Since the early 1980s the government of Ghana has grappled with the need to reform the public sector. Several public service reform initiatives have been undertaken, but they have fallen short of the intended outcomes. In large part this occurred because political constraints prevented tackling the underlying problems: inefficient decision making, excessive staffing, lack of effective incentives and sanctions, and poor quality of services. Progress in resolving these problems was further impeded by a stop-and-go approach to reforms.
Early Reform Efforts
Early civil service reforms failed to attract sufficient attention from the Ghanaian leadership because of the large number of competing structural reforms vying for its attention and because of the early recognition that these reforms would be costly and, although generally desirable, not universally supported. As a result, existing civil service policies were allowed to linger on. In essence, Ghana’s initial reform efforts were driven by the executive. In the early 1980s President Rawlings came to power with an anticorruption agenda and devoted his energy to stamping out the squandering of the country’s wealth by politicians, traditional authorities, and public officials. However, the Ghanaian economy was in such difficult straits that for many years the attention of the government was focused primarily on restoring the basic functions of the economy, Ghana’s managerial depth was limited, and this constrained the number of initiatives that could be carried out simultaneously.
By 1987, however, it had become clear that the large and inefficient public sector was becoming a serious drag on the economy. The Civil Service Reform Program was launched to reduce overstaffing while increasing civil service wages to more competitive levels and decompressing the wage structure. The public sector wage bill was to be limited to about 4.5 percent of GDP, an objective achieved in 1989 but undone in 1992, and it has not been reached again since. The pay differential between the highest and the lowest paid civil servants was raised from 5.4 to 1 in 1988 to 9.4 to 1 in 1990 and to 10.5 to 1 in 1991. Also in 1991, preparatory work for the introduction of a new civil service grading system was completed and a performance appraisal system for civil servants introduced. Under the Civil Service Reform Program, some 50,000 mainly unskilled civil servants were removed from the payroll during 1987–90, although hiring of new staff meant that the net reduction did not exceed 30,000. To make this process possible, a “special efficiency budget” was established in 1987 and used to cover severance and retraining costs. The retrenchment continued in the following two years but increasingly became a mere exercise of transferring civil servants to subvented agencies. In 1991 the net reduction in civil servants was 6,300, In 1992, although the core civil service was reduced by 11,000, subvented agencies* staff increased by 8,000.
With the advent of democracy in 1992, efforts to reform the civil service slowed as the new parliament and even the government’s party were looking for safeguards for their constituents. Concerns about employment led many in parliament to voice their opposition each time a new initiative was presented for legislative approval. The solution found in 1994 was to adopt a National Institutional Renewal Program (NIRP), which put together initiatives to strengthen financial management, policy formulation, and implementation in the public sector; to introduce a new remuneration and grading system; and to streamline subvented agencies. A National Oversight Committee (NOC) composed of 20 members from the executive, parliament, and the private sector was appointed in December 1994 to serve as the executive board of the NIRP, charged with overseeing the work of the NIRP secretariat.
Unfortunately, the approach initially failed to restart the reform process. Although it broadened the participation of civil society in the public sector reforms, the NOC proved incapable of moving the reform process ahead. The NIRP secretariat did not present sweeping proposals to the NOC, because it was unsure of what the decision makers would consider acceptable, particularly since the NOC was primarily made up of officials with an interest in maintaining the status quo. The result was hesitation and a lack of direction, which hampered implementation and the establishment of a comprehensive action plan. An agenda of specific measures to reorganize functions, reduce staff, rationalize hiring practices, and decompress salaries was to have been completed by the end of 1995, but the deadline passed without much progress.
Recent Reform Efforts
In 1997, to jumpstart the stalled public sector reforms. President Rawlings appointed Vice President John Evans Atta-Mills as chairman of the NOC, in charge of overseeing all public sector reform activities. Since then the pace of reforms has increased, and political support has strengthened. In particular, the focus of the NIRP was changed to emphasize better coordination of the reform programs, particularly those aimed at making the public sector more compact and efficient: altering the compensation structure to enhance motivation, skills, and good governance; and establishing monitorable indicators of progress and a timetable for completing the reforms.
The public sector reforms carried out under the NIRP have covered a large number of subcomponents, some of which were started early but only gathered momentum after 1997. The most important have been the Civil Service Performance Improvement Program, the Public Financial Management Reform Program (PUFMARP), and the Public Sector Reinvention and Modernization Strategy (PUSERMOS). PUSERMOS is meant to consolidate and push forward the main themes of the public sector reform agenda. Other related programs included decentralization (Box 6.1), public service wage reform, state enterprise reform (see section VII), and the establishment of independent commissions to regulate the provision of public services.
Civil Service Performance Improvement Program
The Civil Service Performance Improvement Program, launched by the Office of the Head of Civil Service in March 1995, covers 190 organizations with about 81,000 employees.18 It addresses issues of motivation and performance, focusing on divisional and individual performance improvement plans, consolidation of the computerization process, and manpower ceilings. Feedback from managers resulted in the identification of 16 common problem areas in the civil service, and discussion groups have been set up to help resolve these issues. The program is scheduled to run until 2006; as of mid-1998, 125 organizations had passed the problem identification stage.
The provision of consistent and reliable data on personnel management is of crucial importance for human resource and financial management. The existing system– the Integrated Personnel and Payroll Database (IPPD)–which covers the core civil service, was introduced in July 1995. but migration from the old payroll system was only completed in December 1998. Unfortunately, the system has limited scope for expansion and was not year-2000 (Y2K) compliant. Therefore, an IPPD Consolidation Project had to be carried out with financial assistance from the U.K. Department for International Development. Its first phase aimed at making the system Y2K compliant and ensuring that the system meets the government’s needs until new software is put into place. The second phase envisages adopting new software and developing a replacement system, which should cover the entire public sector. The first phase was completed by December 1999. The second phase should start in the fall of 2000 and end in 2001.
Public Financial Management Reform Program (PUFMARP)
In 1997 the government embarked on a comprehensive effort to improve its procedures for budgetary planning, implementation, and monitoring under the PUFMARP. which was launched in July 1995. The program is supported by the World Bank and other donors. As a first step, the government designed a new budget and public expenditure management system (BPEMS), which is expected to be fully operational by the end of 2000. It will improve budget preparation and implementation, streamline cash management, and strengthen accounting procedures. Accompanying the BPEMS is the medium-term expenditure framework, designed to merge sectoral policy decisions with budget constraints (See Box 5.1). The sectoral policies are calibrated on a rolling three-year macroeconomic and budgetary framework.
The costs of activities are calculated on a per-unit basis, such as primary education for one child or rehabilitating a mile of feeder road, to determine what policy targets can be achieved within a ministry’s overall budget ceiling. The medium-term expenditure framework became fully operational in 1999, when the budget was prepared on the basis of a macroeconomic framework consistent with the Policy Framework Paper and budget guidelines for each entity, and consistent with available resources and covering a three-year horizon.
Decentralization Efforts
Ghana is divided for administrative purposes into 10 regions and 110 districts. The regions are directly controlled by the central government through the appointment of the regional ministers, but the districts are governed by district assemblies, two-thirds of whose members are elected. Within each region a regional coordinating council is made up of the presiding members of the district assemblies in the region and their district chief executives, under the chairmanship of the regional minister. This council is charged with coordinating development planning activities in the region and ensuring their consistency with national policies. The district assemblies (and the cross-sectoral planning groups) are, however, expected to be the basic units on which Ghana’s system of decentralized development planning is based.
The powers divested to the district assemblies were made clear by the 1992 constitution, which “commits the government to a decentralized public administration system which seeks to effectively transfer functions, powers, means and competence to local government bodies” (Government of Ghana, 1992, vol. II, p. 62). To provide the district assemblies the means to carry out that mandate, the constitution created the District Assembly Common Fund, which provides financing for development programs from central government contributions amounting to at least 5 percent of its domestic tax revenue. Donors, however, can also channel funds through the common fund, as is done, for example, in the case of the World Bank’s Village Infrastructure project.
Despite the government’s efforts toward decentralization, problems were noted early on. Many district administrations were fairly weak, and efforts had to be made to strengthen their capabilities for expenditure allocation and project implementation. Progress has been slow, and in 1995 the long-standing practice of funding half of local government staff salaries through the central government budget was abandoned in favor of full central government funding. This step was taken at least in part to allow the central government to regain a measure of control over the hiring policies of the district assemblies, which tended to absorb all available resources. Nevertheless, district tender boards have been established since 1995, and the award and payment of contracts up to limits set by the Ministry of Finance are now made locally.
A major new decentralization effort is presently under way. A Fiscal Decentralization Sub-Committee of the PUFMARP Steering Committee was set up in April 1999. This subcommittee will oversee the fiscal decentralization project undertaken by the government with assistance from the Canadian International Development Agency. Key objectives include identifying changes in the framework for financial management to adapt it to a decentralized environment; and defining the content, structure, and format of district composite budgets aimed at making the allocation of development resources more responsive to local needs, while ensuring transparency and accountability. Government agencies in a given region, district, and locality are expected to be merged, with the goal of providing decentralized services and avoiding duplication. The implementation of the district composite budgets is scheduled for 2000 and is expected to result in a single document listing all the development programs taking place in a district, whatever their sources of financing. A crucial aspect of the design of the district composite budgets will be to ensure that these budgets are fully compatible with the medium-term expenditure framework (See Box 5.1). A Local Government Act is expected to be presented to parliament in 2000, creating the administrative and accounting framework within which the district authorities should operate.
Public Sector Reinvention and Modernization Strategy (PUSERMOS)
In October 1997 the government issued its proposed strategy for public sector reinvention and modernization (Government of Ghana, 1997c). This strategy, approved by the cabinet in December of that year, provides a framework for integrating, sequencing, and monitoring reforms that address the key weaknesses of the public sector: lack of qualified staff in critical areas such as economics, finance, policy analysis, general management, and information systems, combined with inadequate incentives; a confused regulatory and organizational environment; and weaknesses in system planning, budgeting, implementation, and monitoring combined with an inefficient management regime (Box 6.2).
Public Service Wage Reform
In 1996 the Public Services Commission, with assistance from the international accounting firm Price-Waterhouse, began a review of the salary structure of the public service. The aims of the review were to eliminate ad hoc approaches to income policy in the short term; establish a medium-term remuneration policy based on a job classification exercise; and develop a wage structure for the long term in line with the objectives laid out in Ghana—Vision 2020.
The Public Sector Reinvention and Modernization Strategy
When the NIRP was created, information on the various public sector agencies was collected using a common review instrument, the Self-Appraisal Instrument. By the end of 1996, 250 institutions had been analyzed, and the information was used for a national workshop organized in May 1997. The workshop focused on coordination of reforms and development of human resources and was chaired by the vice president of Ghana, who had just been appointed chairman of the NOC. It developed a strategy for the next 10 years covering the broad spectrum of public sector reforms, titled the Public Sector Reinvention and Modernization Strategy (PUSERMOS). The cabinet approved this strategy on December 18, 1997, and the NIRP was instructed to begin a pilot phase to be completed by the end of 1999 (later postponed to 2001). Since mid-1998 the NIRP secretariat, with assistance from the World Bank, has developed action plans to revitalize public sector reforms. These plans were further refined by a number of focus group discussions and were approved by the NOC in December 1998.
PUSERMOS seeks to clearly delineate the functions of all components of the public sector, achieve efficient management and execution of these functions, achieve effective delivery of services, and instill good governance. In addition to redefining and clarifying the national policy process through a Policy Management Group in the Office of the President, PUSERMOS stresses reforms in two public service sectors: subvented agencies and central management agencies. These reforms are to be implemented over the next 10 years and would allow a significant decentralization of the government’s activities.
In support of these reforms, the World Bank approved in April 1999 a Public Sector Management Reform Project aimed at raising the efficiency, effectiveness, and quality of public services in Ghana, as well as improving overall fiscal performance in the medium term. The project will be implemented in three phases. The first phase (1999–2001) will concentrate on early steps in priority areas, including a pilot program for the closing down or restructuring of at least 17 subvented agencies, the realignment of central management agencies, development of a public-private partnership initiative, and implementation of a project management system for the reform agenda. The second phase (2002–2005) will extend the reforms to all levels of government and will achieve the closure or restructuring of another 70 subvented agencies and the realignment of 10 ministries and 25 departments. The third phase (2006–2009) will complete the process by dealing with the remaining subvented agencies, 11 ministries, and 25 departments.
Although the main focus of the strategy is not on downsizing but on reallocating functions within the public service, it is expected that the core civil service will contract. To avoid the social unrest often associated with civil service reforms, termination and unemployment benefits as well as costs of retraining have been built into the project. It is estimated that, in the first phase, 10 percent of the employees in the targeted institutions will be eligible for termination and unemployment benefits. These payments are expected to represent 57 percent of total project costs. Although it is difficult to forecast at this time the size of the eventual staff reduction, the project assumes that about 28,000 public servants will become eligible for retrenchment benefits over the course of the project. The World Bank will finance only part of these costs, and the government will need to take severance payments into account when drafting each year’s budget.
In December 1997 Price Waterhouse submitted a report with recommendations for the introduction of a uniform salary and grading structure based on tasks instead of positions. This proposal implies a salary realignment in the short term and a salary adjustment in the medium term. In 1998, while the salary structure and the slotting of civil servants in this structure were being discussed, the government granted an across-the-board 20 percent salary increase as a relief allowance for price increases, instead of the standard annual adjustment of the salary structure. In November 1998 a Central Management Board was constituted to oversee the implementation of the proposed changes that would introduce the new Ghana Universal Salary Structure (GUSS).
The government began the implementation of the GUSS as of January 1999. Negotiations were concluded with civil servants, teachers, judicial service workers, and registered nurses; together these groups make up about 83 percent of public sector workers. The negotiations resulted in an application of a relativity ratio of 2.5 percent, rather than the 2.75 percent recommended by the Price Waterhouse consultants, and in a graduated increase across the salary structure. The application of the GUSS for these four categories of workers required a structural increase of about 10 percent of the wage bill, and a general increase of about 15 percent, in 1999. Other public sector workers who would not go onto the GUSS in 1999 negotiated general salary increases between 10 and 12 percent. Negotiations with other public sector workers began in November 1999 and continued in 2000.
Independent Regulatory Commissions
As the government withdraws from the direct provision of services, both through privatization and through the removal of entry barriers, it is redefining its role to become principally one of regulation of these activities. In this context the government has established an independent Public Utilities Regulatory Commission to review pricing and standards in the electricity and water sectors; an Energy Commission to monitor pricing in the petroleum sector and issue licenses to independent power producers; and a National Communications Agency to oversee the telecommunications sector.19
Lessons Learned and the Future of Public Sector Reform
Public sector reform has come a long way in Ghana. In the process, the government has learned valuable lessons about program implementation and reform ownership. Aware of the political constraints under which it operates, the government is increasing its use of results-oriented mechanisms to bring about reform. It has learned that capable and motivated officials can make the difference. Therefore, recent reforms have tackled management problems, and this, together with the drive to privatize (as discussed in section VII), has allowed the government to strengthen public sector operations.
Not all problems have been resolved, however, despite the progress made. The speed with which public sector reform will be pursued in Ghana will depend on how stakeholders perceive the balance between the costs and the benefits of reform. Foremost among possible concerns is that of preserving employment opportunities. Efficiency measures that include downsizing of the work force will need to incorporate social safety nets that comply with the labor law (that is, severance pay) and allow for retraining of laid-off workers. The government must undertake special efforts to convince lawmakers and traditional authorities of the advantages of letting the private sector undertake some of the services that government now provides, while refocusing government activities on core issues, particularly education and health.
Also very important is the need for the government to rally public opinion in support of the reform program. This may involve more than just asking for the opinions of stakeholders: it may require an active effort to explain the benefits of the proposed policies. The government has heightened its efforts to explain the purposes and the benefits of its reform agenda to civil society. The PUFMARP Committee, for example, now publishes the PUFMARP News, a monthly magazine, which reports on the objectives and progress of PUFMARP activities as well as on other newsworthy government activities, such as success stories of the privatization program.
PUSERMOS, the most comprehensive public sector reform attempted in Ghana to date, will be crucial to attaining the financial targets under the Poverty Reduction and Growth Facility and World Bank programs. Modernization will greatly benefit Ghana’s government machinery, both by making it more efficient and by giving it the desired transparency. Without reforms aimed at containing public expenditure, Ghana will be hard pressed to create enough room for vigorous growth of its private sector and the economy. Similarly, streamlining the subvented agencies will be crucial to increasing the focus of government activities and instilling managerial responsibility. The size of the public sector and the time frame for the reforms will also need to be kept under constant review.
The experience gained thus far by the government bodes well for the future of public sector reform in Ghana, but the international community can play a crucial role in helping the government accomplish these difficult tasks. Financing of job classification programs in the civil service and of social safety nets, provision of technical assistance in privatizing strategic state enterprises, and help for government officials and the public in placing their concerns in perspective are all important areas in which donors can help push the process along. In doing so, donors will need to maintain a careful balance between encouraging the authorities to move ahead forcefully and ensuring that they do so in a well-thought-out manner.
Ghana’s experience makes it clear that although a democratic environment may pose some challenges to the reform process, it should not impede progress. Moreover, it may increase the public’s sense of ownership of reforms, making it less likely that they will be delayed or abandoned. However, success in this environment depends on a more participatory approach, strong leadership, good communications, and mechanisms to protect key stakeholders from the immediate adverse effects of the reforms. A key pitfall to guard against is allowing reforms to become excessively drawn out. Delay militates against sustained public support and encourages complacency among decision makers.