How can civil service reforms in programs supported by the World Bank and the IMF, particularly in low-income countries, be made more effective? On September 6, the IMF’s Fiscal Affairs Department and the Public Sector Group of the World Bank’s Poverty Reduction and Economic Management Network cosponsored a workshop on civil service reform with the goal of reaching a consensus on this question. This workshop was the second thematic discussion held in the context of strengthening the two institutions’ collaboration on country programs and streamlining conditionality. About 60 staff, including managers and country teams from the Bank and the IMF, participated. The first workshop, held in July, focused on privatization of public enterprises (see IMF Survey, September 3, page 283).
In her opening remarks, Teresa Ter-Minassian, Director of the IMF Fiscal Affairs Department, noted the limited success of civil service reform. According to an IMF evaluation of Structural Adjustment Facility and Enhanced Structural Adjustment Facility programs during 1985–95, some wage bill reductions were obtained in the short run, but insufficient emphasis on strengthening institutions made it difficult to sustain employment and wage reductions. Cheryl Gray, Director of the World Bank’s Public Sector Group, emphasized that civil service reform is crucial for improving governance and reducing poverty and that there is scope for strengthened collaboration between the Bank and the IMF to ensure consistency between the sometimes conflicting goals of short-term fiscal discipline and longer-term structural reforms.
Challenges of civil service reform
In the first session, participants discussed the motivation for civil service reform. The purpose of such reform is to ensure that payments for wages and pensions do not create fiscal imbalances and that incentives for civil service workers are adequate to ensure efficient service and good governance. In general, Sanjeev Gupta (IMF, Fiscal Affairs Department) noted, IMF-supported programs have assigned high importance to the wage bill and pension payments to ensure macroeconomic stability and sustainability. This was the case in Bank-supported programs as well. The difficulty with this emphasis is that it ignores structural objectives and fails to set priorities in the medium term, he said. Mike Stevens (World Bank, Africa Region) noted that there is no blueprint for civil service reforms because each country may require a different sequencing of reforms.
Participants generally stressed the need to better understand the political environment in which reforms are designed. In particular, Barbara Nunberg (World Bank, East Asia Region) observed that, in many cases, governments have been unable to resist pressures to increase wages or to provide employment as a social safety net to some segments of the population. Yasuhiko Matsuda (World Bank, Latin America and the Caribbean Region) and Amit Mukherjee (World Bank, Europe and Central Asia Region) pointed out that the public often supports the demands of underpaid civil servants for higher wages and does not support civil service reform. Cyrille Briancon (IMF, African Department) noted that strong labor unions may make it difficult to change wage and employment practices. Furthermore, insufficient “country ownership” of reforms has delayed the implementation of policies and therefore reduced the effectiveness of the reforms.
In summarizing the challenges, Helen Sutch (World Bank, Public Sector Group), who chaired the session, stressed that civil service reforms must be justified to policymakers and the public on the basis of their impact on poverty and on the effectiveness of government.
Bank and IMF approaches
The second session, chaired by Ke-young Chu (Deputy Director of the IMF’s Fiscal Affairs Department), examined civil service reform in Bank-and IMF-supported programs. Describing how long it took for such reforms to bear fruit, Brian Levy (World Bank, Africa Region) pointed to Tanzania, where the current success took 10 years to achieve.
In the discussion, Valerie Mercer-Blackman (IMF, Western Hemisphere Department) noted that the timing of reforms is critical in some IMF-supported programs. Participants generally agreed that the timing and sequencing of reforms in Bank- and IMF-supported programs have sometimes been in conflict, in part because of differences in time horizons in the programs they support. In some cases, the need to reduce government spending has run counter to long-term structural reforms, such as the need to decompress wages and improve pay structures. Unless resolved, differences in strategy between the two institutions can reemerge as conflicts of timing.
Participants agreed that advice should be given in the context of a medium-term framework in order to assess all aspects of civil service reform and emphasized the need for early and frequent dialogue between the two institutions. Chu stressed that the macroeconomic and the structural components of civil service reform are inseparable and called for improved collaboration between the two institutions, particularly in light of the greater expertise of the World Bank in this area.
With regard to IMF conditionality, some participants questioned whether the IMF should impose conditions on civil service reform at all. In some cases, they argued, the wage bill may be the main fiscal problem, which would require that the IMF include conditionality to ensure that reforms actually take place. Moreover, according to Barbara Nunberg, IMF conditionality on short-term fiscal problems can lead to opportunities for longer-term reform and, in some cases, is critical in bringing about a consensus for reform. Nonetheless, in light of the track record with civil service reform to date, the sense of the meeting was that it was important that conditionality on future civil service reform be more selective and more sharply focused, with the Bank taking the lead on structural conditionality. Participants agreed that, to optimize Bank-IMF conditionality, care should be taken to ensure that the reform programs are consistent.
Chairing the final session, Gupta solicited participants’ views of the extent to which civil service reforms could be embedded, at least for low-income countries, in the poverty reduction strategy paper (PRSP) process. Claudia Costin (World Bank, Latin America and the Caribbean Region) emphasized the importance of engaging stakeholders early, being open about the strategy, and using high-quality data on employment and wages in formulating a reform strategy. For such a strategy to be implemented, the sequencing of civil service reforms needs to be tailored to each country’s circumstances. Nick Manning (World Bank Public Sector Group) noted the need for data on wage structures and sector-specific information that could be used to model the various reform scenarios. Participants also discussed various examples of countries in which data availability helped define the need for reform and convince the population of its benefits.
Summary and conclusions
Participants reached consensus on several key issues. First, they agreed that the staffs of the Bank and the IMF should engage in early dialogue during the implementation of civil service reforms in programs supported by the two institutions and should make the objectives of the reforms explicit from the outset. Second, the workshop concluded that short-run objectives–aimed at protecting the macroeconomic framework–and long-run objectives–linked to structural reforms–should be set in a medium-term fiscal framework, within which the fiscal impact of reform measures can be incorporated. Third, explicit recognition of the political challenges of reform is crucial. Fourth, conditionalities should be used sparingly. Consistent with the lead agency concept, the Bank would normally take the lead in advising on medium-term structural reforms and in designing conditionalities. Fifth, explicit discussion between the Bank, the IMF, and the government on timing and sequencing of reforms should precede decisions on key actions, such as civil service censuses, functional reviews, design of retrenchment programs, or monetization and consolidation of benefits. Sixth, national ownership is critical for the success of civil service reform. A first step in ensuring national ownership could involve identifying stakeholders and discussing with them what the tradeoffs, in terms of public sector provision and tax burden, would be in the absence of reforms. In low-income countries, this discussion could take place in the context of the PRSP process. Finally, further efforts are needed to identify the core data that are needed to support civil service reforms and, in particular, to model different reform alternatives.
To build on the findings of the workshop, Teresa Ter-Minassian and Cheryl Gray agreed that the Bank and the IMF would select a group of focus countries to begin enhanced collaboration in the area of civil service reform. Bank and IMF staff working on these focus countries would agree on the priority macroeconomic, structural, and governance objectives for civil service reform, taking into account the medium-term fiscal framework within which these reforms must take place. In addition, country teams will agree on the sequencing of reforms, conditionalities, and the core data to support the reforms.
Photo credits: Denio Zara, Padraic Hughes, Pedro Márquez, and Michael Spilotro for the IMF, pages 325–326, and 331–335; and Paul J. Richards for AFP page 327.