| Appr Acual Comments Esimate Economic Policy Policy Analysis Division Carry out consultancy study. 1991 1991 Study financed by DFID and completed before start of project. Ghana Statistical Service Set targets for publications. January 1991 Ongoing No targets were set. Expand and improve printing facilities. December 1991 New printing equipment Agree on plan to broaden dissemination. January 1991 purchased but now outmoded. Improve dissemination channels. December 1991 Dissemination improved through regular country-wide seminars. Debt Management Install machines and software and train staff at December 1991 1993 Machines installed. BOG and CAG. Expand system to cover domestic debt. December 1992 Fiscal Management Establish Budget Improvement Working Group January 1991 1991 Agree on TORs for requirements study for January 1991 Study completed, but computerized budget information system. computerization managed Carry out budget requirements study. May 1991 internally at MOF Budget Design system. August 1991 Division, resulting in Install systems. September 1991 1994 installation of stand-alone computers. Finalize recommendations to improve aid April 1991 1993-1995 Review completed in 1995. management. Implement recommendations. June 1991 1995 Renovate/equip residential training school for June 1992 1995 Completed and in use. CAG. Contract with consulting firm. designate Project February 1991 Major cost overruns in Manager. appoint Steering Group, Technical and software. Users Committees for integrated Poor timing of consultant personnel/payroll management system. services. Finalize user requirements. May 1991 Delays in resolving Select software and hardware. October 1991 procurement issues for Install hardware and software. November 1992 hardware. System design and development. March 1992 Current problems with Data conversion, parallel runs and system test. June 1992 maintenance agreement. New system. September 1992 Post-implementation review. March 1993 mid 1995  19 Table 5: Key Indicators for Project Implementation (continued) Acl | Appraa Acual | Conimats Domestic Resource Mobilization Extend ASYCUDA system. December 1993 partial Extension to airport and ports extension: only. Border points delayed 1993 because of lack of stable electricity supply. Finalize proposal to link ASYCUDA with December 1991 not done Canceled because of domestic sales and excise taxes. preparations for VAT. Install system. December 1993 not done Decision to implement a credit system of sales January 1992 Early 1991. Pilot ran from 1991-1993. tax. Replaced by VAT Implementation of credit system of sales tax. December 1993 preparations in 1993 Resumed after VAT withdrawal in 1995. Undertake requirements study for December 1991 August Study completed. computerization of IRS. 1993 Computerization not done Install system. December 1993 because of underestimation of not done cost. Civil Service Reform Prepare implementation program for introduction March 1991 Interim regrading of new grading system. implemented for civil and Implement new grading structure. January 1992 1992 education services. Establish Salary Administration Unit in the August 1991 1992 Done. Payroll and Personnel Management Division. Implement revised performance appraisal December 1991 1992 Done. system. Equip accommodation for Management Services July 1992 1996 Construction 1992-1996. Division of OHCS. Furnishing and equipment 1996.  20 Table 6: Key Indicators for Project Operation MajorAcdvty Expected Oitput Actual Oulcome Economic Policy Strengthen staffing of Policy Improvement of Ministry's capacity Division strengthened. but staff still lacks Analysis Division. to analyze policy issues, review basic computer skills and a strong base in economic trends and performance. macroeconomic analysis. Work program and provide advice on economic includes involvement in policy dialogues policy. and negotiations. Upgrade office technology to Improvement in quality, timeliness Timeliness of CPI improved. other data still improve access to sources of and availability of reports. and have long publication lags which exceed the data and speed up production increase in production quantities. IMF targets. of reports. Fiscal Management Effect further improvements Installation of a database Budget currently stored on 10 stand-alone in the budget process through management systems: integration PCs. Budget preparation on new computer automation. into budget of information on aid. system. Budget and Public Expenditure investment projects and actual Management System being developed by expenditures; switch over to a later operation. broad-based budgeting system. Extension of current debt Strengthened aid management in Computerization needs further training and management system to aid International Economic Relations equipment to be effectively operational. management. Division in MOF. Debt Strategy Working Group formed for coordination on aid and debt analysis between BOG. MOF and CAG. Expand computerization of Enhancement of computer main Not done. functions both at CAG frame systems at Headquarters and Headquarters and in the field. installation of computer works stations in regions. Support staff training in Completion of residential training Residential training facility in use and accounting procedures. facility and expansion of in-house resumption of regular training courses at training courses. CAG. Install an integrated Establish coordination between IPPD installed, but not yet effectively personnel/payroll management manpower and expenditure controls. operational. Coordination between system which would improve manpower and expenditure control not yet management of payroll. achieved. Current efforts in NIRP and PUFMARP to address this. Support modernization of Elimination of backlog in Backlog in annual audited public accounts methods for the examination submission of Auditor's annual eliminated. Timely production ongoing. and reporting of public reports and timely production accounts. thereafter. Improve facilities and access Resumption of regularly scheduled Resumption of regular training courses at to in-service training training programs to v'grade skills Audit Service. programs. of senior and junior level staff.  21 Table 6: Key Indicators for Project Operation (cont.) Domestic Resource Mobilization Higher tax revenues. Tax revenues increased from 12% GDP in 1989-92 to 16% GDP in 1993-1998. Streamlining of customs CEPS: Extend ASYCUDA network ASYCUDA extended to Kotoka. Takoradi procedures and lower costs of to Kotoka, Takoradi, Head Office and Tema only. Data used by GSS. but not revenue collection. and border points. by CEPS. IRS: link ASYCUDA systems to Not done, partly because IRS tax collection systems. computerization not completed. Support training aimed at NRS, CEPS and IRS: Development In-country training program developed and improving staff skills relating of an in-country training program ongoing. to tax collection, management tailored to the needs of CEPS and and manpower development. IRS. Civil Service Reform Continue support to civil Improvements in civil service pay New pay and grading structure designed, service reform program. and grading structure; increased but not implemented. IPPD installed, but efficiency and effectiveness of only partly operational. Few achievements personnel management systems and in improved personnel management. services. Redeployment completed, but control of recruitments and manpower numbers weak.  22 Table 7: Studies Included in the Project Computerization of Budget Implementation plan. Completed Recommendations not implemented and preparation. computerization of budget preparation handled in-house at Ministry of Finance. Revenue Service Information systems Completed Not implemented due to cost overruns and Computerization. feasibility study for IRS. change of priorities. Four macroeconomic Economic policy analysis. Completed None of the four models are currently in models. use. Aid Management Implementation plan for Completed Software system installed in 1995. Information System. aid management information system. Core Welfare Indicators Interim living standards Completed First national CWIQ in the world Questionnaire (CWIQ). survey. completed in 1997. which ended a lapse since 1993 in social data and allowed time to prepare GLSS4 for implementation in 1998. Various Management Job inspections and human Completed Completed redeployment program. The Services Division reviews resource needs assessments outsourcing of many of these studies and of civil service divisions, to inform redeployment the concentration of technical assistance departments and agencies' program and manpower on OHCS limited the building of hearings. widespread capacity in HR analysis. Broad Based Budget Guidelines for the Completed Ministry of Finance produced annual Summary. production of a "broad- budget summaries from 1992 to 1995 based" budget summary. when they were in anticipation of a fully integrated budget. 1. The Credit financed most of the work done at the Management Services Division of the Office of the Head of the Civil Service during the life of the project. This work comprised studies to evaluate the human resource needs of agencies in the civil service.  23 Table 8A.1: Project Costs by Category Category Apprisal E Actual Esdimate (USSM) (USSM) I.Consultant services & studies 2.75 3.63 2.Training 2.12 2.51 3.Equipment. supplies and materials 8.26 8.65 4.Operating & maintenance costs 0.32 0.45 Refund of SPPF 0.05 0.05 Contingencies 1.50 0.00 TOTAL (excluding taxes) 15.00 15.27' 1. Increase due to changes in SDR/US$ exchange rate. Table 8A.2: Project Costs by Component Appraisal estimate (SSM) IDA disbursements (US$M) Compnents Local Foreign Total Local Foreign Total costs costs costs costs 1. Economic Policy Analysis 1.12 1.61 2.73 n.a. n.a. 2.85 2. Fiscal Management 2.29 5.11 7.40 n.a. n.a. 9.49 3. Domestic Resource Mobilization 1.14 2.02 3.16 n.a. n.a. 0.70 4. Civil Service Reform 1.73 3.39 5.12 n.a. n.a. 2.18 Contingencies 1.09 1.58 2.67 n.a. n.a. 0.05 TOTAL (excluding taxes) 7.37 13.71 21.08 3.02 12.27 15.27 Table 8B: Project Financing Appraisal0;0x :f etme T ,, p i(UtSS( I Adual (USSM) __dl_ Local costs Foregn costs Total Loca costs Foreign costs Total IDA 5:22 9.78 15.00 3.02 12.27 15.29 UK ODA/DFID' 0.00 3.93 3.93 0.00 n.a. n.a. Government 4.10 0.00 4.10 2.42 0.00 2.42) TOTAL 9.32 13.71 23.03 n.a. n.a. n.a. 1. Despite efforts by the project team to coordinate ODA inputs, few government records are available on ODA contributions because these disbursements were made directly to suppliers. 2. Estimated cost for govemment counterpart funds and taxes. Records not available for in-kind contributions. Table 9: Economic Costs and Benefits It is not possible to do any empirical analysis of cost effectiveness because most of the developmental objectives were not achieved and because there are no consolidated records of the government's in-kind and financial contributions to the project activities (other than the required counterpart funds for certain categories of expenditure). The original project document does not include cost/benefit or cost effectiveness analyses.  24 Table 10: Status of Legal Covenants Original Actual Description Covenaw Present fuiffllment fulrilment Of Section type status Date date covenant 4.01 (a) Accounts/audit Complied Borroxwer shall maintain adequate records and accounts. 4.01 (b) Accounts/audit Complied 06/30/92 annual Requires annual audit of special accounts. 4.01 (c) Accounts/audit Complied annual Requires periodic statements of expenditure and supporting documentation. 3.03 Monitoring & Complied 03/01/92 annual Requires the Borrower and report the Association to exchange views annually on work program and progress. Source: OIS and Legal Agreements Library Table 11: Compliance with Operational Manual Statements Statement number and title Describe and comment on lack of conpliance 1. OP 10.02 Accounting, Financial Reporting and Auditing Reports received with minor delays. 2. BP 12.20 Special Accounts Reports received with minor delays. 3. BP 12.30 Statements of Erpenditure Reports received with minor delays. Table 12: Bank Resources: Staff Inputs Planned Actual Stage ofproject cycle Weeks US$'OOO Weeks USS'OOO Preparation to approval 7 51.4 114.7 Supervision 38 L 123.5 361.2 Completion 3 11.1 23.2 TOTAL 48 1 176.0 499.1  25 Table 13: Bank Resources: Missions _________ _________0X -___ I_ __ IPerfenm RaIinI Number Days Spedalized Prejee Cycle<; : : . Dat ef In StaffSkilL InlMea'n Develpment 1T)esof _________ ____________0R: Persons Fidd Rep reweted Satus Objives Probm Identification January 23, 1990 n/a n/a n/a Pre-appraisal June 2-12, 1990 2 8 EC.OA n/a n/a n/a Appraisala None n/a n/a n/a Negotiation January 14-25, 1991 2 11 PO. OA n/a n/a n/a Supervision I June 19 - July 5, 1991 3 16 EC. OA. IT NR NR Supervision 2 July 2-15. 1992 2 13 EC. OA 2 2 PW Supervision 3 November 8-20. 1992 2 12 EC. OA 2 2 PW Supervision 4 March 31- 9 April 9. 2 9 EC. PO. OA 2 2 PW 1993 Mid-Term January 31 - February 3 9 EC. CO. OA 2 2 PW Review 9,1994 _ Supervision 6 March 1-8, 1995 2 7 CO. IT S S Supervision 7 September 10-15. 2 5 CO. EC S S 1995 _ Supervision 8 June 14. 19966 2 10 CO. PO 5 S IC Supervision 9 July 8. 1997t 2 5 EC. PO S S IC Supervision 10 March 27. 1998 I 2 EC. OA S S Completion I I c EC U U a. No appraisal mission, the project was deemed sufficiently appraised at the pre-appraisal review meeting. b. Form 590 date, only. c. Implementation completion work was done by resident mission staff. Specialization Types of Problems CO = Country Officer IC = Implementation Capacity EC Economist PW= Procurement IT = Information Technology Specialist OA = Operations Assistant PO = Project Officer  26 Appendix A IMPLEMENTATION COMPLETION REPORT REPUBLIC OF GHANA ECONOMIC AND FINANCIAL MANAGEMENT PROJECT (CREDIT 2224-GH) The following pages are an excerpt from the Government of Ghana's evaluation report on the EMS project prepared by the Minister of Finance and received on December 21, 1998. The entire report is available in the project files in AFC 10.  In cam of repJy the namber ad daze of this MINISTRY OF FINANCE eDer? shouid bequotei. O. Box M 40 ot*gwO .r..ef.iE;.D a 4A ! f > ........ ACCRA. Your Ref No............. Tel. NO ....s.c OP. GHA.. WC.k....S 1Cj O .......... .. . ...... . .... ........ .......... IMPLEMENTATION COMPLETION REPORT ECONOMIC MANAGEMENT SUPPORT (EMS) PROJECT (CR. 2224 GH). As vou are aware the EMS Project formnally closed on June 30. 1998. 2. The Implementation Completion Report for the project has been completed and is herewith forwarded for your review and records. 3. You mav wish to peruse the document and let us have your comments please. (MR. M.A. QUIST THERSON) HEAD WORLD BANK DESK For: MINISTER FINANCE THE COUNTRY DIRECTOR WORLD BANK MISSION ACCRA  Document ot MINISTRY OF FINANCE IMPL