rl The World Bank S ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~J u L Y ;'''I!i 2002notes PUBLIC SECTOR Recent trends in lending for civil service reform Civil service reform contributes to good govemance and has implications for a wide range of World Bank projects. This note assesses how the volume, distribution, structure, and objectives of Bank lending for civil service reform have changed in recent years. In Bank operations, civil service reform usu- service reform lending have changed since ally refers to interventions that affect the a 1999 Operations Evaluation Department Even projects organization, performance, and working (OED) review of such lending covering conditions of employees paid from central, 1980-97. Only lending is considered; the without civil service provincial, or state government budgets. assessment does not cover economic and It excludes reforms that affect police, the sector work. reform components armed forces, public health care workers, primary and secondary public school teach- Volume of new lending are affected by Bank ers, and employees of state enterprises. The OED review found that between the (Administrative employees of ministries of late 1980s and 1997 the Bank financed a support for such health and education and other school growing number of standalone civil service administrators are included.) In addition, reform projects. But between fiscal 1999 reform civil service reform generally does not cover and 2001 only 4 of 62 civil service reform interventions involving employees paid by interventions were standalone; the rest were local or district governments. components of larger lending operations. Assessments of such reform have wide rel- (Data for 1998 were not included in the evance. The findings can strengthen the OED review or the review conducted for this Bank's civil service reform efforts. They can note.) also help client countries improve gover- Civil service reform appeared most often nance, because a well-functioning civil ser- as part of public sector management reform vice helps foster good policymaking, effective projects-with financial management reform, service delivery, and accountable resource decentralization, legal reform, and procure- use. Finally, the findings are important for ment reform as the other components- Bank sectors and projects that do not directly accounting for 26 of the 62 interventions. aim to improve governance. Although the After that health and economic policy oper- Bank is promoting efforts that give com- ations were the most likely to include civil ser- munities more say in how central govern- vice reform, each with 7 projects involving ment services are delivered, most Bank such reform. Other projects with civil service projects are implemented by the civil ser- reform components involved education, vices of client countries. So, even projects transport, private sector development, agri- without civil service reform components are culture, and social protection. affected by Bank support for such reform. Exact lending for civil service reform is This note assesses how the volume, dis- difficult to determine because the Bank's tribution, structure, and objectives of civil data on its operational activities are primarily FROM THE DEVELOPMENT ECONOMICS VICE PRESIDENCY AND POVERTY REDUCTION AND ECONOMIC MANAGEMENT NETWORK FIGURE 1 NEW CIVIL SERVICE REFORM PROJECTS AS A SHARE OF ALL had the fewest new projects involving civil WORLD BANK PROJECT APPROVALS, FISCAL 1999-2001 service reform, while other regions-par- ticularly Europe and Central Asia and South Percentage of all new Bank projects Asia-saw a significant increase in such 20 projects. Amount of new lending for civil service reform projects Structure of new lending 15 The Bank uses two main lending instru- ments. Standard investment loans have a 5-10 year focus and finance goods, works, 10 and services in support of economic and social development. Standard adjustment Number of new civil service loans have a 1-3 year focus and provide quick- 5 reforn projects disbursing assistance to countries in need of external financing, to support policy changes and structural reforms in a sector or the econ- 0 1999 2000 2001 omy as a whole. Of the 62 new loans involv- ing civil service reform during fiscal 1999-2001, 40 were investment operations Source: World Bank 2000a; http://www4.worldbank.org/sprojects. and922 were adju stment operations. and 22 were adjustment operations. In recent years the Bank has created new at the project level. Even loan documents investment and adjustment lending prod- do not fully separate the cost of civil service ucts, called programmatic loans, to respond reform components from other interven- to longer-term needs. Programmatic invest- tions. Thus efforts to measure the volume ment loans are used when a country requires of civil service reform lending must isolate a sustained path of investments, each build- specific reform components from the larger ing on the one before, and usually support projects and sectors of which they are a part. a phased program of sector restructuring. Between fiscal 1999 and 2001 the num- Adaptable Program Loans are the Bank's ber of-civil service reform projects fell in tool for such lending. Programmatic adjust- real terms-with 24 in 1999, 21 in 2000, and ment loans are used when a country needs 17 in 2001-and as a share of total Bank incremental policy changes over a longer projects approved (figure 1). (During these period (usually about 10 years), focusing on three fiscal years the Bank approved loans step-by-step capacity building. Programmatic for 806 projects.) But as a share of Bank Structural Adjustment Loans are the main lending, loans for civil service reform tool for such lending, although the new increased. Fewer projects and larger loans Poverty Reduction Support Credits will be suggest that civil service reform is increas- largely programmatic. ingly associated with high-value projects. Among new investment loans involving civil service reform in fiscal 1999-2001, 18 Distribution of new lending were Specific Investment Loans. Thus these During fiscal 1999-2001 the regional break- accounted for 29 percent of new loans down of new loans involving civil service involving civil service reform, making them reform was broadly similar to that in the the most common type. Technical Assistance OED review for 1980-97 (table 1). Sub- Loans accounted for 21 percent, Adaptable Saharan Africa had the most, with 23 new Program Loans for 11 percent, Learning projects, though its share of the total has and Innovation Loans for 2 percent, and fallen considerably: from 74 percent in Sector Investment and Maintenance Loans 1987-93 to 37 percent in 1999-2001. Latin for 2 percent. America and the Caribbean was second, with The 1999 OED review found that during 10 new projects. East Asia and the Pacific 1980-97 more than half (55 percent) of civil PREMNOTE 71 JULY 2002 TABLE 1 NEW CIVIL SERVICE REFORM PROJECTS BY REGION AND YEAR OF BOARD APPROVAL, 1980-2001 (percentage of total) Region 1980-86 1987-93 1994-97 1999-2001 East Asia and the Pacific 0 6 14 8 Europe and Central Asia 18 2 5 15 Latin America and the Caribbean 27 12 11 16 Middle East and North Africa 5 5 8 11 South Asia 0 1 3 13 Sub-Saharan Africa 50 74 59 37 Total 100 100 100 100 n=22 n=65 n=37 n=62 Note: Does not include data for 1998. Civil service reform Source: OED 1999; authors' calculations. projects focus on service reform interventions were compo- In investment operations, pay and grad- correcting fiscal nents of adjustment operations. That share ing structures are improved by reclassifying plummeted in fiscal 1999 and 2000, when jobs, providing training on human resource imbalances, just a quarter of new loans with civil service management, improving payroll manage- reform components were adjustment oper- ment, introducing human resource man- adjusting pay and ations. But in fiscal 2001 adjustment oper- agement information systems, and ations dominated new loans involving civil enhancing salaries selectively. Adjustment grading structures, service reform, accounting for 60 percent. operations pursue this objective through Overall, during fiscal 1999-2001 Structural policy reforms that change pay and grad- and improving Adjustment Loans accounted for 26 percent ing arrangements. of new loans involving civil service reform, Investment operations improve account- accountability and followed by Programmatic Structural Adjust- ability and service delivery by funding man- ment Loans (5 percent), Sector Adjustment agement training, information technology service delivery Loans (3 percent), and Special Structural and office equipment, appraisal systems, com- Adjustment Loans (2 percent). petitive performance improvement and challenge funds, and restructuring and Lending objectives reengineering. In adjustment operations such This note categorizes civil service reform efforts usually involve strengthening meri- objectives under three broad headings: cor- tocracy and user voice, organizational restruc- recting fiscal imbalances, adjusting civil ser- turing, and accountability improvement. vants' pay and grading structures, and During fiscal 1999-2001 adjustment oper- improving accountability and service deliv- ations showed a broad balance between the ery. (The OED review, by contrast, grouped three main reform objectives, with most pro- civil service reform projects into those that jects pursuing all three. Investment opera- downsized the civil service, built adminis- tions focused on accountability and service trative capacity, and reformed institutions.) delivery, but always in tandem with other The same objectives are pursued differently structural reforms. Training and provision depending on whether a civil service reform of information technology and office equip- intervention is part of an investment loan ment were rarely undertaken unless they or an adjustment loan. supported a larger structural reform. Correcting fiscal imbalances through Regionally, objectives differ substantially. investment operations involves funding civil Projects in Latin America and the Caribbean service censuses or redeploying employees. focus more on fiscal imbalances and pay Through adjustment operations it involves and grading structures than do projects in cutting wages or employees. other regions (table 2). PREMNoTE 71 JULY 2002 TABLE 2 OBJECTIVES OF NEW CIVIL SERVICE REFORM PROJECTS BY REGION, FISCAL 1999-2001 Adjusting Improving Correcting pay and accountability Total fiscal grading and service Region new loans imbalances structures delivery East Asia and the Pacific 5 1 1 5 Europe and Central Asia 9 0 5 9 Latin America and the Caribbean 10 6 7 7 Middle East and North Africa 7 2 5 7 SouthAsia 8 3 1 8 Sub-Saharan Africa 23 8 8 19 Questions remain Total 62 20 27 55 Note: Many projects have more than one objective. about the financing, Direction of future analyses Further reading structure, and If the trends in 1999-2001 are representa- OED (Operations Evaluation Department). tive, this note raises some issues for further 1999. Civil Service Reform: A Review of World effectiveness of civil consideration. First, financial supportfor the Bank Assistance. Report 19599. Washing- technical components of civil service reform ton, D.C. service reform loans is being provided through investment lend- World Bank 2000a. Annual Report 2000. ing, in parallel with larger programmatic Washington, D.C. [http://www. adjustment operations. Programmatic adjust- worldbank.org/html/extpb/annrep/]. ment operations strengthen political incen- . 2000b. Reforming Public Institutions tives for long-term reforms, while investment and Strengthening Governance: A World Bank lending provides more detailed support. Strategy. Washington, D.C. [http://wwwl. What are the optimal combinations of such worldbank.org/publicsector/strategy. financing, and under what circumstances? htm]. Second, any examination of new lending also raises questions about the shape of the This note was written by Ranjana Mukherjee underlying portfolio. While this note has (Consultant, Public Sector Group, PREM Net- not updated the OED review of the effec- work) and Nick Manning (Lead Public Sec- tiveness of civil service reform operations, tor Specialist, Public Sector Group, PREM their record of effectiveness appears mod- Network). est. Many operations approved prior to this Ifyou are interested in similar topics, consider period were still active in fiscal 1999 and joining the Administrative and Civil Service 2000. Taken together, what is the structure Reform Thematic Group. Contact Ranjana of the overall portfolio? If the portfolio is Mukherjee (x34301) or click on Thematic Groups changing, what impact is that having? on PREMnet. _ _ This note series is intended to summarize good practice and key policy li nk U findings on PREM-related topics. The views expressed in these notes are those *L | | |of the authors and do not necessarily reflect the views of the World Bank. P itflflfl | PREMnotes are distributed widely to Bank staff and are also available on flfl Mthe PREM website (http://prem). If you are interested in writing a PREM- note, email your idea to Sarah Nedolast. For additional copies of this PREM- note please contact the PREM Advisory Service at x87736. PREMnotes are PMvReduclonandlconmicManausnent edited and produced by Communications Development Inc. Prepared for World Bank staff