Document or · Public Disclosure Authorized The WORLD BANK FOR OFFICIAL USE ONLY CONFIDENTIAL Report No. 9827-ZA Public Disclosure Authorized REPUBLIC OF ZAMBIA PUBLIC SECTOR MANAGEMENT REVIEW VOLUME I Public Disclosure Authorized EXECUTIVE SUMMARY and MAIN REPORT October 31, 1991 Public Disclosure Authorized COFY t~l)!•J r', 1 D.S1'J'f i A;"_; h:etJc,i---L t~c). 83~. 7 -· ZP1 Tyr.·e: t ~:.~EC) CHANG-f,'0 Yi X34411 1 ~11117'1/ AF6<;0 ... Country Operations Division Southern Africa Department Africa Region Tim document bas a restricted dutribution and may be used by recipients only in the performance or their ofracial duties. Its contents may not otherwise be dbclosed without World Bank authorization. REPUBLIC OF ZAMBIA PUBLIC SECTOR MANAGEMENT REVIEW CURRENCY EQUIVALENTS Kwacha (k) YLl 1980 1.3 1 198S 2.7 1 1990 27,S 1 August 9, 1991 64.7 1 ABBREVIATIONS and ACRONYMS AG Auditor General (Zambia) 80Z Bank of Zambia CDEs Classified Daily Eq,loyees (Zad>ia) CG Consultative Group (Organization of Government and Donors) cso Central Statistics Office CY Current Year DETC Department of Economic and Technical Cooperation (NCDP/Zambia) DPE Total Expenditure Net of Interest Payments DPl,., District Plaming Unit (Znia) DPU Data Processing Unit (MoF, Zari:,ia) EmN Economic Cooperation and Debt Management Department (NCDP/Zambia) Ea.I Economic Cooperation Unit CNCDP, Zambia) ER Establishment Register (Zambia) ERC Economic Recovery Credit (World Bank) ERP Economic Reform Program (Zambia) FINNIDA Fimish International Development Agency (Finland) Fll>P Fourth National Development Plan (1989·1993; Znia) F.R.G Federal Republic of Germany FY Fiscal Year GDP Gross Domestic Product GRZ Governnent of The Republic of Zambia HIID Harvard Institute for International Development IBRD International Bank for Reconstruction and Development (i.e., World Bank) JDC Industrial Development Corporation (Zambia) INF International Monetary Fund IPR Investment Plaming and Research Department (NCDP, Zambia) It Kwacha (Zad>ian currency) LAS Local Adninistration Service (Zambia) LGER Local Government Establishment Register (Zari:,ia) LIU Loans and Investment Unit CNCDP, Zant>ia) NCC Menmer of Central Conwnittee, UNIP (Zambia) NGA Ministry of Agr:culture (Zambia) NoF Ministry of Finance (Zambia) NoPTC Ministry of Power, Transport, and Conm.mication (Zari:,ia) ICDP National Conmission for Development Plaming (Zanmia) IENIC National Economic Monitoring and Iq,lementation Conmittee (Zambia) IGO Non-Governmental Organization IIPA N,::onal Institute of Public Adninistration (Zambia) IJCEs t,; mal Joint Counci L Eq,loyees <Zanmia) IIORAD No,...egian Agency for Development ODA overseas Development Agency (The United Kingdom) OIN Operations and Maintenance PD Plarv,ing Division (MoA, Zambia) PER Public Expenditure Review PFP Policy Framework Paper PIP Public Investment Program PNO Prime Minster's Office (Zari:,ia) PPEU Project Preparation and Evaluation Unit CNCDP, Zant,ia) PPS Provincial Permanent Secretary (Zambia) PPU Provincial Plaming Unit (Zambia) PSN Public Sector Management PSRP Public Service Reform Program (Zambia) PU Plaming Unit CMoPTC, Zant>ia) RDCs Recurrent Departmental Charges (Zambia) RIU Revenue Inspection Unit (MoF, Zambia) SIDA Swedish International Development Agency Sim Special Investigation Team on Economics and Trade (SITET, Zambia) SPD Sectoral Plaming Department (NCDP, Zambia) UIIDP United Nations Development Progrmnne IIIFPA united Nations Family Plaming Association UIIDO United Nations Industrial Development Organization UIIP United National Independence Party (Zambia) USAID United States Agency for International Development ZCDJ Zambia Consolidated Copper Mines Corporation, Ltd. ZINCO Zambia Industrial and Mining Corporation, Ltd. iii REPUBLIC OF ZAMBIA PUBLIC SECTOR MANAGEMENT REVIEW VOLUME I TABLE OF CONTENTS EXECUTIVE SUMMARY •••••••••••••••••• •• •••••••••••••••••••••••••••••••••• •••••• •• ••••••••• •••••• •••••• vii I. INTRODUCTION •••••••••• ••••••••••••••••••••••••••••••••••••••• ••• ••••••••• ••••••• •• ••• ••••••• 1 Background Scope II. CURRENT PUBLIC SECTOR MANAGEMENT OBJECTIVES ••••••••••••••••••• 3 Medium-Tenn Policy Context Priorily Sub-Sectors Priorily Functions Short-Term Fascal Stabili7.ation Objectives Fiscal Targets Revenue System Reform Expenditure Restructuring Public Investment Program Public Sector Performance Objectives Phased Action-Program: Principles and Approach DI. RECENT FISCAL PERFORMANCE...................................................... 8 Public Expenditure (1975-1990): Background Composition or Expenditure Public Employment Non-Development Expenditure Development Expenditure Fmancing or Public Expenditure 1986-1989 1990 - Present '• iv 11/. PUBLIC SERVICE REFORM ••••••.•...••••..•.••••••••..•...•..•....••.••.••••............ 17 Objectives Approach Issues Policy Priorities Staffing Incentives Personnel Management Financial Management of Public Employment Process Recommendations Objectives and Phasing Short-Term: Recruitment Short-Term: Retrenchment Short-Term: Foundation/or Improved Performance Medium-Term: Improved Effectiveness Longer Tenn: Improved EjJiciency V. PLANNING, BUDGETING, AND FINANCIAL MANAGEMENT ............... 47 Introduction Issues Integration Budget Accuracy Staffing F.xpendilure Control: Background Accounting and Auditing Recommendations Phase 1: Expendilure Co11trol Phase 1: Resource Mobilization Phase 11 and Beyond VI. DECENTRALIZATION •............•.•••...••...........•.....•..•••.....•.........••...•.... 64 Current System Structure Functions Employment Issues Central Government Control Dejkil Fmancing Eco~omic Policies Recommendations Fiscal S:abi/ization Organizational A"angements Illustrative Conclusion VII. RECOMMENDATION: AN INTEGRATED STRATEGY ........................... 80 Program Integration Phasing Building-Blocks Short !':!· Longer Term V VII. RECOMMENDATION: AN INTEGRATED STRATEGY (continued) Technical Assistance and Training Technical Assistance (TA) Training Donor Coordination Integrated Management of The Reform Effort Assumptions and Risks Assumptions Risks Response 10 Risks FIGURE:;; III. l(A) Public Expenditure: Percentage Distribution 1980-90 (Subsidies & Constitutional) 10 III. l(B) Public Expenditure: Percentage Distribution 1980-90 (RDCs & Capital) 10 IV. 1 Optimum Level of Wage Expenditure 18 IV. 2 Public Sector Retrenchment: Sequencing Flow Chart 20 IV. 3 Comparison of Government & Parastatal Wage Scale 26 V. 1 Implementation of 1983 Recommendations 48 V. 2 Organii.ation and Functions Chart: Major Financial Management Functions; Central Responsibilities 51 V. 3 Budget Preparation Process 53 V. 4 The Accounting Process 56 VI. l Linking Subnational and National Planning Systems 67 VI. 2 Donor Support to PPUs (Provincial Planning Units) 68 VI. 3 Budget System and Procedure: Flow Chart at Provincial and District Levels 72 TABLES II. 1 Zambia - Key Economic Indicators 5 III. 1 Index of Total Discretionary Expenditure and Share of GDP 8 III.2 Public Expenditure by Economic Function (by percentage distribution) 11 III. 3 Public Expenditure by Economic Function (in millions of 1985 Kwacha) 11 III.4 Public Expenditure by Economic Activities (in millions of 1985 Kwacba) 13 III. 5 Public Expenditure by Economic Activities (by percentage distribution) 14 III.6 Share of Total Expenditure Provided by Revenue (percent, period averages) 15 IV. 1 Formal Employment, 1975 - 90 22 IV. 2 Public Service Employment 23 IV.3 Living Costs/Average Salary of Public Employees 24 IV. 4 Personal Emoluments 25 • vi TABLES {continued) IV. S Cumulative Costs of Housing Allowances 27 IV. 6 Projected Number of Housing Allowance Beneficiaries 28 IV. 7 Distribution of Civil Servants (1990) 29 IV. 8 Analysis of Public Employment and Benefits: Option 1 36 IV. 9 Analysis of Public Employment and Benefits: Option 2 38 V. 1 Authorized to Actual Staff: Financial Managehtent Functions 54 ATTACHMENT A. High Priority Sub-Sectors and Functions vii REPUBLIC OF ZAMBIA PUBLIC SECTOR MANAGEMENT REVIEW EXECUTIVESUMMARY I. SCOPE AND ORGANIZATION OF REPORT 1. The preparation of this Public Sector Management Review bas been undertaken with two operational perspectives in mind: (i) the need to improve the Government's capacity to perform its appropriate functions and (ii) the need to rationalii.e public expenditure to meet fiscal stahili7.ation objectives without seriously damaging the first of these two objectives. To meet both of those requirements necessitates a comprehensive understanding of the current situation in Zambia and a phased and clearly targeted approach to resolving iong- standing issues which avoids the •sin of comprehensiveness;• 1/ i.&., the diagnostic part of strategy formulation requires comprehensive understanding but the programmatic response requires a clear focus on limited actions to be effectively implemented. Those two principles have guided the scope and organi7Jltion of this Report. 2. The ultimate purpose of this Report is to assist the Government of Zambia to formulate a medium- term strategy for rationalizing its planning. budgeting. and non-parastatal public employment systems in the context of its Policy Framework Paper (PFP); including both short and medium-term action programs for the implementation of the strategy. Thus, the Main Report (Volume O fOC1.1SoeS on current issues in those three systems and provides recommendations for: (i) a comprehensive policy with prioritized objectives; (ii) a program disaggregated into short, medium, and longer-term phases; and (iii) an implementation process which maximizes the possibilities of efficiently integrating political and social factors with technical requirements. Given the scope of the Review, it is assumed that many readers might require additional background information. Therefore, several Annexes and additional statistical tables are provided in a Technical Annex (Volume m. Thus, Volume II includes more detailed background information and descriptions of current organi7.ational structures, procedures, legislation, and implementation policies. 3. The specific purpose of this Executive Summary is to provide decision-makers within both the Government of Zambia (GRZ) and the international donor community with an overview of the main issues affecting efficient and effective performance of the non-parastatal public sector in Zambia and the key themes for Public Sector Management Reform presented in the Main Report. The five main issue areas addressed in the Review, and, hence, this Summary, are: (i) fiscal constraints and the changing role of the public sector; (ii) public service reform; (iii) improvement of planning, budgeting, and financial management systems; (iv) the impact of Government's current decentrali7.ation program on economic management issues; and (v) the need for an integrated policy and action-program to address those issues. The Executive Summary concludes with remarks about assumptions, risks, and responses to such risks • • l./ See Arturo Israel, Institutional Development: Incentives for Performance (Baltimore: The Johns Hopkins University Press, 1987). viii II. FISCAL CONSTRAINTS AND THE CHANGING ROLE OF THE PUBLIC SECTOR A. Fiscal Constraints 4. The analyses provided in Chapter Ill clearly demonstrates that, given dwindling fiscal resources and a shrinking expenditure program, Government is no longer capable of performing all of the functions it bad previously assigned itself. That was not always the case because for many years copper revenue aHowed the GRZ to maintain an unusually high level of central go,·emment expenditure in its attempt to deliwr a wide range of services. With such resources available, the public sector dominated the economy through its parastatals and various policy interventions; which, in tum, requir&I it to perform a comprehensive set of management and technical functions. S. Since the mid-1970's, that earlier situation has changed dramatically. Nevertheless, despite declining government expenditures in both relative and re..l terms, public sector partic~~ation in the economy and its role and fonctions continued to expand until very recently. Thus, Government responded to the economic crisis by further tightening its grip on the economy through price controls, administrative allocations of foreign exchange (exc~pt during the period 1985-86), imposition of additional import restrictions and export bans, direct subsidies of agricultural production and marketing, and indirect subsidies of parastatals. 6. Of even more concern from the perspective of this Report is that, in addition to inefficient macroeconomic policies, Government's operational priorities resulted in an anti-developmental expenditure pattern. Specifically, increasingly limited resources were allocated to the public employment wage bill, subsidies, and maintenance of the military at the expense of non-wage recurrent operations and capital investment. In addition, the allocation of financial resources for development purposes did not conform to established priorities; being driven by either donor financed initiatives or across-the-board distributions. The result was misallocation of scarce resources and the inability to sustaui many post-investment operations. B. Chan1in1 Role or The Public Sector 7. Government bas responded to the fiscal crisis summariud above by embarking on a new Economic Recovery Program, the objectives of which are discussed in Chapter II. Suffice it to point out here that a basic principle inherent in that program is that the role of the public sector in the economy will be transformed from the primary producer of goods and services to: (i) the financier (but not necessarily producer) of public goods and services and (ii) the regulator of appropriate private sector economic behavior. 'l:l Thus, it is expected that the public sector will focus its attention on providing an enabling environment for the private sector to play an expanded role in a medium-tE>':'ID. recovery of economic growth and investment. 8. If Government is to give practical effect to its dramatically new economic policy objective and redefined role in the economy for the achievement of that objective, it must take into account the Iimitd resources available to it. That, in tum, requires fundamental decisions as to its investment and operational priorities; in terms of both sub-sectors and economic management functions. 9. Priority sub-sectors have been identified in Government's preliminary consideration of a public expenditure program and its established public investment programs. These priority sub-sectors share the following characteristics: (i) clear public sector comparative advantage; (ii) high priority economic and/or social return; and 2/ Although there are additio:.al functions which must be perfonned for the appropriate, but limited, regulatory role of the Government, such functions have not been addressed by this PSM Review. Thus, the Government should consider additional assistance for the design of an appropriate institutional framework and the efficient and effective perfonnance of such a regulatory framework. ix (iii) focus on requirements for the creation of a sustainable enabling environment for private sector development and growth. 10. Priority functions are those which tl;e Government must perform if priority public good~ are to be efficiently and effectively planned and financed. Such functions include: (i) identification of priority public goods and services (including assessment of its technical feasibility and economic viability); (ii) the specificatior of construction and/or operational requirements; (iii) preparing investment and recurrent financing plans; (iv) the generation of finance; (v) supervision of production of the good or service (which might be produced by the private sector under co~tract to the public sector); and (vi) evaluation of the financing and production of such goods and services for enhanced future planning and decision-making. 11. The principles und' , :ying the changing role of the public sector provide the ~ for the specific recommendations provided in the Report. Thus, priority should be given to improving the performance of those functions identified above (para. lC') in those priority sub-sectors included in its public investment and expenditure pro,1rams. Actions taken to streamline the public service and meet fiscal stabilization objectives through retrenchment should, during the initial three years of the recommended program, focus on those functions and those sub-sectors not included in the above priorities. From that perspective, emphasis in the Report is placed on: (i) mechanisms which the Government could employ for implementing retrenchment of personnel who are nonessential by that criteria (Chapter Iv) and (ii) improving the performance of the essential cross-cutting financial functions of planning, budgeting, and financial management (Chapter V). Recommendations for improving performance of unique sector-specific technical functions, including management and supervision, within each high-priority sector is not within the scope of this Review. III. PUBLIC SERVICE REFORM A. Obiectives 12. There are two criteria for determining the si1.e and distribution of non-parastatal public sector employment for achieving maximum efficiency and effectiveness 'J_/ First, the type and number of tasks which need to be performed to achieve Government's priority objectives. Second, the appropriate balance between wage and non-wage recurrent expenditure within the aggregate limits which Government can afford. Applying those two criteria, the strategy and action-program recommended in the Report set forth actions phased over the short, medium, and longer term along two separate, but equally necessary, tracks. The destination of the first track is a contribution to fiscal stabili1.ation in the shortest feasible time. The destination of the second track is significant improvement in the performance of Government operations over the medium and longer term. Travelling oniy along the first track would ultimately be self-defeating. Travelling only along the second track is ultimately impossible. Successfully travelling along both tracks, however, requires the changes in the role of the public sector outlined in the Main Report. 13. From a practical, shorter tenn, operational perspective, however, the immediate objectives of the PSRP are: (a) Design and initiate a comprehensive medium-term capacity-building program for the improvement • of public sector performance of key functions in high priority sub-sectors; 'J.I As used in this Report, the tenn effectiveness is defmed in tenns of capacity to formulate, manage, and implement measures to achieve policy objectives; the tenn efficiency is defined in terms of the expenditure of the least amount of human and financial resources sufficient to achieve those objectives. X (b) Identify specific l.mits on public expenditure for non-parastatal public sector employment to be effected through retrenchment of staff in non-priority sub-sectors; and (c) Determine the formula and procedures for redistributing savings from non-priority sub-sectors to staff performing key functions in high priority sub-sectors. B. Issues 14. Based on analysis of past experience and current conditions in Zambia, five fundamental issues emerge as requiring significant attention: (a) Absence of explicit policy priorities for public sector staffing; (b) An imbalanced structure of public sector staffing given the jmplicit requirements of current macroeconomic policy priorities; (c) Inadequate incentives for performance, including reliance on an inappropriate structure of personal emoluments; (d) Weak personnel and financial management; and (e) Absence of sophisticated understanding of the crisis in public sector employment among key stakeholders and mobilization of necessary support, through negotiations, for a program to implement acceptable solutions. C. Recommendations IS. Recommendations are presented in Chapter IV under the following six headings: (i) Objectives and Phasing; (ii) Short-Term: Recruitment; (iii) Short-Term: Retrenchment; (iv) Short-Term: Foundation for Improved Performance; (v) Medium-Term: Improved Effectiveness; and (vi) Longer Term: Improved Efficiency. The sequence of the recommended retrenchment program is based on the principle that targets should focus initially on those easiest to achieve, both technically and politically. Thus, the sequencing of the recommended program is essentially as follows: (a) Identify and eliminate "Ghosts" on the payroll at both central and district government levels; (b) End obligatory Government employment of all those trained at Government post-secondary training institutions; (c) Identify and begin to dismiss obviously redundant staff (such as the substantial ~xcess of drivers over available Government vehicles); (d) Identify those eligible for normal retirement or subject to expiration of contract who are performing non-priority functions and/or whose career streams are !!Q!!•priority and enforce such retirement/refuse to renew contracts; • (e) Design and offer a financially viable voluntary early retirement ra(kage for persons with more than 4 years to normal retirement who perform nor:-priority functions in any sector; (t) Design and offer a financially viable voluntary early retirement package for persons with more than 4 years to normal retirement assigned to non-priority sub-sectors; and xi (g) Dismiss other staff non-voluntarily only after all the above mechanisms are used and as fiscal requirements and/or effective performance at the organizational level demand them. That process is characterized by movement from the use of relatively crude instruments for decision-making to the application of carefully refined instruments for such decisions. IV. IMPROVEi> PLANNING, BUDGETING,AND FINANCIAL l'dANAGEMENT 16. Planning and revenue generation, budgeting, and financial management are key functions of Government regardless of its role in the economy. These various financial functions are essential for transforming Government's decisions into action. A. Objectives 17. An effective financial management system bas four broad purpo:ieS: (i) implementing expendil'Jre controls to ensure that public resources are expended only in amounts and on purposes legitimately approved; (ii) promoting efficiency in resource use and ensuring that, for a partic.llar program objective, minimum inputs are expended; (iii) promoting the effectiveness of governmental expenditures by ensuring optimum allocation of resources among programs and appropriately linking expenditures and objectives; and (iv) balancing governmental resources and expenditures over the long-term so as to ensure financial health and credit worthiness. 18. A complete financial system requires that all four purposes are met. The proposed strategy is designed to produce a complete financial management system capable of accomplishing all four purposes in the medium and loniier term. However, that strategy stresses the need for a phased approach, with initial emphasis on budget preparation and expenditure control. Other processes, such as strategic, program, and financial planning and resource mobilization (including taxation, debt management and foreign aid coordination) are understood, in the fiNt instance, as supporting the budget process. 19. It should be noted that Government bas attempted to implement the recommendation to establish an integrated Program Planning and Budgeting System lPPBS) as recommended in a 1983 World Bank report.~/ However, under pressure to respond to crisis &ituations, several elements of the recommended system have not been sustained. The primary lesson learned from that experience is that such an undertaking was, and continues to be, too ambitious given Government's current capacity. Further, the attempt to implement that recommendation reinforced a central planning approach to investment and expenditure BL-l resulted in a dispersion of focus and effort. That lesson has informed both the approach to, and specific recommendations for, reform of the planning and financial management systems. B. ~ 20. Although a myriad number of problems plague Zambia's planning and financial management systems, those requiring immediate attention are: (a) The absence of medium-term financial planning; • ~/ Planning and Budgeting in Zamb_i! (Washington, DC.: The World! ..ink, 1983), xii (b) The lack of realism in prepared budgets, particularly as regards actual expenditure intentions, resulting from weak costing and estimating techniques and the absence of actual prior year expenditure data during budget preparation and in the budget format: (c) Absence of effective budget ceilings, lack of basic and effective expendit1.1re control, and non- adherence to audit findings; and (d) Insufficient numbers of gualified staff. Underlying these issues is the limited capacity of the budgeting and fmancial management system to support the full range of four purposes identified above; as well as a systems design problem. C. Recommendations 21. Although a complete planning, budgeting, and financial management system requires that all four economic management purposes ue 11womplished, an improvement strategy requires priorities to be set and improvements phased. Therefore, the first priority during the short-term should be to improve budget preparation and expenditure control. The objective should be to ensure that fmancial policy-making and implementation is conducted in a transnarent and explicit manner, rather than the accidental result of many, uncoordinated actions by operating agencies (as is now largely the case). Budgetary and accounting systems should be wodified so that financial policy, as expressed in the budget, is proposed by MoF through established procedures and, once established, that compliance with such policies becomes, in actual practice, the norm. More specifically, Government should consider measures to improve the realism of proposed budgets, particularly revenue and cost estimates, and strengthen management of budget execution. Further measures to develop accounting, auditing, and resource mobilization need also to be introduced. V. DECENTRALIZATION 22. Alternative forms of public sector decentralization can have important differential effects on economic performance at both macro and sector levels. Although it can be argued th&t some forms of decentralization are, at least theoretically, necessary for the achievement of economic growth with equity, it should also be noted that decentralizing economic management functions can also result in maimaining llWlY historically rooted public sector inefficiencies. That latter result clearly characterizes the current situation in Zambia. A. Objectives 23. Government's current official policy is to progressively devolve~/ a wide range of responsibilities, currently performed by central Government ministries, to district councils. The objectives of that policy are, essentially, to: (i) improve management efficiency; (ii) improve financial performance; and (iii) increase popular participation in governmental decision-making. To those ends, the Local Government Act (1980), as amended, provides the legal basis for district councils to exercise substantial discretion in the performance of a wide • ~/ Devolution involves the transfer of discretionary responsibilities for a range of sub-sectoral activity to subnation&l levels of government, such as a district council. For example, if district level education officers are responsible directly to a District Council, perfonnance of the education functions assigned to that officer have been dev~ by the central government to the Council. If, on the other hand, the Officer is assigned to a subnational level, but remains responsible directly to the central government Ministry, authority has been deconcentrated. xiii range of responsibilities, which they are expected to finance largely by themselves. §/ Chapter VI provides a detailed discussion of Zambia's: (i) current decentralized organiutional structure; (ii) assignment of responsibilities for, and capacity to perform, the k"Y provision functions of planning, budgeting, and financial management within that structure; and (iii) current employment policy and practice at subnational levels. B. ~ 24. None of the three objectives specified above (para. 25), with the possible partial exception of improved channels for popular expression of demands for Government services, have been, or are likely to be, achieved under current circumstances. Systemic inefficiencies in public sector performance have been dispersed through decentraliution, rather than improved, and deficit financing of expenditure at the district level is having a significantly negative affect on Government's fiscal stabiliution program. District councils in Zambia have historically: (i) established parastatals, monopolized productive sectors and distorted the terms of private sector participation and competition; (ii) engaged in deficit financing of their expenditures; (iii) received preferential treatment by financial institutions and borrowed a disproportionate share of locally available credit; (iv) employed excess staff with inappropriate qualifications; and (vi) attempted to implement their responsibilities through inappropriate institutional structures and inefficient procedures. 25. Thus, under current conditions in Zambia, changing the structure of the public sector through decentrali1.8tion has not resulted in changes in macroeconomic policy behavior that could foster greater efficiency. Unless substantially new approaches are taken to the structure and objectives of Government's decentralimtion program, increased participation in public sector investment and service delivery decisions is likely to increase demand pressures on gov-,rnment resources; to say nothing of the likelihood that decentralimtion will increase, rather than decrease, aggregate public sector employment. 26. In the short-term, three major issues will need to be resolved if Zambia's decentralization program is to result in a positive contribution to achievement of the dual objectives of fiscal stabilimtion and improved efficiency and effectiveness of the public service. These three issues are: (i) excessive central government control; (ii) deficit financing; and (iii) inappropriate economic policies. Each of those issues are discussed in Chapter VI. C. Recommendations 27. The recommendations proposed, in combination, to address those three issue areas are divided into two categories: (i) Fiscal Stabilimtion and (ii) Organimtional Arrangements. The essence of the recommendations are that Government should clarify its intentions concerning: (i) the extent to which district councils, as public sector entities, are expected to adhere to Zambia's current fiscal stabilimtion objectives and principles underlying the Economic Reform Program; (ii) the primary functions that district councils should be expected to perform in which specific sub-sectors; and (iii) the manner in which central and provincial government will be organized so as to foster mutual support and complementarity among the three tiers of government. These are matters which Government as a whole needs to address. Such issues are at the core of Government's ability to improve its performance. Responsibility for addressing such policy issues should not be viewed as solely, or even primarily, the responsibility of the Decentralimtion Division of the Prime Minister's Office (PMO) . • §./ In spite of the provisions of the 1980 Act, the constitutional system in Zambia is that of ll Unitary, rather than Federal, state. Therefore, the SS district councils are creatures of the national government and any discretion exercised by them is based, legally, on voluntary grants of authority by Central Government. xiv VI. INTEGRATED POLICY AND ACTION-PROGRAM 28. As stated at the outset, the purpose of this report is to assist the Government of Zambia to formulate a medium-term strategy for rationalizing its performance of economic, financial, and personnel management functions in the context of its overall macroeconomic reform progra;.n. In that context, the Main ~ presents specific recommendations for action in the short, medium, and longer terms. Nevertheless, Government's past experience with attempts to reform its planning, financial management, and public service systems clearly demonstrate the need to integrate the various elements of Reform into a Program which it can reasonably expect to manage. As discussed in Chapter vn, priority should be placed on: (i) program integration; (ii) a structure for effectively managing the overall Public Sector Management Reform Program (PSMRP); (iii) new approaches to Technical Assistance (TA); and (iv) improved coordination of international donor financing by Government. The first two of those four priorities are further elaborated here. A. Pro~am Intmation 29. A careful reading of the Report demonstrates the mutual contingencies among planning, financial management, and personnel management functions. Thus, for example, if reform of public employment is to support Government's economic policy of growth with equity, it will need to be considered in terms of both fiscal stabilization on the one·hand and, on the other hand, improved operational performance of the public sector's necessary functions. At a minimum, it is clear that without effective expenditure control, no program of staff retrenchment in Zambia is possible. Further, to improve Government's performance of its operational responsibilities, streamlining the structure, procedures, and staffing of the public sector is necessary; even if staff retrenchment was not required to meet more limited fiscal stabilization requirements. That, in tum, requires an ability to monitor the sii.e and distribution of public employment in terms of Government's planned priorities over time. An important element of an effective personnel information system is an effective, computerii.ed, payroll system. Such a system can serve not only to monitor expenditure, but also to produce key information on the sii.e and distribution of personnel in an easily manipulable ma.: 1er. 30. Therefore, effectiveness of any Public Sector Management Reform Program (PSMRP) will need to integrate key elements among the planning, financial management, and personnel management functions. That, in tum, requires a clear articulation of Government's policy in that regard. Such a polky should focus on three interrelated objectives: (i) design and initiation of a comprehensive medium-term program for achieving the dual objectives of fiscal stabilization and improved performance in the context of the overall goal of economic growth with equity; (ii) establishment of specific limits on public expenditure based on clear cross-sectoral priorities established in terms of its Economic Recovery Program (ERP); and (iii) effective improvement of its performance in the high-priority sub-sectors established in the context of its overall strategy and financial constraints. 31. In particular, it must be stressed here that appropriate reforms of the public service are contingent on the broader restructuring of the public sector implicit in the requirements of both Government's ERP and its policy of decentralization. In that regard, the focus should be on the entire public sector at all levels, not just central government organizations. Given the demonstrated impact of local government economic, financial, and employment behavior on public sector performance, Government should give serious attention to resolving fundamental policy issues concerning the appropriate roles and responsibilities of central, provincial, and district level authorities and the relationships among them. 32. Under current economic and financial conditions in Zambia, three related elements provide the foundation for the effective accomplishment of PSMRP objectives: (i) a clear articulation of Government's policy, including clear specification of objectives and priorities for reform of the public sector; (ii) fiscal stabilization and redistribution of available financing so as to support high priority development objectives and functions; and (iii) control of expenditure so as to ensure that act>.1al disbursements reflect budgetary decision-making. That foundation xv is of particular importance to any sustainable progress toward public service reform during the initial three-year period. 33. Government's current capacity to articulate an integrated policy framework and substantially improve expenditure control is adequate to meet short-term requirements. Thus, it is feasible to meet such requirements in the short-term. However, with regard to the other key economic management functions, which have been described and assessed in detail elsewhere in this report. Government's current capacity is weak. Sustainable responses to those requirements necessitate a longer term institutional development approach; although interim assistance can be provided to meet other short-term requirements. B. Intmated Mana1ement of the Reform Effort 34. Because the objectives of short-term fiscal stabilization, medium-term economic growth, and the need to improve the efficiency and effectiveness of Government's role in providing an enabling environment are so inextricably linked, it is necessary for the Government to integrate responsibility for operational decisions regarding the various elements of its economic reform program. That is not now the situation. Rather, operational decisions concerning the short-term fiscal impact of Government performance is vested in the Ministry of Finance; responsibility for addressing longer term impact of performance on economic growth is vested primarily in NCDP; staffing and improved efficiency of the national level public service is vested in Cabinet Office; and staffing and improved efficienc:t of district councils is vested in the Decentrali7.ation Division in the Office of the Prime Minister. 35. Under such arrangements, it is natural that government officers focus on their own particular mandates, resulting in uncoordinated and, sometimes, contradictory actions. Thus, those responsible for improving the performance of public employees can be expected to focus their attention on improving incentives while those concerned with immediate fiscal stabili7.ation requirements are more likely to focus on the need to restrict such incentives because they increase the Government's financial deficit. 36. Resolving such potential contradictions while achieving necessary improvements in public sector performance requires both: (i) integrated management and (ii) mobili7.ation of participation and support from among a critical mass of stakeholders. Therefore, NEMIC should be assigned responsibility for overall coordination of the PSMRP. However, it will also be necessary for Government to appoint an official with sufficiently high rank (i.e., no less than a Senior Permanent Secretary) to manage the Program on NEMIC's behalf. Such an assignment should be long-term and full-time. VII. ASSUMPTIONS AND RISKS A. Assumptions 37. The most important and obvious assumption is that Government's officially articulated policy for reforming the public sector, although not well advertised within Zambia itself, represents substantial commitment. The history of previous reform attempts, however, does not augur well for effective implementation of any PSM reform effort. At the very beginning of the Main Report, it is argued that "[several] factors have influenced prior failures and, therefore, it would be seriously misleading to ~ that such failures have been due solely to lack of Government commitment and/or duplicity on the part of those responsible for implementing past reform efforts." Rather, it was argued, it would be more productive to assume that "the primary causes of failure have been: (i) absence of an appropriate and well-under.·tood policy context; (ii) lack of sh811)1y focussed priorities and systematic phasing of program implementation; and (iii) too much faith in technical remedies which have ignored political and xvi social realities.• Thus, the primary assumption underlying this Report is that the approach recommended here ad~iUltely takes into account the various reasons for those prior failures and provides an effective response to them. 38. More specifically, it is also assumed that Government: (i) will be able and willing to meet its obligations to the international donor group currently financing its Economic Recovery Program; (ii) was fully aware of the implications for the changing role of the public sector inherent in the ERP and, therefore, that its commitment to that program will be sustained over the medium and longer term without regard to the emergency nature of current external balance of payments financing; and (iii) is concerned with improving public service performance, and all that entails, beyond its immediate concern for the fiscal stabili1.ation objectives of such a program. Finally, it is also assumed that, if all of the above is correct, that any changes in Government which might occur over time, given Zambia's recent return to a multi-party electoral system, will not result in significant discontinuities in the implementation of an effective strategy once underway. B. Biw 39. The obvious risks are essentially the obverse of the assumptions identified above. Nevertheless, there are also other risks beyond the control of Government. Among such risks are: (i) lack of response on the part of the international donor community (especially those capable of providing grant aid); (ii) lack of cooperation by other Zambian stakeholders (such as the relevant labor unions) in spite of Government's best efforts to take a broadly participatory approach to design and implementation of specific elements of the program; and (iii) the strong possibility that implementation c,f a balanced approach to achieving the dual objectives of the program will be unbalanced because of external ,,ressures to meet potentially unrealistic fiscal stabilization targets at any cost. C. Response to Risks 40. To the extent that the primary risks are that Government and/or the international donor community might not be prepared to sustain their commitment to the full implications of effectively implementing the program recommended in these pages over the long-term, there is little that the program itself can do to counter that risk. Nevertheless, the approach underlying the recommended strategy and action-program is self-consciously directed at minimizing such risks. Thus, the approach taken in this Report emphasi7.es the need to: (i) tightly focus the reform program a limited number of achievable priority objectives; (ii) phase program implementation in a systematic manner, focusing initially on those objectives easiest to achieve, both technically and politically, and defer eventual initiation of the most difficult actions until accomplishment is practicable; (iii) develop broad-based understanding and support for the program through effective articulation, communication, and participation among both domestic and external stakeholders; and (iv) design a clearly articulated process for developing such understanding and support. ■ I. INTRODUCTION 1. At the invitation of the Minister of Finance and Planning, this World Bank Public Sector Management (PSM) Review bas been undertaken to assist the Government of Zambia to formulate a medium-term strategy for rationalizing the planning, budgeting, and non-parastatal public employment systems in the context of its Policy Framework Paper (PPP); including both short and medium-term action programs for the implementation of the strategy. 2. The work of the PSM Review Mission bas been undertaken with two operational perspectives in mind: (i) the need to improve the Government's capacity to perform its appropriate functions and (ii) the need to rationali7.e public expenditure to meet fiscal stabilization objectives without seriously damaging the first of these two objectives. A. Backw,und 3. These two policy objectives are not new in Zambia. Serious efforts were undertaken during the 1980's to rationali7.e and improve the performance of Government's planning and financial management systems, as well as the Public Service as a whole. 1/ Such efforts included attempts to implement most of the recommendations of a 1983 World Bank Report to establish a Planning, Programming, and Budgeting System (PPBS) 'J,./ md to rationali7.e the organir.ation of the non-parastatal public sector; including significant reduction of public employment levels. Nevertheless, such efforts have, to date, not been fully implemented nor have they succeeded in improving performance. 4. This report attributes such failures to the attempt to accomplish too much change in too short a time; priorities were not established nor actions appropriately p ~ . The result bas been substantial sub- optimiz.ation on all aspects of the program. S. The primary causes of past failures were: (i) lack of commitment to the adjustment program; (ii) absence of an appropriate and well-understood policy context; (iii) lack of sharply focussed priorities and systematic phasing of program implementation; (iv) absence of a strategy to transform the role and functions of the government in support of economic reforms and private sector development; and (v) too much faith in technical remedies which have ignored political and social realities. All five of these factors have influenced prior failures and, therefore, it would be seriously misleading to assume that such failures have been due solely to lack of Government commitment and/or duplicity on the part of those responsible for implementing past reform efforts. B.~ 6. With those lessons in mind, this report presents recommendations for: (i) a comprehensive policy with prioriti7.ed objectives; (ii) a program disaggregated into short, medium, and longer-term phases; and (iii) an implementation process which maximizes the possibilities of efficiently integrating political and social factors with technical requirements. 7. The report is presented in two volumes: (i) an Executive Summazy and Main Report (Volume I); and (ii) a Technical Annex (Volume II). Volume ll includes more detailed background information; descriptions of current organizational structures, procedures, legislation, and implementation policies; and statistical tables. !/ The tenn Public Service is used throughout this report to include all non-parastatal public sector employees, while use of the term Civil Service is limited to those employed at national and district levels under the terms of a Civil Service Statute which provides them with the status of permanent staff. 'Ji Planning and Budgeting in Zambia <Washington, D.C.: The World Bank, 1983). 2 8. This Main Report (Volume I) places primary emphasis on a limited number of immediate priorities for action, balancing fiscal requirements, desired improvements in public sector performance, and current implementation capacity. The objectives of public sector management reforms are discussed in Chapter II, as are the principles on which recommendations for the phased action-program are based. In order to provide sufficient background for an understanding of the very real fiscal constraints on programs to improve performance, Chapter m provides a summary of Zambia•s recent fiscal performance. Chapters IV through VI provide justifications for, and explanations of, the various recommendations integrated in Chapter VII. More specifically, Chapter IV describes public service efficiency and effectiveness objectives, 'J/ current performance, and an assessment of various mechanisms available for improved performance of necessary functions and retrenchment of staff responsible for unnecessary functions. Chapter V follows with a discussion of current planning, budgeting and financial management sy8tems and linkages among them, with an emphasis on the short-term priority concern for control of actual expenditures. Chapter VI discusses the impact of the role and performance characteristics of district governments on public sector efficiency and effectiveness. Finally, this Main Report provides recommendations for an integrated strategy and program of action for the improved performance of key economic management functions focussed on an initial five year period (1992-1996; Chapter VII). It should he noted, however, that because of the critical importance of improving the financial efficiency of Government performance in the !l!2a.: .tm:m, this Volume focusses primarily on such short-term imperatives. More detailed background information concerning planning, financial management, and personnel management systems are provided in Volume II. 'J./ As used in this Report, tt&e term effectiveness is defined in tenns of capacity to fonnulate, manage, and implement measures to achieve policy objectives; the tenn efficiency is defined in tenns of the expenditure of the least amount of human and financial resources sufficient to achieve those objectives. 3 Il. CURRENT PUBLIC SECTOR MANAGEMENT OBJECTIVES A. Medium-Term Policy Context 9. The overall objective of Zambia's economic recovery program is the promotion of economic growth with equity. A basic principle inherent in that program is that the role of the public sector in the economy will be transformed from the primary producer of goods and services to: (i) the financier (but not necessarily producer) of public goods and services and (ii) the regulator of appropriate private sector economic behavior. !/ Thus, it is expected that the public sector will focus its attention on providing an enabling environment for the private sector to play an expanded role in a medium term recovery of economic growth and investment. 10. If Government is to give practical effect to its dramatically new economic policy objective and redefined role in the economy for the achievement of that objective, it must take into account the limited resources available to it. That, in tum, requires fundamental decisions as to its investment and operational priorities; in terms of both sub-sectors and economic management functions. Priority Sub-Sectors 11. Priority sub-sectors have been identified in Government's preliminary consideration of a public expenditure program 2,/ and its established public investment programs (as identified in Attachment A to this Volume). These priority sub-sectors share the following characteristics: (i) clear public sector comparative advantage; (ii) high priority economic and/or social return; and (iii) focus on requirements for the creation of a sustainable enabling environment for private sector development and growth. Priority Functions 12. Priority functions include: (i) activities necessary to identify a public good or service (including assessment of its technical feasibility and economic viability); (ii) the specification of construction and/or operational requirements; (iii) an investment and/or recurrent financing plan; (iv) the generation of finance; (v) supervision of production of the good or service (which might be produced by the private sector under contract to the public sector); and (vi) evaluation of the financing and production of such goods and services for enhanced future planning and decision-making. 13. It should be noted that the list of functions which Government must necessarily perform to fulfill its primary functions do not include the actual production of public goods and services. Thus, for example, a decision by a public sector body, such as a local government council, to finance a feeder road and/or to maintain such a road requires it to: (i) weigh the benefit of such a road compared to other alternative goods or services that the Council might provide; (ii) assess its feasibility and viability; (iii) determine appropriate specifications; and (iv) mobilim finance for, and supervise, its construction and/or maintenance. However, it does not require the Council, through its own departments and employees, to actually carry-out such construction or maintenance itself. Rather, it can assign, through contracts, that production function to the private sector. !/ Although there are additional functions which must be performed for the appropriate, but limited, regulatory role of the Government, such functions have not been addressed by this PSM Review. Thus, the Government should consider additional assistance for the design of an appropriate institutional framework and the efficient and effective performance of such a regulatory framework. '2,/ The Bank is currently assisting Government to establish a public expenditure program, the result of which will be a Public Expenditure Review (PER) report scheduled for completion by May, 1992. 4 14. The principles underlying the changing role of the public sector provide the criteria for the specific recommendations provided in this report. Thus, priority should be given to improving the performance of those functions identified above (para. 13) in those priority sub-sectors also identified above (para. 12). Actions taken to streamline the public service and meet fiscal stabilization objectives through retrenchment should, during the initial three years of the program recommended in these pages, focus on those functions and those sub-sectors not included in the above priorities. From that perspective, emphasis in this report is placed on: (i) mechanisms which the Government could employ for implementing retrenchment of personnel who are nonessential by that criteria (Chapter IV) and (ii) improving the performance of the essential cross-cuttin~ financial functions of planning, budgeting, and financial management (Chapter V). Recommendations for improving performance of unique sector- specific technical functions, including management and sui>ervision, within each high-priority sector is not within the scoix~ of this Review. B. Short-Term Fiscal Stabiliytion Objectives 15. The major fiscal objective of the Government is to substantially reduce inflation by curtailing recourse to bank financing and generating the resources necessary for a recovery in growth and investment. For the next three years, key policy objectives are: (i) increase the revenue/GDP ratio with particular emphasis on nonmetal revenue sources; (ii) reduce the real level of non-interest expenditure and restructure public exp!nditure in favor of productive and high priority activities; and (iii) strengthen monitoring and control of the financial operations of the central government. Measures to reduce public sector expenditure include a cap on personal emoluments during 1991, except for an allowance for increases due to promotions not to exceed eight percent of the aggregate wage bill (Chapter IV). As shown in Table 11.1, the primary balance, excluding grants, is scheduled to decline substantially from a deficit of 4.S percent of GDP in 1990 to 0.S percent in 1991, before moving into a small surplus of about 0.4 percent of GDP in both 1992 and 1993. Thus, it is expected that recourse to domestic bank borrowing will be minimum in 1991, and that substantial repayment to the banking system will take place in 1992 and 1993. Revenue System Reform 16. In response to negative revenue generation trends outlined in Chapter m below, the Government intends to introduce structural reforms of the tax system to: (i) broaden the tax base; (ii) improve tax compliance; and (iii) strengthen the revenue function. Broadening the tax base, however, is the first priority. To broaden the tax base, the Government intends to implement the following specific revenue measures over the next three years (1991-93): (a) Broaden the base of personal income tax by taxing fringe benefits (including housing benefits); (b) Broaden the base of sales tax including all imports (rather than only dutiable imports, as has been the case), all locally manufactured consumer goods (including excisable goods), and several other services; (c) Broaden the base of import taxes and excise duties by eliminating virtually all exemptions and reliefs; (d) Broaden the base of company tax by limiting the claims of the parastatals for exchange losses to those realized during the tax year; and (e) Create a Revenue Board staffed by special cadres with clearly defined responsibility and accountability, with compensation directly linked to performance. 5 w·•11ww-~:,~· · . ..... . .... . . ,,.,•,•:r~tq~r:·:;~v·~;-•·w,• . . . . ' ;~½~•··;,1;;;::;;:~;ii0;t~;::·:;;'. I ~ I l :.!_i,: Expenditure Restructurina 17. The consequence of severe resource constraints in recent years bas been a serious deterioration in the provision of economic as well as social services. Government's immediate response to fiscal constraints include attempts to: (i) reduce the budgetary burden of subsidies; (ii) target investment and sufficient recurrent expenditure on clearly established high priority sub-sectors; and (iii) strengthen expenditure control and budget execution. Thus, sufficient expenditure will be provided for the key sectors of agriculture (particularly extension and other support services), roads (for maintenance and rehabilitation), and health and education (for supplies and maintenance equipment). Measures for expenditure monitoring and control, include: (i) strict enforcement of relevant procedures; (ii) requiring ministries to maintain commitment ledgers on a timely basis; (iii) limiting 6 transfers to district councils; and (iv) undertaking a thorough review of the levels of expenditure allocated to grant- aided institutions; particularly with regard to payroll and fringe benefits. lmpleroentation issues in this regard are discussed in more detail in Chapter V. Public Investment Prom:a,m 18. In response to the fiscal constraint, and the need to mobilize donor development financing within a common budgetary framework, the Government bas prepared priority Public Investment Programs (PIP) on a revolving basis since 1989. The objective of the PIP is to specify priority investment projects and programs consistent with available domestic and external financing. Inter-ministerial consultations during the course of preparing the PIP are expected to facilitate stronger integration of all planning and budgeting decisions; whether financed from domestic resources or by donors. 19. The use of PIPs as a guide for public sector investment decisions represents a significant step toward integrating development planning into the annual budgetary process. However, to achieve maximum benefit from PIP planning, such plans need to be consistent with a policy framework which clearly articwates priority sub- sectors. In addition, although effective implementation of such decisions through budgetary allocations and actual expenditure is an obvious requirement, such behavior has been the exception rather than the rule. C. Public Sector Performance Qbiectives 20. The si.ze and distribution of non-parastatal public sector employment is important from the perspective of both fiscal efficiency and performance effectiveness. Even if the Government was not confronting a fiscal crisis, excessive public sector employment would remain as a significant problem. Supervision of excess employees, especially those at unskilled levels, places a significant and unnecessary burden on middle level managers, while the observable inactivity of such redundant employees creates morale problems for those staff with actual tasks to perform. 21. Nevertheless, although Government bas long recogniz.ed the need for reform of the Public Service, there is no clear articulation of policy objectives in that regard; events have clearly overtaken past policies and Government employees, including those responsible for impJementing any reform program, are unsure of what is Government's current policy. In addition, there is a clear tension within Government among those who address public service reform issues from the perspective of improving performance effectiveness and those who approach such issues from the perspective of fiscal stabilization requirements. The former tend to view retrenchment of public sector employment, if driven primarily by short-term fiscal stabilization targets, as threatening the ability of the public service to perform even essential services and regard those advocating such programs as unsophisticated about the difficulties of implementing such programs; to say nothing about the need to preserve necessary capacity to perform appropriate Government functions. The latter tend to view those concerned with the very real and complex issues of improving public sector performance capacity as obstructionists who are not able to transcend what appear to them as parochial interests in the face of financial crisis. 22. The current tension outlined above is particularly unfortunate because it is more apparent than real. Achievement of both fiscal stabiliz.ation and improved performance effectiveness objectives is equally necessary and possible; achievement of neither objective is desirable without th.- other. However, it will be necessary to clearly articulate the importance of both objectives in an explicit statement of Government policy and a program to implement that policy will require careful crafting. 23. The approach recommended in this paper to achieving both fiscal stabilization and improved public sector performance will be detailed in Chapter IV. Suffice it to state here that the recommended retret1cbment process is characterized by movement from the use of relatively crude instruments for decisions regarding staff I 7 performing low priority functions in non-priority sub-sectors to the application of carefully refined instruments for such decisions regarding staff performing high priority functions or assigned to high priority sub-sectors. D. Phased Actlon-Promm; Prireiples and Approach 24. The principles on which recommendations for specific actions an1 sequencing, as presented in Chapter VII, are based include the following: (a) Under current conditions in Zambia, even minimally adequate public sector performance is not possible without a stable fiscal environment and reallocation of wage and non-wage recurrent expenditure; (b) Requirements tc..c efficient and effective performance of Government's redefined functions should be the sole determinant of the size, distribution, and skill mix of Zambia's public service; (c) Given current Government capacity, implementation of any reform program needs to be tightly focussed on a limited number of achievable objectives, the priority of which are clearly understood by those required to implement them; (d) Implementation targets should focus initially on those easiest to achieve, both technically and politically, and eventual initiation of the most difficult actions should be undertaken only as it is practicable to accomplish them; (e) Those stakeholders capable of significantly affecting implementation of the reform program, either positively or negatively, should be directly involved in planning, decision-making, and/or negotiations about the various actions included in the program, as appropriate; (f) An explicit process for facilitating the involvement of important stakeholders is necessary for the effective implementation of the reform program; and (g) Notwithstanding the need for broader-based participation of stakeholders in the formulation and implementation of the reform program than has hitherto been the case in Zambia, decisions will still need to be based on considerations of teclbtical and financial feasibility and economic viability. 25. In order to integrate both fiscal stabilimtion and public sector performance requirements, the approach recommended here posits that fiscal stabilimtion objectives will •drive• reform of the public sector in the non-priority sub-sectors during an initial three-year period, while improved efficiency and effectiveness in high priority sub-sectors through capacity-building efforts will "drive" the reform program in later years (although necessary studies and proposals for action in those priority sub-sectors should he initiated in parallel from FY1991). That approach avoids delay in achieving necessary fiscal objectives by focussing on a small number of priority short- term actions during an initial three-year period; including improvement efforts regarding expenditure controls and reduction of employment levels in non-priority sub-sectors commencing with the current 1991 budget year. Thus, the long drawn-out, but necessary, process required to improve the performance of broader public sector planning, revenue generation, and financial management functions, plus determination of the impact on performance of a staff retrenchment program in high priority sub-sectors, need not delay the achievement of immediate fiscal objectives (see Chapters IV and V). 1/ 'J./ In the context described herein, the consulting study currently being prepared by VisionConsult (a subsidiary of ZCCM) for the Cabinet Office should be viewed as an effort to establish the necessary data base and analyses required for that lo11ger term effort. 8 m. RECENT FISCAL PERFORMANCE 26. As indicated above, fiscal stabilization is a necessary condition for any broader improvement of public service performance. This Chapter assesses Government's capacity to finance the wide range of functions it has previously assigned itself, in terms of both the total amount of resources available to it and the allocations of those limited resources. A. Public Expenditure (197S - 1990); Background 27. For many years copper revenue allowed the GRZ to maintain an unusually high level of central government expenditure in l'll attempt to perform a wide range of functions. For example, during 1975-80, central government expenditure averaged about 40 percent of GDP per year. This compares with 27 percent for middle- income African countries as a whole, 26 percent for Tanzania, 22 percent for Malawi, and 20 percent for Kenya during the same period. !/ The range of functions and types of services Government attempted to provide reflected both its relatively wealthy status and its official philosophy. With such resources available, the public sector dominated the economy through its parastatals and various policy interventions. Copper revenue also enabled Government to extend social services to all citiz.ens and act as a primary provider of employment opportunities. 28. Since the mid-1970's, copper revenue has declined dramatically. In addition, e--..onomic growth has not been sustained. Indeed, per capita income declined from $729 in 1981 to $390 in 1989. As a result, discretionary public expenditure (OPE) 'Z/ has declined from 27 percent of GDP in 198S to 20 percent in 1990 (Table ill.I). The siz.e of Government's discretionary budget in real terms declined by about 25-30 percent from 198S tc the period of 1989-90. To some extent, the reduction of Government expenditure is also the result of an eroding tax revenue base and lower tax compliance. 29. Despite declining government expenditures in both relative and real terms, public sector participation in the economy and its role and functions continued to expand until very recently. Thus, Government responded to the economic crisis by further tightening its grip on th~ economy through price controls, administrative allocations of foreign exchange (except during the period 1985-86), imposition of additional import restrictions and export bans, direct subsidies of agricultural production and marketing, and indirect subsidies of parastatals. Such expanded intervention in the economy was increased further following the abandonment of ,.m earlier adjustment program in 1987. In addition, rApid growth of the overall rAJpulation has resulted in proportionally increased demands for employment, public health care, . and general education services. Such population !/ Planning and Budgeting in Zambia (Washington, D.C.: The World Bank, 1983). 2/ DPE = expenditure not including service payments for public debt. 9 growth has been compounded by the demographic distribution in Zambia; which is highly urbanized. Vocal demands by the urban population for continuing the previously high levels of Government services at no cost to themselves has severely constrained the level of resources available to the naral poor. :iO. It is now readily apparent that, given dwindling fiscal resources and a shrinking expenditure program, Government is no longer capable of performing all of the functions it bad assigned itself prior to the mid- 1970's when the current economic crisis began. The sharp decline in the level of real government spending should have induced a recomposition of expenditure in order to protect priority areas of Government operations. This could not have been achieved unless Government took measures to considerably reduce expenditures in non-essential and low priority areas so as to ensure adequate funds for provision of high priority goods and services. However, for the most part, such restructuring of public expenditure bas not yet taken place. Therefore, much has yet to be accomplished in terms of Government redefining the scope of its operations, its role and functions in the economy, and the restructuring of its expenditure programs. The current period constitutes the earliest phase of a potentially profound restructuring of the role of the public sector in the Zambian economy. B. Composition of Expenditure 31. That the pl :>jected restructuring of the public sector has not yet occurred is evidenced by the composition of expenditure. The most prominent feature is that there has been a growing diversion of Government's resources away from sur""rt for recurrent operations (i.e., RDCs) and capital expenditure, towards more for public employment, subsidies, an~ .:nilitary expenditure (see Figures m.l[A] and m.l[B] and Tables IIl.2 and 111.3). Public EmplOJ'ment 32. The budgetary definition of "Personnel Emoluments" is seriously misleading and clearly illustrates the lack of transparency in the Budget format. Thus, a review of central government budgets for the period 1985 to 1990 shows that allocations for Personal Emoluments, narrowly defined in terms only of that specific Budget Heading, declined sharply from 2S percent of total expenditure to 12 percent. In real terms, this expenditure item was reduced by 58 percent from 1985 to 1990. 'J/ Those statistics, however, mask the fact that costs of salaries, wages, and all other fringe benefits bas actually increased as a share of total government expenditure. That reality is not observable in the Budget because much of the increase in the total wage bill ~/ is financed through two other items in the Budget: (i) Grants and Other Payments and (ii) Recurrent De,partmental Charges. Thus, for example, the actual wage bill for 1990, including expenditures from these two other Budget items, is estimated to be k6,980 million; fully k2,498 million (55%) more than thst attributed solely to the Personnel Emoluments line item. By that measure, personnel expenditure accounted for 18 percent of total expenditure in 1990. This problem 'J.I Although this calculation does not take into account the recent increases of salaries and housing allowances in mid- 1990 (see Chapter IV), such increases have only reduced the extent of the real decline, not reversed it. That picture remains essentially correct even though, as argued above, the specific allocation for Personal Emoluments presented in the Budget does not, in fact, capture the entire amount of expenditure disbursed for salaries, wages, and benefits for all persons employed in the non-parastatal public service. Thus, although limited primarily to those persons employed by the central government and enrolled on the computerized payroll system (see Chapter IV), the category of persons to which these statistics are applicable remain the same. Therefore, although aggregate expenditure on salaries, wages, and benefits has probably increased, not decreased, substantially as a share of total expenditure, the amount allocated to central government employees on the computerized payroll has substantially decreased. The significance of this is that the population of public employees most easily monitored and for whom expenditure can most easily be controlled have been the subject of such decreases, while employees not enrolled on the computerized payroll system are estimated to have increased substantially and, thus, personnel expenditures for that category of personnel has escalated (see Chapter IV). ~/ The Wage Bill is defined here as the total costs of wage, salary and all other fringe benefits (see Chapter VI). 10 figure m. uA> 11 is discussed further in Chapter IV (in the context of public service reform) and in Chapter V (in the context of Budget presentation). Non-Development Expenditure 33. Despite the Government's professed policy goals of reversi· •g the prolonged economic decline, the pattern of non-ministerial expenditure constitutes a clear trend during the period 1985 - 90, moving away from general development purposes to spending on mibsidies, national defense, and less development related purposes. 12 Increased expenditures for subsidies and the military are particularly noteworthy for that period: (i) in 1985 prices, subsidy costs rose from k18f million in 1985 to k287 million in 1990 while (ii) military expenditure increased from k287 million to k338 million during the same period. That occurred within the context of an overall aggregate decline, in real terms, of 22 percent for ministerial expenditure and a decline of only 7 percent for non-ministerial expenditure. ~/ This data is significant because, on thn whole, non-ministerial expenditures are less directly related to economic development purposes than are ministerial expenditures. 34. The result of these expenditure patterns is that in a large number of ministries, shortage of operating funds has left government employees with insufficient t.upplies and materials to carry out their assigned tasks. Thus, the lack of operating funds, as well as the substantial erosion of civil servant's salaries, have seriously undermined the morale and quality of the civil service sector and, thus, constrained the ability of Government to perform its essential functions and to deliver key goods and services (See Chapter IV). Development Expenditure 35. Tables Ill.4 - Ill.5 show the distribution of expenditure by economic ministries. The first notable feature of such a sectoral distribution is that expenditure on Economic sectors ~ . agriculture, mining and manufacturing) has increased, albeit moderately, since 1985, both in real terms and as a percentage of total government expenditure. The bulk of that increase bas occurred in agriculture, increasing that sector's share of total expenditure from 4.7 percent in 1985 to above 7 percent in 1990. This reflects the high priority Government has assigned to the development of agriculture as an alternative to the declining mining industry. 36. Another notable feature is that real expenditures for infrastructure and social service sectors have declined significantly; the latter experienced the most severe reduction, dropping from 24 percent in 1985 to 17 percent in 1990. In small part, this was the result of fiscal adjustment measures introduced by Government. For example, some of that reduction has been offset by cost recovery measures introduced in higher education institutions and for certain types of health services. But, most importantly, such reductions were the consequence of a prolonged economic decline which took its toll relatively more on the social sectors (which often became the first victims of spending cuts). It is reasonable to assume that disproportionate reductions in non-wage recurrent budgets generally, and the health and education sectors more specifically, have occurred because such actions are viewed by decision-makers as politically less threatening than reduced public sector employment, military spending, maiu subsidies, and financing of the United Independence Party (UNIP). C. Financln& of Public Expenditure 1986-1989 37. Revenue (including grants) financed approxillllltely 58 percent of total expenditure during 1986-89 and 72 percent in 1990; borrowing paid for the remaining 42 percent during 1986-89 and 28 percent in 1990. Put another way, the Government spent approximately 72 percent more than it received during 1986-89 and 39 percent more in 1990. Compared with the 1970's, when government revenue financed about 66 perc,,nt of government expenditure, this reduced level of revenue financing reflects: (i) a loss of mineral revenue in 1986; (ii) a decline in donor grants as the Adjustment Program was suspended during 1987 and 1988; and (iii) a general lack of tax compliance and erosion of the tax base. 5_/ Thus, as a share of total spending, ministerial expenditures were only 54 percent in 1990 as compared with SB percent in 1985. Non-ministerial expenditure grew from 42 percent of the total in 1985 to 46 percent in 1990. 13 "ffl" ) " : , - ~ - ~ ~ ~ ~ ~ l (: ( j~ : I . 1 . I » f 3 j J ? ~~~~~ ..~~~. . . . ~~~~~..i';!i!~~~~~~ fflffl~~!i'-'lj~~~~~....~~~~ ""-"'-'~~~~ ij~~!!!!ii!i'-'lj ❖- '@tWII' ~ !'~!!J!l~t:!r!l~li17J!lf!f ~ ;\,lf ,~;:=::=-;'.;:·;:;;. £; , i',;~- ~·-;1, \, :, -=·2-;·:~ :·-n·;3.·: · .· ❖-,~-:. ;.,x·:_•'. ·:•\':_· •❖·< · L- '_•·.~- ;❖·:.~t·; ~'., : F,;-•--.:.:-··_·.A•·-1.·:\:,··;\:-f }5&-'::r. -· ;•:·;, . :;~;:; ;\~: ~ :&tfug;;fsfa-::f:~tt-<}2.}t''.L" •-:S:,·,·:.;_.iWk:A~Jt 0 .. ·;··.__ u·_:.',_\/._ .. ,·:··:;.r,!._,':,·•·-··;_;_!,·_ 38. With reference to the lack of tax compliance, several studies §./ have noted that Tax Inspectors are poorly motivated, are provided with inadequate operating budgets and supplies, and operate with insufficient latitude for decision-making. Nominal measures adopted by Government to raise corporate and personal income tax rates have further eroded tax compliance; increased rates have led to increased evasion. Thus, a vicious circle bas been created: (a) Over-extension of the public sector bas resulted in the inability to finance the full range of Government's recurrent operations; §/ Two recent examples are: (i) Zambia: Rehabilitating Tax Administration, Fiscal Affairs Department (Washington, D.C.: The International Monetary Fund, 1991) and (ii) Glenn P. Jenkins, James Owens, and Ian Roxan, Zambia: Income Tax Reform - Administration and Policy. mimeo, 1991. 14 (b) Lack of prioritization bas resulted in inadequate financing of all operations and, as a part of that broader problem, inadequate financing of the revenue generation function; (c) Inadequate financing of the revenue generation function bas resulted in the absence of personnel qualified to address tax policy issues ~ . broadening of the tax base); (d) That, in tum, has resulted in the simplified response of increasing tax rates for those historically responsible for paying taxes; and (e) Those bearing increased tax rates resort to widespread evasion which cannot be effectively addressed because of inadequate financing of the Government agencies responsible for enforcing compliance with tax regulations. The consequence of that vicious circle is that Government has bad to finance its expenditure program through heavy foreign and domestic borrowing in order to maintain its comprehensive role in the economy (see Table 111.6). 39. Thus, in recent years, Government borrowing has been about three times as much as capital expenditure (compare Tables 111.6 an<l 111.3). It is generally accepted that borrowing can cover a portion of capital 15 expenditure, since investment assets can be expected to generate a future stream of income to repay loans. However, the ratio of borrowing to capital expenditure during this period clearly indicates that a major part of borrowing (85 percent during 1986-89) must have been used to cover recurrent expenses which do not directly create assets or generate future income; even if they do maintain existing assets. 40. Such levels of borrowing have had two important effects. First, it increased debt service expense and, thus, reduced the amount of funds available in future years for other uses. Second, to the extent that the Government borrowed from domestic sources, it reduced the amount of credit available to the rest of the economy (to the extent that monetary expansion was limited) and/or increased inflation through the printing of money to meet credit needs. More specifically, Government borrowing from domestic sources absorbed S2 percent of the net domestic credit created during 1986-89. This indicates that, indeed, proportionally very little credit was left for the rest of the economy. That, in turn, has had a negative effect on the production and profits of, and tax revenues from, parastatals and the private sector, which depend on sufficient credit being available to them. Those effects appear to have been particularly strong because the bulk of the public sector deficit is composed of spending on subsidies, public employment, and the military. 1990 - Present 41. Revenue performance during 1990 showed considerable improvement over the 1986-89 period. This was due to at least three factors: (i) with the adoption of a Policy Framework Paper in September, 1989, the donors accelerated their assistance to Zambia; (ii) mineral revenue improved significantly (not only because of the rebound of copper prices, but also because of the specific mechanism adopted in 1990 to collect mineral tax); and (iii) with the introduction of a Second Window exchange rate for foreign allocation and valuation of import duty, revenue from import duties also improved. 42. Thus, as emphasi.zed in Chapter II, Government's current economic adjustment program necessarily requires: (i) adoption of revenue increasing tax measures supported by: (ii) measures to contain expenditure to achieve deficit reduction and fiscal stabili7.ation; (iii) application of the principle that Government finance, but not necessarily produce, only the highest priority public goods and services; and (iv) identification of 16 high priority sub-sectors and functions for that purpose. Those various elements can, thereby, provide criteria for restructuring public expenditure, selection of public investments, and targeting and phasing retrenchment of public employment. 43. Because of the excessive level of Government expenditure devoted to non-parastatal public employment on the one band and the central role that the Public Service must play if efficiency and effectiveness of Government is to be improved, this Report turns next to the subject of Public Service Reform (Chapter IV). 17 IV. PUBLIC SERVICE REFORM 44. This Chapter addresses requirements for significant reform of the Public Sector over the short through long-term. This is an extremely complex issue with a long and well knoY1n, but unsuccessful, history in Zambia. Therefore, the discussion here is limited to the objectives and approach to reform, key issues to be addressed through the reform program, and specific recommendations for action. A. Objectives 45. As a core element of Government's effort to improve the performance of the public sector, the objectives of a Public Service Reform Program (PSRP) should correspond to that more comprehensive program. The ultimate objective of the PSRP must be improved management performance of public sector investment and recurrent operations within the context of the changing role of Government in the economy (see Chapter II). That, in turn, requires the establishment of an appropriate organizational structure and staffing pattern. The criteria for thinking about what would be the appropriate staffing pattern for the public service is the type and number of tasks which need to be performed to achieve Government's priority objectives. However, another important criteria is the appropriate balance between wage and non-wage recurrent expenditure within the aggregate limits which Government can afford. As bas been clearly demonstrated in Chapter m, capital expenditure bas not suffered the crippling degree of reductions visited on non-wage recurrent expenditure. 1/ Therefore, an important objective of the PSRP in Zambia should be to reduce the proportion of wage expenditure relative to non-wage recurrent expenditure. That is necessary because the current ratio among those expenditures is clearly inimicable to maximizing the output of public goods and services. Figure IV .1 illustrates in graphic form the impact of distortionate wage expenditures on labor productivity. 1/ 46. From a practical, shorter term, operational perspective, however, the immediate objectives of the PSRP are: (a) Design and initiate a medium-term capacity-building program for the improvement of public sector performance of key functions in high priority sub-sectors; l/ The discussion in Chapter m has clearly outlined the sharp decline of RDC spending in both real tenns and as a share of total Government expenditure. That is in sharp contrast to the growing size of government employment, the expanding share of wage expenditure, and the depletion of capital stock (due largely to inadequate resources for operations and maintenance [O&M]). The clear implication is that while an increasing share of the recurrent . budget is being diverted to finance the government payroll, public sector employees are being under-employed for lack of non-wage recurrent budget to carry out their assigned tasks. 2/ The purpose of Figure IV .1 is to illustrate the relationship between the amount of goods and services which can be produced given a particular ratio between wage and non-wage recurrent expenditure (this is a modification of the well known "Laffer Curve"). If wage expenditure is an inftnitely small percentage of total recurrent expenditure (TRB), then the amount of Government goods and services produced by a given number of public employees with a given budget (including both capital and recurrent) will be infinitely small; i.e., close to zero. On the other hand, if wage expenditure accounts for 100 percent of the recurrent budget <.i&.,, no non-wage operating budget), then the amount of Government goods and services will be zero. Thus, the maximum output level can be reached if wage expenditure is at the point shown in the Graph as "A." Obviously, the critical judgement is whether Zambia's current wage bill expenditures have passed the point of maximum output. Based on the evidence presented in Chapter Ill, the current situation is clearly to the right, downward sloping, side of point "A" (as represented by the range "xy"); even though at the present time available empirical data does not allow for calculating its precise location because some part of the wage bill is financed out of RDCs and Grants and Other Payments (see Chapter ID). 18 (b) Identify specific limits on public expenditure for non-parastatal public sector employment to be effected through reduction of the wage bill, in part through retrenchment of staff in non-priority sub- sectors; and (c) Determine the formula and procedures for redistributing savings from non-priority sub- sect ors to staff performing key functions in high priority sub-sectors. 47. Analyses presented in Chapter III clearly shows that under current conditions in Zambia, even minimally adequate public sector performance is not possible without a substantial reduction of the fiscal deficit. This is so because in the absence of such stabiliution, inflationary growth of wage expenditures tends to displac" other recurrent expenditure. There can be no substantial doubt that reduction of the size and fiscal burden of public sector employment is absolutely necessary if sustained economic growth is to be achieved over the medium and longer term. On the other hand, it has also been argued in this Report that requirements for efficient and effective performance of Government's redefined functions should be the sole determinant of the size, distribution, and skill mix of Zambia's Public Service. Thus, an essential element in resolving the potential tension between short-term fiscal stabilization requirements and the need to improve the efficiency and effectiveness of public sector performance of necessary functions is distinguishing between high and non-priority sub-sectors; as discussed in Chapter II. B. Approach 48. As argued in Chapter II (para. 22), achievement of both fiscal stabiliution and improved performance effectiveness objectives is equally necessary. This is particularly the case of public service reform. Therefore, the strategy recommended in this paper is directed at establishing an integrated approach which allows for meeting staff retrenchment targets while protecting the core functions of Government in the short-term and providing sufficient time for the design of programs to improve public sector effectiveness in the medium-term. In effect, a protective boundary would be placed around high priority sub-sectors during the earliest phases of staff 1 retrenchment and government reorganization. Improved efficier,, effectiveness through capacity-building efforts could then drive the reform program in later years (althou~ ;sary studies and proposals for action in those priority sub-sectors should be initiated in parallel from FYh!7 J.J, That approach avoids delay in achieving the fiscal objectives by permitting significant reduction of employment levels in non-priority sub-sectors starting with the current fiscal year. Thus, the long drawn-out, but necessary, process to determine the impact on performance 19 of a staff retrenchment program in the high priority sub-sectors need not delay the achievement of immediate fiscal objectives. 'J./ 49. The sequence of the retrenchment program recommended in this Chapter focusses initially on those targets which are easiest to achieve, both technically and politically. ~/ Thus, as illustrated in Figure IV .2, the sequencing of the recommended program is essentially as follows: (a) Identify and eliminate "Ghosts" on the payroll at both central and district government levels; (b) End obligatory Government employment of all those trained at Government post-secondary training institutions; (c) Identify and begin to dismiss obviously redundant staff (such as the substantial excess of drivers over available Government vehicles); (d) Identify those eligible for normal retirement or subject to expiration of contract who are performing non-priority functions and/or whose career streams are non-priority and enforce such retirement/refuse to renew contracts; (e) Design and offer a financially viable voluntary early retirement package for persons with more than 4 years to normal retirement who perform non-priority functions in any sector; (f) Design and offer a financially viable voluntary early retirement package for persons with more than 4 years to normal retirement assigned to non-priority sub-sectors; (g) Dismiss individual staff for malfeasance on a case-by-case basis; and (h) Dismiss other staff non-voluntarily only after all the above mechanisms are used and as fiscal requirements and/or effective performance at the organizational level demand them. That process is characterized by movemt>nt from the use of relatively crude instruments for decision-making to the application of carefully refined instruments for such decisions. c.~ SO. Based on analysis of past experience and current conditions in Zambia, five fundamental issues emerge as requiring significant attention: (a) Absence of an explicit link between the structure of public sector staffing and macroeconomic policy priorities; 'J/ In the context descnbed herein, the consulting study currently being prepared by VisionConsult (a subsidiary of ZCCM) for Cabinet Office should be viewed as an effort to establish the necessary data base and analyses required for that longer tenn effort. ~/ See Baibara Nunberg and John Nellis, Civil Service Refonn and the World Bank, Working Paper 422 (Washington, D.C.: The World Bank, May 1990) and Louis de Merode, Civil Service Pay & Employment Reform in Africa: Selected Implementation Experiences, AFI1M Division Study Paper No. 2 (Washington, D.C.: The World 8ank, June 1991). 20 FIGURE IV.2 PUBLIC SECTOR RETRENCHMENT: SEQUENCING FLOW CHART Suhffl:toa High Priontx Non-prjontx ... . (i) Normal . - C c:.. (ii) retirement/ expiration or c.ontra..1; Early (iii) re:irment; Dismissal . (i) Norma! (i) . Normal retirement/ retinnent/ expintioa expir1tion cf C:ODtract; orc.ontnct; ... - "" 0 (ii) E.uly · retirmeac; 2 (ii) E.rty . retirement; ...I,. (iii) Dismissal. (iii) Dismis~al. I C Q :z: ·Ghcsts;· I (i) - CZI Normal - ;, (ii) >w ~et:i:-eoe:ic/ =e = C ;, u C I expiration .J -= C": - :: -- :ICQ Q - : >,, :, _, _, C c:. :> "C E.t:: «I C er: :, I (iii) of c.ontract: Early - - Retire.-nent: - - - (iv) Dismissal. - I ·t Improved Effic:i~nc:y and Effec:tivness t Reduced Negative Impact of Public: Service Performance of Wages and Benefits on Public Servic:e Fiscll Stabilization Targets Ptirnaa Objectives 21 (b) An imbalanced structure of public sector staffing given the implicit requirements of current macroeconomic policy priorities; (c) Inadequate incentives for performance, including reliance on an inappropriate structure of personal emoluments; (d) Weak personnel and financial management; and (e) Absence of sophisticated understanding of the crisis in public sector employment among key stakeholders and mobilu.ation of necessary support, through negotiations, for a program to implement acceptable solutions. Each of these issues are discussed below. Policy Priorities S1. government Function,. As suggested in Chapter 11 of this report, a basic principle inherent in Zambia's economic reform program is that the role of the public sector in the economy will be transformed from the primary producer of goods and services to: (i) the financier (but not necessarily producer) of public goods and services and (ii) the regulator of appropriate private sector economic behavior. Nevertheless, although that principle is inherent in Government's current economic reform program, it is not explicit in any Government document accessible to most of its own staff or the public. Thus, clear signals have not be given that the priority functions which Government expects to perform in future are largely limited to: (i) activities necessary for decisions to provide a public good or service (including assessment of its technical feasibility and economic viability); (ii) the specification of construction and/or operational requirements; (iii) an investment and/or recurrent financing plan; (iv) the generation of finance; (v) the contracting for and supervision of production of the good or service; and (vi} evaluation of the provision and production of such goods and services for enhanced future planning and decision- making. Thus, the functions which Government expects to continue to perform into the indefinite future~ include the actual production of public goods and services. 52. Behavior vs, Policy. Government has delineated a rather complete policy framework relating Public Service reform to its broader economic reform objectives.~/ However, its own behavior bas not been consistent with that policy framework; in part because it has not been communicated to staff, and, thus, has not been well understooc: throughout the Public Service and, in part, because it is not broadly supported (for approaches to ameliorate that problem, see paras. 98, 125 - 126 below). For example, in spite of Government's well established commitment to contain the wage bill, an adjustment of the wage bill in July 1990 included across the board 85 percent increases in salaries and wages and an increase of housing allowances by over 100 percent. The consequence of that adjustment is that prices accelerated during the second half of 1990, raising the consumer price index to over 100 percent for the year as a whole. Stamna 53. Due to decades of economic stagnation, and deliberate policies to limit private sector development, the public sector in Zambia has become the main provider of formal employment. As illustrated by the data provided in Table IV.1, public service employment in 197S accounted for only 22 percent of formal employment while in 1990 it had increased to more than SO percent. Even if the Government was not confronting a fiscal crisis, excessive public sector employment remains a significant problem. Supervision of excess employees, especially ~/ See "Letter of Development Policy,• Government of Zambia, dated February 1991. 22 those at unskilled levels, places a significant and unnecessary burden on middle level managers while the observable inactivity of such redundant employees creates morale problems for those staff with actual tasks to perform. 54. Types of Public Service Employ- mmi. There are four major types of public service employees: (i) the regular Civil Service; (ii) non-civil service employees, mostly Classified Daily Employees (CDEs); (iii) Local Administration Service (LAS) employees; and (iv) National Joint Council Employees (NJCE), the equivalent of CDEs at local government levels. Suffice it to note here that there has developed an obligation for Government to retain CDEs and NJCEs indefinitely once they have been employed for six months. Thus, large numbers of theoretically temporary employees are now considered by Government to be permanent wclassified employees, w rather than wdaily contract employees. w As a result, the terms and conditions of their employment have become similar to those of regular civil servants. 55. Size and Distribution. Govern- ment's payroll system provides the best available source of information on the sue and composition of non-parastatal public service employment. Public service employees (excluding parastatals) are paid through two payroll systems: (i) a computerized payroll and (ii) manually maintained payment lists. S6. Non-parastatal public sector employment is characterized by substantial over-staffing at lower levels and inadequate staffing at middle and higher levels. Over-staffing at lower levels is largely a function of extensive employment of CDEs and NJCEs. The former currently exi:-,eed 73,500 (see Table IV.2); acr..ounting for more than 41 percent of total central government employment and representing an increase of 31 percent from the 1983 level. That pattern of employment suggests that the trade-off between employment of lower-level employees and salaries of higher-level officers is a critical one. In particular, the much more rapid growth of CDEs and NJCEs is a cause for concern. S1. Based on the computerized payrolls at the MoF and information from the Decentralization Division of the Prime Minister's Office (PMO), the best estimate of total non-parastatal public employment as of June 1990 is 208,149 (Table IV.2). Of that number, government employees on the computerized payroll totalled approximately 121,437 (58.3%); of which 107,602 were civil servants and 13,385 were non-civil servants. 58. Of total non-parastatal public sector employment, 87 percent (or 179,818) are employed at the central government level and the rest at provincial and district level. From another perspective, 55. 7 percent (115,877) of the total are regular civil servants and 44.3 percent (92,272) are non-civil servants. The majority of Civil Servants (92 %) work for the central government and the rest for local district councils (8%). About 80 percent of non-civil servants are employed by the central government and the rest by district councils. 59. The estimates of district council employment calculated for this report are provided in Table IV.2. Employment of permanent LAS staff at district level is 8,275; a considerable increase from approximately 4,000 23 in 1983. This trend is closely associated with the Government•s decentralization policy, which has resulted in a replication and multiplication by district councils of the same type of activity by the central government which the Economic Recovery Program is meant to change (see Chapter VI), 60. Perhaps the most distressing aspect of public employment at the district level is the excessive use of supposedly temporary NJCE, the number of which is estimated at 18,772. That number represents an approximately five- fold increase over such employment levels in 1983. A substantial proportion (perhaps as much as 35 %) of NICE are engaged in local government owned and managed industrial and commercial ventures. Most of these ventures have never generated a positive return on investment (or even on operating costs). Such negative results have now been compounded by the recent wage and housing benefit increases, which apply to all NJCE and have raised the wage bill for them to a level substantially higher than equivalent positions in the private sector (for further discussion of this issue, see Chapter VI). Incentives 61. Incentives are what transform capacity into performance. Incentives can be positive, negative, or the sum of both, but in most cases will be based on some relatively complex combination of interests. Such interests combine in different ways for different people; which complicates attempts to structure incentives in support of improved performance. Such combinations of interests are affected, at least to some degree, by the calculation of the opportunity costs of performing assigned activities and/or complying with established procedures. 62. The efficiency and effectiveness of Zambia•s public service have repeatedly been cited as causes for concern. This Report confirms the findings of another World Bank report issued in 1986, §./ which also concluded that: (i) Zambia•s public employees were poorly motivated; (ii) the size of public employment was excessive, particularly at lower levels; 11 and (iii) the public service was insufficiently productive. §/ In addition, wage bill costs were identified as one of the main factors contributing to the Government•s fiscal deficits. 63. At the same time, the broad decline in real income across all sectors of the economy has made employment conditions within Zambia•s public sector increasingly unattractive to technical, managerial, and professionally skilled personnel. That effect bas been compounded by Government's policy to compress the wage structure within the public sector so as to "narrow the income gap" between skilled and unskilled employees; an §./ Zambiaj Wage Policy and the Structure of Wages and Employment (Washington, D.C.: The World Banlc, 1986). 11 Zambia: Public Expenditure Review (Washington, D.C.: The World Bank, 1986). §/ Ibid and Planning and Budgeting in Zambia (Washington, D.C.: The World Banlc, 1983). 24 explicit policy goal by the government during the 1970s and much of the 1980s. 2/ The result has been perpetuation of reliance on expatriate skills and insufficient wage employment opportunities (see paras. 68 • 79). 64. Subsistence Compensation. Too many public employees in Zambia are paid salaries and benefits below subsistence requirements. The analysis of payroll data undertaken for this report clearly indicates that most Division Il and m employees are paid at or below the Zambian subsistence level; while most Division I employees are only marginally above it. With the rising cost of living and general erosion of public service pay and employment conditions, it is increasingly difficult for public employees to maintain even currently inadequate real income levels. Table IV.3 clearly shows the deterioration during the period 1986 to 1990. In 1986, a person at level S12, would find that a nutritionally minimal level of food for a family of four persons would require 57 percent of bis/her salary. By 1990, the same minimal amount of food for a family of four required 100 percent of salary. Such costs are for food only and do not include other costs of living. Thus, personnel at such levels are clearly below subsistence. It should be obvious that most such subsistence public servants will not be sufficiently motivated to perform at an acceptable le\-el. 65. For those above subsistence, the ,,____,,~,_,,_,_,,,,,,,._,, __ .,,_,_,7,:um,f.Sy+.)l....+'1 ~'~,.,__,_-:- specific incentives which might improve staff performance are likely to be more complex and particular-istic than for those below that level. However, the extent to which Government can identify and provide such facilities to staff whose remuneration exceeds subsistence levels, especially as compared to such conditions offered elsewhere, can have significant impact on performance. 66. Recent Increase in Personnel Emolwnents. Proposals to increase salaries and wages for government employees were seriously considered in preparing the 1990 budget and personal emoluments were programmed to increase by k3,000 per person. Such increases would have represented an annual growth rate in salaries and wages ranging from 7 to 30 percent; a rate of increase which was feasible within a fiscally sound budgetary framework. In the event, however, an extra-budgetary wage and benefit package was adopted in July 1990 without full consultation with responsible officials at the Ministry of Finance. This new package included an 85 percent salary and wage increase for all government employees and 100 percent to 300 percent increases in cash housing allowances for those not accommodated in Government housing. 67. Nevertheless, total nominal increases in July 1990 were far from sufficient to offset the wage and salary erosion resulting from consumer price increases in 1988 (56 pet.:ent) and 1989 (117 percent). In fact, during the last six years (1985 - 1989), total personal emoluments declined by more than SO percent in real terms (see Table IV.4). Given the rapid growth of public service employment during that same period, the reduced amount of aggregate expenditure on persona, emoluments covered increasing numbers of people. Thus, average compensation has been even further reduced. 68. or Compregion Wge Structyre. Overall inadequacies of the public service salary and wage structure are compounded by Government's long-standing political objective or-closing the income gap• (para. 63). 'J.l See Zambia: Wage Policy and the Strucpre of Wage and Employment. OJ!,Cit. 25 The need to contain wage expenditure is e direct result of the fiscal constraint, while the attempt to close the income gap between the highest and lowest salary and wage earners in the public service has been motivated by the egalitarian principles endorsed by the official philosophy of the State. 69. Thus, the structure of personal emoluments has historically been severely compressed. That has been even more the case in recent years. For example, in 1983, the most senior civil servants' salaries and benefits were approximately seven times that of the most junior civil servants. However, the current range from lowest to highest equals only a factor of about four (see Volume Il, Table A.9). Thus, from 1975 to 1990, the highest salary scales depreciated by 89 percent, while the lowest salaries' depreciated by only 55 percent. The July 1990 wage and benefit increases continued further the policy of "narrowing the gap• established by previous salary review commissions and again resulted in substantially higher cash increases for lower level civil servants (including CDEs) than that awarded to higher level civil servants; the former receiving an 85 percent basic wage increase plus kl 8 thousand per-year housing allowance while the latter received only the 85 percent basic salary increase. 10/ overall, the declining level of wages, salaries, and benefits in real terms has adversely affected the morale and effectiveness of all public employees, but the impact has been particularly acute on professional, technical, and managerial staff within the regular civil service. 70, The Competitivenes., of Public Pay. As reported above, the policy of narrowing the gap by granting successively lower percentage increases to higher level staff and higher percentage increases to lower level staff has resulted in an increasingly compressed wage structure. At the upper end, non-parastatal public sector salaries, wages, and benefits are not sufficiently competitive to attract and retain professional and managerial personnel. However, at the lower end, the situation is the opposite; remuneration at those levels are excessively generous relative to personnel income in the private and parastatal sectors. With specific reference to comparison with parastatal wage scales, unskilled Government employees have consistently received more compensation than their counte,parts employed by ZIMCO and its subsidiaries (with the exception of ZCCM). Thus, unskilled Government employees at lower levels received 40 percent higher salaries than their ZIMCO counterparts in 1983; by 1990 that margin had increased to 63 percent (as illustrated in Figure IV.3). A similar pattern also exists as between unskilled government and private sector workers. Thus, during the period 1975 to 1980, an unskilled worker in Government received compensation comparable to an equivalent worker in a private construction company. However, by 1990, the same unskilled Government employee received 30 percent more than his/her counterpart in the construction company. 71. Given the situation described above, the public service suffers from the worst of both worlds; it is quite attractive to unskilled and semi-skilled persons at the lowest levels, but is not competitive at the higher levels. Government still attempts from time-to-time to raise higher level personal emoluments so as to more closely approximate parastatal salaries, wages, and benefits. However, the net effect of such attempted adjustments has been for ZIMCO to use such increases in the non-parastatal public sector as the basis for determining increases in the personal emoluments of its own employees. The result has consistently been that increases in either sector have fueled increases in the other sector, without any substantial change in the gap between them. 10/ This calculation is based on the assumption that most higher level staff are provided with actual Government houses, while lower level staff receive the cash benefit in lieu of actual housing. 26 Figure IV.3 COMPARISON OF GOVERNMENT & PARASTATAL WAGE SCALE (in units of 1975 Kwacha per annum) Wage 7,000 i:----------------------- 6,500 6,000 5,500 5,000 4,500 4,000 3,659 3,500 3,000 2,500 2,000 ...,_____ 1,794 1,500 1,000 - - 500 - - 475 450 o ------'----------'------ --!l:=----_,J 1980 1985 1990 Time Government Parastatal 27 72. Benefits and Allowances. Allowances currently constitute an excessive proportion of total personal emoluments (for a comprehensive List of Allowances, see Volume II, Annex VII). The real decline of personal remuneration at the upper level has been further compounded by a similar pattern of benefit allowances; i&.,, across the board increases without reference to individual performance and without regard to differentiation of salary scales. The most important of the benefits contributing to the distortion of the overall wage structure is the Housing Allowance. 73. Housing and Housing Allowances. All officers are entitled to Government assistance for their lodging. This assistance is provided through actual accommodation in Government housing or through a monthly cash housing allowance. It is estimated that approximately 30 percent of public sector employees (or about 3S,000 officers) are provided actual accommodation, at a nominal charge of 10 percent of base salary, in Government housing. Provision of houses has represented a considerable benefit to staff in the past, but this is no longer the case with the increased cash allowance authori2:ed in July 1990; which those accommodated in government housing do not receive. 74. The number of Government employees reported to be receiving a housing allowance include only those paid through the computerized payroll system. As such, they comprise only 30 percent of the total number of employees paid through that payroll system. Three factors account for the fact that only 30 percent of those persons are receiving the cash housing allowance: (i) many of the employees who are paid through the computerized payroll system are also accommodated in Government houses; (ii) a large number of the persons enrolled on the computeriz.ed persons are precisely those who are actually accommodated in government housing (and, therefore, do not receive the cash allowance); and (iii) many of those recently eligible for receiving the benefit have not yet applied for it. At the same time, many of the CDEs and NJCEs, most of whom have been eligible for the cash housing allowance only since July 1990, are not paid through the computeriz.ed payroll system and, therefore, are not reported in it. Thus, the cu..rrent number of applicants (i.e., 36,359) will certainly increase during the coming months and that likelihood representatives a figurative fiscal •time bomb.• 1!/ 7S. The fact that !ill non-parastatal public sector employees are now eligible to receive the housing allowance means that the actual increase will clearly exceed the extra-budgetary allocation of 106 percent over the previously estimated total wage will for 1990. As people continue to apply, the overall amount will escalate monthly. Expenditures for housing allowances were k22S .1 million during the first two quarters of 1990, but jumped to kSS9.8 million during the third quarter; an increase of 149 percent (see Table IV .S). 76. Provisional 1991 estimates for the housing allowance, based on estimates of the numbers of all non-parastatal public sector employees, suggest that costs could rise as high as k9.6 billion during the 11/ A major problem for the Payroll Unit in MoF will be to ensure that people who are already accommodated in government housing do not also end up with cash housing allowances or vice versa. 28 current year. If, for purposes of iJJustration, only 70 percent of Government staff receive the housing allowance, the cost would still be excessive; LA, k7 .6 billion. Assuming that only 70 percent of total central government employees are finaUy enrolled for housing allowances during 1991, the number of beneficiaries would increase over ; : ~ : ; ; ; ; : ; ; ~ ~ ~ ; ; ; ;,~~,, T , ,, . . • ~ ~]1~ , •~ "benefits" are not part of the official .... terms and conditions of employment, they are a wen established part of actual practice. The financial value of such benefits are, by definition, not easy to measure. Examples of such indirect benefits include: (i) time made available for pursuing economic activities outside of government service; (ii) availability of government transport for use by family members; and (iii) the services provided by •messengers• and other marginal staff in the form of shopping for food and other household supplies. Thus, in terms of current practice in Zambia, it is much more difficult to determine the extent to which particular public employees are actually above or below subsistence. Personnel Manaeement 78. Three fundamental problems plague Government's attempt to effectively manage public sector employment: (i) lack of a clear and explicit policy; (ii) inadequate monitoring systems; and (iii) cumbersome and inflexible retirement and dismissal procedures. 79. Policy. In the absence of a comprehensive and clearly articulated public service reform program (see para. 124), the primary problem inhibiting rational and systematic staffing levels within Government bas been the breakdown of the policy which limits the recruitment of most public sector personnel to positions established and published in the Establishment Register {ER). The process of having positions officially established based on clear job descriptions and standards of qualification for candidates for such positions and requiring such candidates to pass examinations based on such qualifications became excessively bureaucratic. Thus, Government ministries and other entities increasingly resorted to an ml hQ£ process of hiring staff on "provisional appointments• and as supposedly temporary CDEs and NJCEs. However, such appointments result, in practice, in de facto regularization and in terms of tern.s and conditions of appointment are treated as if they were regular Civil Servant appointments. Thus, the Establishment Register (ER) has proven to be of little relevance to staffing decisions at the departmental and ministerial level. Although there bas been little increase in the number of authoriz.ed ER positions, the computerized payroll system clearly shows a totally different level of civil service employment (Table IV. 7). This is in large part due to the bypassing of the ER through the delegation of authority to employ such persons to sectoral ministries and provincial government agencies. Thus, recruitment regulations allow important exceptions to supposed mandatory adherence to the Establishment Register. such as those for the ministries of Education, Health, and Agriculture and the Zambia Police. Such exceptions have progressively run out of control. Thus, a major discrepancy of 34,000 civil servants (45% of the ER) can be observed between the computeriz.ed payroll (107,602 employees) and the ER for 1990 (73,126 positions). U/ U/ These excess persoMel, considered by MoF as "provisional appointments," arc employed primarily by the above listed ministries, but not exclusively. Approximate distributions of "excess" penonnel are: Ministry of General Education, Youth and Sports (21,000); Ministry of Agriculture and Cooperatives (3,500); Ministry of Health (2,200); Zambia Police (2,000); Ministry of Water, Land and Natural Resources (2,000); Ministry of Legal Affairs (2,000); Presidency and State House (680); and Ministry of Works and Supplies (65S). 29 ····-·· ·,:.; - 30 80. Unrestrained and dispersed authority to employ CDEs and NJCEs has been compounded by Government's practice to, in effect, confer permanent status on such employees after six consecutive months of service. That practice, recognized in labor union contracts, 13/ severely constrains Government's ability to adjust employment of temporary workers as requirements change. Since the requirement for CDEs and NJCEs varies substantially over time and across different sectors, maintaining a large permanent complement of such employees adversely affects the effectiveness of the public sector. 81. Monitorig. Although the computerized payroll system, operated by the Data Processing Unit (DPU) at the Ministry of Finance (MoF), covers virtually all central government civil servants (as well as those civil servants assigned to provincial levels or seconded to the district councils), the vast majority of CDEs employed at both central and provincial levels and all NJCEs employed by district councils are paid through lists maintained manually within their employing agencies. Although it can be argued that government staff should be paid by their employing agencies, ~uch employment should be monitored by central government so that it can know what the entire wage bill is for fiscal purposes. Yet, no central agency has been able to maintain any comprehensive list of all Government employees. This is in part due to the fact that many government agencies and district councils report employment and personnel expenditures to the Personnel Division much too late and in too partial a manner to be of any relevance for effective personnel and expenditure management. 82. Classified Daily Employees (CDEs). Only partial COE employment records are maintained in the computerized payroll system; which, according to the estimate calculated for this Report, indicates only 19 percent of total CDEs. 14/ Ministries and Departments which exercise discretionary authority for hiring CDEs are required to provide reports on the number of their employees to the Budget Office of MoF to support budget requests. However, adherence to that requirement has not been systematic. For example, during the FY1990 budget exercise, personnel officers in charge of COE employment did not often maintain updated information and, in most cases, such records are at least two years in arrears. Control and monitoring of COE employment is particularly weak for those ministries, such as Health and General Education, with significant operations at district levels, because coordination and communication between headquarters and field offices are inadequate. Finally, lack of effective monitoring and management of COE employment is also the result of insufficient funding for recurrent operations; i.e., no travel budget for central government to undertake field visits for supervision and monitoring purposes. 83. It is clear that, despite the decision taken by the Government in 1986 to reduce the number of CDEs, employment of such workers has expanded. No effective system exists for reporting, monitoring. and evaluating the employment of CDEs within ministries. 84. National Joint Contract Employees lNJCEs). In the absence of a computerized payroll system for district council employment, updated and reliable information on size and composition, either permanent or contractual, is difficult to obtain. Central government ministries have little control or knowledge of actual employment and personnel expenditures at subnational levels. This is due to two factors. First, central government has no control nor effective mechanisms to monitor actual district council expenditures from •Grants and Payments" transfers to district councils. One of the consequences of such inadequate information is that the Budget Office of the MoF simply assumes that 60 percent of such grants and payments are allocated by district councils to the 13/ These contracts committee the government to provide the same tenns and conditions of employment to all its employees without differentiating between regular civil servants and non-civil servants such as CDEs and NJCES. 14/ That most CDEs are not included in the computerized W!!!!! system is as it should be, since including personnel who are, at least theoretically, temporary in such a system tends to reinforce the perception that they are pennanent. Nevertheless, in the absence of a computerized persoMel monitoring system, the computerized payroll system is the only systematic instrument available at national level such monitoring (see para.144). 31 personal emoluments of staff; a proportion which intetviews at three district councils suggest is substantially too low. Thus, it is believed by MoF and some district council officials that an excessively large proportion of "Grants and Payments" is being absorbed by wage and salary payments; though the exact amount is not known. 8S. Retirement and Discharge Provisions. Government's General Orders specify condit:ons for termination of appointment 1S/ and retirement and conditions for dismissal. l§l 86. Retirement. Retirement provisions apply to regular civil setvants and LAS employees having attained "the pensionable age prescribed in the Pensions Laws;" which is currently SS years. However, Article 44 specifies cases where "an appropriate Commission may require an officer setving on permanent and pensionable terms of setvice to retire." Such cases include: (i) abolition of office; (ii) reorganization of ministries or provinces; (iii) medical disability; (iv) the national interest; (v) the 1 , ,blic interest (in cases of failure to perform assigned duties, incompetency, and/or disciplinary offenses); and (vi) changes in the prescribed retirement age. Except for medical grounds and the public interest, "forced" retirements require three months notice or the payment of three month's salary as a severance bonus. 87. A significant disincentive for early retirement is the inadequacy of pensions and the loss of such benefits as housing and education benefits. The basis for determining the level of pension is the gross salary or wage (excluding allowances). As bas been shown above, salaries and wages have been substantially eroded, while the levels of allowances, particularly the housing allowance, bas fared much better. Thus, any longer term solution to the redundancy issue will need to address this problem. Further, any significant early retirement program for Government staff will significantly expand the burden on the Pension Fund: the number of retirees will be higher while the average duration of pension payment is likely to increase. Thus, the primary constraint to the implementation of an effective retrenchment program which empbasi7.es enforcement of normal retirement regulations is the financial cost and the inadequate condition of the Pension Fund. Resolving those cost issues will be a key issue for the design of any sustainable retrenchment program. 88. Dismissal. For all practical purposes, dismissal is only permitted for reasons of major misconduct or criminal offense. Significantly, a permanent officer cannot be dismissed for strictly economic or financial reasons. Nonprejudicial dismissal of a permanent officer prior to normal retirement age can only be effected as "early" retirement with pension. 89. Termination of appointment only applies to: (i) officers on probation, by giving three months notice or paying the equivalent of one month's salary; (ii) officers hired for a specified period of setvice, by giving due notice or paying the equivalent of one month's salary; and (iii) officers hired on temporary setvice, by giving the period of notice prescribed in the employment agreement or providing a salary payment equivalent to that period. 90. In any event, current procedures for dismissing public sector employees are extremely cumbersome and inefficient. In recognition of that problem, the most recent approach by Government, in its attempt to achieve previous retrenchment targets, was to avoid dismissals and rely, instead, on inducing voluntary early retirements in response to a generous early retirement benefit package. That approach resulted in two other seriously negative consequences: (i) the benefits were so attractive because of their high cost and were, therefore, financially nonviable; and (ii) many staff who voluntarily retired were those the Government could least afford to lose in terms of efficiency and effectiveness criteria. The result was that Government was forced to essentially abandon the program. That early retirement benefit package is still the one currently in force while nothing bas been done to revise dismissal procedures. Thus, recent experience in Zambia and elsewhere in Africa suggests that dismissals of pensionable Civil Setvice employees for malfeasance should not be relied on as an effective instrument for IS/ General Orders, Chapter m. 16/ Ibid, Chapter IV. 32 meeting retrenchment targets because of the time required to accomplish them. Nevertheless, enforcement of dismissal procedures should be vigorously pursued for the purpose of enforcing discipline. Financial Manapment of Public Emplqyment 91. Under current conditions in Zambia, three financial issues are paramount: (i) reasonable limits on the aggregate wage bill; (ii) increasing the availability of financial resources for actual operations; and (iii) establishing and maintaining reasonable costs of necessary retrenchment. 92. Aaareeate Waae Bill Ceilin,. Without the establishment of an explicit wage ceiling, it is estimated that total wage bill costs would likely reach k14.8 billion (Government wage bill = k12.4 billion and District wage bill = k2.4 billion) for 1991. A particular cause for concern in that regard is the potential fiscal impact of the previously discussed Housing Allowance increases. 93. In response to the recogni7.ed fiscal consequences of such increases, Government has established an aggregate wage bill ceiling equal to the FY1990 wage bill plus 8 percent for grade creep (i.e., k9.0 billion); 17/ k6.6 billion at central government level and k2.4 billion at district level. The difference between the estimate for an uncontrolled increase resulting in an aggregate wage bill of k14.8 billion and the established ceiling of k9 billion equals kS.8 billion. That amount (kS,8 billion} establishes the fiscal target which must be achieved through fiscal restraint during FY1991. 94. Qperations Support. The effectiveness of the public service has been affected by insufficient provision of operating budget for civil servants to carry out their assigned tasks; which, in turn, is a symptom of the low priority given to actual delivery of goods and services as compared to employment of personnel as an end in itself. Real RDCs in 1990 were only 80 percent of the 1985 level, and even the level of RDCs in 1985 was considered insufficient. 18/ That general problem of insufficient financing has continued to date. Examples of such constraints include the fact that: (i) personnel officers in charge of employment of CDEs do not have travel budget to undertake field visits to supervise pay and employment practices; (ii) agriculture extension workers do not have vehicles to deliver consulting service to the field; and (iii) the Central Statistical Office does not have sufficient operating budget for its routine collection of statistics and publication of important data. 95. Nevertheless, it should be noted that "insufficient financing" is not, in and of itself, a constraint. Rather, the absence of adequate financing for the performance of a specific function or task is a consequence of: (i) an absence of prioritization among functions or tasks within the limits of financial resources and/or (ii) the assignment of low-priority to the function or task being considered. Thus, the current situation in Zambia suggests that either the Government has simply neglected to establish its own priorities as between the value of extensive public sector employment and efficient operations or among various functions and sub-sectors 2! it has consciously decided that employment is more important than actual operational performance. 96. Costs of Retrenchment. The scope of indemnities in case uf separation is a key financial issue. Based on Government's General Orders and other related regulations, it is estimated that a retrenchment program would generate net savings in personnel expenditures from approximately k949 million to kl.098 billion for 1991 and 1992 alone (see para. 137). In the case of dismissal, indemnities owed to employees are determined by the labor law, supporting regulations, and union contracts. With regard to union contracts, it should be noted that, if current stipulations in such contracts regarding separation allowances continue in force, the estimated savings reported above would be substantially reduced during the short-term phase of any retrenchment program. An eight percent annual grade creep is substantially higher than the norm for other countries; i&,,, more on the order of 2 - 3 pen:ent. Zambia-Public Expenditure Review (1986), op.cit.. 33 97. The supporting regulations for Civil Servants stipulated in the Qeneral Orders and other Jaws and regulations establish minimum conditions for retrenchment costs. However, more important under current practices are the terms of the various agreements between Government and the four relevant labor unions (see Volume II, Annex VIII, for the 1990 Agreement). Expectations with regard to terms and conditions of employment, based on the history of labor negotiations, far exceed the minimum conditions embodied in law and regulations. An indicator of the spirit of such nesotiated union contracts is the provision that none of the terms of any subsequent union contract can reduce the value of any prior contract. Procms 98. An almost complete lack of participation in the design and implementation of previous Government programs to reform the Public Service has contributed significantly to Government's failure to effectively implement such programs in the past. Thus, consideration of earlier reforms were closely kept by Government; responsibility being assigned to a small Working Group of senior officials assisted by external consultants. Currently, primary responsibility for addressing reform issues from a technical perspective has been contracted to a consulting subsidiary of ZCCM (VisionConsult), while Government's review is in the hands of a select Committee of Permanent Secretaries. Other stakeholders with a clear interest in the various issues of Public Service Reform have not been included as yet in the analytical nor negotiation processes required for the establishment of an effective program. Even within Government, a central player as important as MoF has been excluded from participating in union contract negotiations and, as shown in July 1990, can even be by-passed when decisions are made concerning wage bill increases (para. 52). 99. At the same time, it is apparent that such groups as relevant labor unions and informal groups of middle and lower level employees are viewed by Government as potential obstacles to implementation of an effective program and, thus, that they should be •kept in the dark" as long as possible so as to present them with a fait accompli. Such an approach is likely to result in such stakeholders "assuming the worst" and attempting to pre-empt the program even before they know precisely what are its actual elements. Evidence from Public Service reform efforts in other countries suggest that the level of political opposition from such non-governmental stakeholders does not, in fact, reach the significant levels normally anticipated; especially if they become a part of the process from an early stage. J!J/ It is also important to note, however, that the recommendation for such widespread consultation does not obviate the need for the few key agencies, such as MoF and the Personnel Department, to take ultimate responsibility for making necessary decisions. It is precisely to gain as much support as possible for such decisions that a process of broader consultation is necessary. D. Recommendations 100. Recommendations are presented in this Section under the following six headings: (i) Objectives and Phasing; (ii) Short-Term: Recruitment; (iii) Short-Term: Retrenchment; (iv) Short-Term: Foundation for Improved Performance; (v) Medium-Term: Improved Effectiveness; and (vi) Longer Term: Improved Efficiency. In combination, these recommendations address each of the issues discussed in Section C above. Objectives and Phasin& 101. The ultimate objective of Public Service Reform must be the efficient and effective performance - of Government's ap_propriate functions. That, in tum, requires staff who are well qualified in the necessary skills and motivated to use those skills efficiently and effectively in the service of Government. As discussed above, motivation results not only from financial rewards linked to performance, but also from an enabling envimnment consisting of political support, prestige, availability of supplies and equipment, supportive (rather than restraining) 19/ Barbara Nunberg and John Nellis, op.cit. and Louis de Mcrode, ~ - 34 bureaucratic procedures, and an appropriate working environment. None of those facilitatins conditions currently exist in Zambia nor can they be provided without a concerted high priority effort over the medium-term. No such reform program will be possible, much less sustainable, without sufficient finance; which will continue to be limited in Zambia for the foreseeable future. Therefore, the first and absolute precondition to achieving the ultimate goal of efficient and effective performance is to concentrate available finance on the most important priority Government functions. There are no short-cuts to such a strategy. 102. As has been stressed throughout this Report, a critical element in the recommended reform strategy is a fundamental change in the role of Government in the Zambian economy. As discussed in more detail in Chapter II above, such a fundamental change is inherent in the nature of Zambia's current economic recovery program. From a practical perspective, the Government's economic policy objective, its redefined role in the economy for the achievement of that objective, and the limited resources available to it for the performance of its appropriate roles, requires decisions as to its investment and operational priorities. Such priorities are classified in terms of two interrelated aspects: (i) sub-sectors which encompass priority public goods and services (identified in Chapter II, para. 12) and (ii) the priority functions which the Government must perform to provide such public goods and services (identified in Chapter II, para. 13). 103. The principles underlying the changing role of the public sector provide the criteria for the specific recommendations provided in this report. Actions taken to streamline the public service and meet fiscal stabili7.ation objectives through retrenchment should, during the initial three years of the program recommended in these pages, focus on those functions and those sub-sectors not included in priorities identified above. 104. In addition, it is important that Government's strategy and action-program carefully consider priorities and phasing (Chapter II, paras. 15 - 24). Therefore, the recommendations for action presented below are set forth in terms of phases over the short, medium, and longer term. Further, it is also important that the strategy and associated actions clearly distinguish among those which address fiscal stabili7.ation and improved efficiency and effectiveness objectives along two separate, but equally necessary, tracks (paras. 25 and 48; and Figure IV.2). The destination of the first track is a contribution to fiscal stabili7.ation in the shortest feasible time. The destination of the second track is significant improvement in the performance of Government operations over the medium and longer term. Travelling only along the first track would ultimately be self-defeating. Travelling only along the second track is ultimately impossible. Short-Term; Recruitment 105. The burden on the budget caused by the excessive size of the civil service, and its associated wage bill, must be reduced. Therefore, the need to grant discretionary authority to the ministries to make necessary staffing decisions must be balanced with the fiscal implications of maintaining excessively large civil service employment. In this regaro, the role of the Personnel Division and the Public Service Commission need to be strengthened to enforce existing regulations, including: (i) observance of General Order provisions on recruitment and (ii) strict respect of the Establishment Register (as stipulated in Provision #9 of the General Orders). Furthermore, there is a need to strike a balance between the demand for and supply of trained personnel. The current recruitment process is largely •supply driven,• because these ministries are obliged to employ all trainees graduating from their pre-service training institutions without regard to the established recruitment process and actual needs. This practice should be discontinued immediately and avoided in the future. Thus, the Government's pre- service and in-service training policy should be revised in order to adapt the siz.e of classes to the projected demand in the public sector. 106. In addition, the recruitment of CDEs and NJCEs should be drastically reduced. With particular reference to district councils (see Chapter VI), authorization to hire NJCEs should be limited, until such time as districts are largely self-financing, through a sensible reduction of budget allocations for such employment. Ministries and provinces should also be clearly instructed that such reduced allocations constitute absolute ceilings for which they will be held accountable (see Chapter V). 35 Short-Term; Retrenchment 107. The criteria for the establishment of future wage bill ceilings have not yet been determined. In any event, the determination of specific ceilings is an issue beyond the scope of this paper. 2Q/ Therefore, the analysis and recommendations provided below focus primarily on the mechanisms that could be employed to implement a retrenchment program to meet whatever targets are established. That approach is illustrated, in part, by focussing on the wage bill limitations established for FY1991 in Z.am.bia's current Economic Reform Program. 108. In order to meet the fundamental fiscal stabilization requirements of the FY1991 phase of its Economic Reform Program, Government bas decided to hold the current aggregate wage bill at k9 billion (of which k2.4 billion is allocated for personnel employed by district councils) during that year (see para. 93). The implication is that the real wage bill level for 1991 will be reduced by domestic inflation. 109. Against this target, the Government faces two options: (i) allow inflation to cut the wage bill across the board for everyone in the system m: (ii) introduce retrenchment of employment in some part of the system to reduce obvious redundancies and apply the net savings of retrenchment, if any, to reform the salary structure. Given the severe compression of the wage bill structure and consequent steady loss of managerial and technical skills to the parastatal and private sectors, Government bas decided to embark on the second course of action. To accomplish that objective, it will be necessary to define a level of wage bill reduction and translate this level of reduction into a given level and mix of staff retrenchment which would generate net savings (i.e., minus costs of grade creep and retrenchment) for redistribution to the upper scale of the wage bill system. 21/ 110. Nevertheless, a cautionary note is in order here. It should be understood that while retrenchment of staff in the short-term is a necessary precondition for subsequent redistribution of savings for decompression of the wage structure, all but token actions in that latter regard will necessarily be deferred to the medium-term (see para. 137). In any event, to translate such targeted savings into numbers of staff to be retrenched, the Government bas two further options. 111. Option one involves prgportional distribution of retrenchment across regular civil servants, LAS employees, CDEs, and NJCEs. The average annual salary for central government civil servants is based on the range of scales from S15-S19 and for CDEs is based on the average annual wage of k20,500 at about COE level 2. Similarly, the average annual salary of district council employees is based on the range of scales for LAS16- LAS20, and, for NJCEs, is based on the average annual wage of k20,000 at about LCDE level 2. As shown in Table IV.8, the result would be retrenchment of a total of 13,054 civil servants and LAS employees from grades S15-S19 or equivalent and a total of 7,717 for CDEs and NJCEs at both central and local government levels. 112. Option two involves disproportional retrenchment with heavier impact on CDEs and NJCEs; i.e., by imposing an aggregate wage retrenchment of 25 percent on such employees as compared to a five percent 2Q/ The issue of appropriate wage bill ceilings would be an appropriate subject of this paper only if it might reasonably be expected that any ceiling that might be established to meet fiscal stabilization targets would seriously affect Government's ability to perform its high-priority functions and maintain the staff necessary to so do. Given the current size and structure of Zambia's public employment, that problem is unlikely to arise any time during the next three yean. 21/ It is often argued that the costs of retrenchment should be treated as a category separate from the wage bill because it is not a recurrent, but a "once and for all" cost. Nevertheless, retrenchment costs are treated here as charges against the recurrent wage bill for two reasons: (i) the established policy in Zambia, as part of its fiscal stabilization program, is that such costs must be financed from savings relative to the current wage bill and (ii) the retrenchment program is expected to continue over several yean and will, therefore, be a recurrent cost for all practical purposes over that period of time. 36 ifJlltif1:lifjillti: '<w~,~=t,,,,,,,i:,i=,:,:t-~,,i@~r:,;,y,im,=r:i,i,::i;:;g$fi1iliiliif (~ I Ii - . ;_0•i~:~r,:;::~1~M"~~~~~!~:;;~ 37 aggregate salary retrenchment for civil servants. Thus, Sections m of Table IV. 9 suggest alternative retrenchment targets. On the same basis of average salaries and wages, retrenchment of civil servants and LAS employees would be only S,023 and CDEs and NJCEs would equal 16,077. Thus, retrenchment of CDEs and NJCEs in option one would equal only 10 percent of their total, while in option two retrenchment of such employees would equal 21 percent. 113. Whichever of the two options are chosen, the mechanisms to be used for retrenchment and the sequence of actions remain essentially the same. Therefore, consistent with the approach outlined at the beginning of this Chapter (paras. 48 .. 49 and Figure IV.2), the focus here is on: (i) wage bill "cleaning;" (ii) immediate retrenchment of "obviously redundant" personnel; (iii) strict application of retirement regulations; and (iv) strict application of General Order provisions leading to dismissals. 114. Wye Bill "ClfflJling". As learned from experience with retrenchment programs elsewhere in Africa, this is the least disruptive and politically threatening mechanism available to achieve reductions in the aggregate wage bill. '1,1,,/ However, it is not simply accomplished. Zambia's own experience in this regard began to gather fruit in 1989, when the Payroll Section of MoF completed a payroll "cleaning" process focussing on retirees, shost workers, and overpayments of salaries and benefits not due within the Ministry of Finance itself. This was accomplished by comparing expenditures over successive months and checking the detailed lists of people enrolled on the computeriz.ed payroJl system in search of discrepancies. During 1990, the Permanent Secretary, MoF, obtained authorization to extend the "cleaning• process to other payrolls (Health and Judicial Departments have been completed; General Education is next on the agenda). These exercises have, as of June 1991, resulted in the identification of nearly 1,000 redundant entries in the computerized payroll system accounting for a total of approximately k3S million in annual savings. 115. Currendy, that exercise is beins followed by a physical check of persons paid through both the computerized and manual payroll lists within selected ministries and at district council levels. Progress in this regard is encouraging. It is conservatively estimated that such physical checking could eventually achieve no less than a saving of k2S0 million or a 7 percent payroll reduction (ghost workers, multiple payments, and so forth); particularly regarding the teachers' payroll. Therefore, it is strongly recommended that the current program be continued. its results validated and additional targets established based on extrapolations from current results, and an ongoin& system for monitoring the payroll for such discrg,ancies be institutionalil.ed; perhaps within the Auditor- General's Office. 116. Retirement and Dismip.gk. Retrenchment objectives can be achieved through one or some combination of the following three mechanisms: (i) voluratary degarture; (ii) early retirement; and/or (iii) dismissal. In the first case, departure is the result of a decision made by the employee. In the second case, some early retirements can also be expected to be voluntary; while yet others will need to be required by Government. In the third case, the decision is made by Government and the employee is obliged to leave. None of these alternatives are without financial cost, since some amount of separation allowance is required in all three cases. Further, dismissals, and required early retirements, are, by their involuntary nature, substantially more difficult to implement. 117. Dismissals: Immediate Retrenchment of "Obvious Redundancies." In the framework of the high/low priority approach, there are many employees in the lower categories, including regular civil servants, LAS employees, CDEs, and NJCEs (scales 15 to 19), who provide marginal or superfll•?us services, or who are obviously in excess. It is expected that immediate measures to retrench such obvious redundancies ~ . 20 drivers for 4 vehicles, three secretaries for one typewriter, and so forth) could apply to as much as 4 percent of regular Civil Servants and 21 percent of non-regular employees assigned to non-priority sub-sectors and/or non-priority '1,1/ Barbara Nunberg and John Nellis, op.cit. and Louis de Merode, op.cit.. . ._:,;.o;. .... $:,:,>f>.*' , .-.•.❖.•-···· fil:!J.h: m,ili~~j ; 39 functions. 'l:11 However, the significance of dismissing obviously redundant staff extends beyond merely relieving the financial burden on Government. Such staff also have a negative aff~t on the morale of colleagues who actually have significant functions to perform, but who receive the same salaries and benefits as those who have little or nothing to occupy them; as well as serving to divert the attention of scarce supervisory staff. Although dismissing even those persons who are obviously redundant requires complex and time-consuming procedures under both current laws and negotiated labor union contracts, the effort to dismiss obviously redundant staff should be initiated immediately. I 18. Strict Application of Retirement Regulations. The provisions of the General Orders pertaining to mandatory retirement should be enforced (see paras. 86 - 87), including more extensive use of the provision for systematic early retirement of peo_ple in lower categories ~ . SIS-S19, or equivalent). If, as the result of enforcing retirement provisions, persons with technical, professional, and/or managerial competence in key areas would be lost to the public service, exceptions, in rare cases, could be allowed. However, such measures should be considered transitory, until appropriate replacements through promotion could be arranged. 119. Significant reliance on voluntary early retirement raises two issues: (i) assured retention of staff with high priority skills and (ii) the costs of an early retirement benefits package. The first issue can be resolved by establishing an explicit and well-understood policy that early retirement by individual staff is a benefit, not a right, and, therefore, individual applicants must be approved by Government. The second issue is, under current Government regulations, more problematic because the current benefits fackage is overly generous and, in aggregate, cannot be finan~ for any significant numbers of staff. Therefore, it is clear that Government will need to substantially revise the current benefits packaae for early retirement. Such a new package will need to offer compensation which is both attractive to the employee and yet not too costly or extravagant. 24/ 120. The potential costs of retrenchment will vary depending on several options available to Government. Nevertheless, assuming that regular dvil servants might find it attractive to leave the public service voluntarily for a reasonably enhanced severance payment (say six to twelve months salary) in addition to a proportional pension (or the reimbursement of their pension contributions), this Report estimates that the potential cost of the retrenchment program during its first two years would be in a range from k277 million to k3S 1 million (see Tables IV.8 and IV.9). For those not willing to retire voluntarily, the provision in the General Orders that provides for required retirement in the national interest might be invoked for the regulatory severance payment of 'JJ/ Since Confirmation of employment following the statutory probation period has evolved into a purely technical process which no longer draws much attention, large numbers of emp?oyees and their supervisors have neglected to formalize it. Thus, many employees remain technically on probation status. Therefore, such employees caMot invoke any legal right to retirement or institutionalized dismissal procedures. Theoretically (and legally), such employees could be dismissed with a severance payment equal to only one month's salary. It might, therefore, be acceptable to retrench such employees without significant resistance for an enhanced severance payment of, say, six months. 'M,/ Such a program should include a component to assist regular Civil Servants (including LAS employees) and Ism&: serving CDEs and NJCEs to find suitable employment in the private sector. Nevertheless, note should be taken of the empirical evidence from such programs in other countries, particularly in Africa, that elaborate programs to retrain public employees for new occupations in the private sector have not been successful; primarily because only a small percentage of redundant persoMel have availed themselves of such retraining opportunities and because of the high per capita cost of such programs for those few persons who do take advantage of the opportunity. Therefore, it would likely be more attractive to any potential beneficiaries (and more efficient as well) if the estimated costs of any proposed retraining program are simply monetized and distributed on a proportional basis to those who do retire early or depart voluntarily. Thus, in addition to an enhanced severance payment, Government should ideally limit its retrenchment bonus program to institutionalized outplacement services; see Barl>ara Nunberg an~ John Nellis, op.cit. and Louis de Merode, op.cit .. 40 three month's salary (which might be extended for, say, six months). Z5_/ It is not possible to estimate whether the latter provision would need to be invoked during the initial three-year period of the retrenchment program until the results of the restructured voluntary early retirement program is known. Obviously, the enforced non-voluntary early retirement option should be invoked only as a last resort during the first three-year phase of the program. 121. Strict AJ!plication of General Order Provisions Leading to Dismissals. The General Orders contain several provisions (Chapter IV) specifying that dismissal can result as a sanction for criminal offenses. These provisions should be immediately and widely applied to all officers and employees convicted of severe lack of discipline or misbehavior; corruption; unjustified absences or desertion; and/or critical professional deficiencies, such as mismanagement or misuse of public funds. In particular, these provisions should be applied to employees caught in attempts to receive undue salaries and allowances. That, in tum, requires that the rate of investigations and, where warranted, legal action against staff accused of such offenses should be increased. Consideration should be given to the formation of a Parliamentary Standing Committee on Corruption which could hold public hearings on charges contained in reports of the Auditor-General. In addition, Government should prepare a "White Paper" which clearly specifies which specific types of conduct it actually intends to punish with outright dismissals (as differentiated from acts which it will punish with demotion, loss of salary or wage increment, fines, and so forth). Government's commitment to taking such action should be explicitly stated in the "White Paper" in a convincing manner. In any event. if the program recommended here is to have any J!ractical effect, it is essential that dismissal procedures are enforced, with vigor, at an accelerated pace and that early retirement not be used as an alternative to dismissal where the latter action is more appropriate. That is especially the case given the current structure of the early retirement benefit package. Short-Term; Foundation for Improved Performance 122. Although fiscal stabiliz.ation objectives must necessarily drive reform of public sector employment in non-priority sub-sectors during the short-term, such a program must not be conducted in a manner which reduces even further Government's capacity to perform necessary functions in high priority sub-sectors. Thus, as emphasized at the beginning of this Chapter (para. 48), a protective boundary should be placed around high priority sub-sectors during the earliest phases of staff retrenchment and reorganiz.ation efforts while necessary studies and proposals for action in those priority sub-sectors can be initiated in parallel during that same period. Basr.:d on such studies and careful review of proposals for action, a systematic improvement program could be implemented in the medium and longer term. Thus, the long drawn-out, but necessary, process required to prepare and implement a systematic improvement program in high priority sub-sectors need not delay the achievement of immediate fiscal objectives. 123. In that context, it is worth reformulating an agenda for action, based on the three issues affecting public sector performance identified earlier in this Chapter (para. SO): (a) Government should establish explicit policy priorities for public sector staffing (for which this Report might serve as a basis); (b) A new structure of public sector staffing focusing on strengthening managerial and technical capacity J and reducing redundancies at the lower level, should be established explicitly based on those priorities; 25/ It is possible that the "national interest" could be invoked to retire employees early based on a claim of economic emergency, but the legality of such a measure is not clear. Under such circumstances, a severance payment higher than the prescribed three months would probably be necessary. On the assumption that such a "bonus" would be necessary as part of any significant retrenchment program, the calculations of retrenchment costs are based on a range of six to twelve months of salary. At a later stage, the nonnal benefits provided by the applicable Jaws could be invoked for retrenchments resulting from a program to improve efficiency and effectiveness. 41 (c) A new structure of incentives for performance, including a substantial improvement of personal remuneration for skilled personnel, needs to be established; (d) Personnel and financial management requires substantial improvement; and (e) An appropriate and effective process for generating sophisticated understanding of the crisis in public sector employment among key stakeholders and mobilizing necessary support, through negotiations, for a program to implement acceptable solutions needs to be designed and implemented. Of these five actions, the formulation of a comprehensive and fully articulated policy (agenda item [a] above) and the design and implementation of an effective process for generating support among key stakeholders (agenda item (e] above) require immediate action. The remaining three agenda items are discussed further below in terms of medium-term effectiveness and longer term efficiency. 124. ~ - For a program of public service reform to be successful, a clear articulation of public sector policy will be necessary. Such a policy should be based on an initial commitment to three tasks: (i) design and initiation of a comprehensive medium-term capacity-building program for the improvement of public sector performance of functions in high priority sub-sectors (see Chapter II for identification of priority functions and sub- sectors); (ii) identification of specific limits on public expenditure for noa-parastatal public sector employment to be effected through retrenchment of staff in non-priority sub-sectors; and (iii) establishment of a formula and procedures for redistributing savings from non-priority sub-sectors to staff performing key functions in high priority sub-sectors. This Chapter has focussed on an analysis of issues affecting implementation of tasks and provides recommendations for a coherent program of public service reform. The key to such a policy is that appropriate reforms of the public service must be consistent with broader changes in the role of the public sector inherent in Government's Economic Recovery Program. Further, the relationship between public service reform and the changing role of the public sector should be explicitly stated in a public document. 125. Pr:«>ceH. Government's current practice of restricting participation in consideration of public service reform options to only a small group of senior officials should be changed. As argued elsewhere, evidence from Public Service reform efforts in other countries suggest that the level of political opposition from potential non- governmental opponents of such reforms has not, in fact, reached the significant levels often anticipated. Therefore, Government should prepare a program for consultation and on-going negotiation with key governmental and non- governmental stakeholders in order to: (i) avoid pre-emptive opposition based on uninformed speculation and (ii) outline a clear, but broad, strategy which can realistically be implemented based on necessary adaptations to the realities of the situation as it evolves. Such a strategic outline could then serve as a framework for those who must make the concrete and hard decisions which are necessary to give practical effect to public service reform. 126. Such a program requires careful crafting. An initial step in that direction is a workshop for approximately 45 Government officials tentatively scheduled for November 1991 for the review of this Report. It is expected that issues concerning the design of a longer term process for inclusion of other stakeholders over the next 12 to 18 months will be addressed at that workshop. Medium-Term; Improved Effectiveness 127. The medium-term is defined here as the period extending from approximately the fourth year of the program through and including the seventh year (i.e., approximately 1995 - 1998). Three full years ought to be sufficient for the priority focus on fiscal stabilization requirements to have largely run its course and to allow a shift to a primary emphasis on beginning implementation of a well-designed program for improved performance. Nevertheless, full implementation of such an improvement program will require many years extending well beyond this medium-term period. Further, improvements are most likely to be in terms of relative effectiveness, rather than 42 significant efficiencies (beyond those aggregate efficiencies already accomplished during the short-term through significant levels of staff retrenchment). 22,/ 128. In addressing improved effectiveness objectives, the recommendations provided here focus on three areas: (i) staffing and personnel management; (ii) incentives; and (iii) financial management. 129. Staffing and Pmonntl Management. There are three technical requirements for the specification and management of an appropriate public service staffing pattern: (a) Disciplined adherence to a rational recruitment policy linked to organiutional requirements; (b) Sufficient flexibility in reassigning staff as requirements change; and (c) A sustainable program to facilitate targeted early retirements and voluntary departures. 130. To meet those requirements, various methods are normally employed, including: (i) data collection and analysis; (ii) job inspections; (iii) organization and method studies; (iv) ratio analysis; and (v) training. 27/ 131. Data Collection and Analysis. Inadequate data for determining appropriate staffing requirements and for aggregate management of the personnel function is a major problem in Zambia. No single individual or organization anywhere in Government bas accurate data on the size, distribution, functions. or cost of aggregate non-parastatal public employment. Although the "wage bill cleaning" effort currently underway will generate partial data in that regard on an incremental basis, more systematic efforts will be necessary to establish an institutionalized data collection and analytical capacity within the Cabinet's Personnel Department. Institutionalizing such a system will be the key to success in this regard. Evidence from Public Service reform efforts elsewhere in Africa clearly indicates that conducting ad hoc censuses of public employees in response to periodic reform efforts are time- consuming, expensive, and inadequate for systematic personnel management requirements. ~/ Therefore, ensuring that some form of computeriz.ed data base is designed and established should be a high priority for the medium-term. Such a data base should be based on quarterly reports from all Government agencies concerning actual filled positions as specified in the ER and monthly reports on the employment of all CDEs and NJCEs and be capable of linking that information to actual expenditures relative to approved budgets; hence the need for the system to be computeriz.ed. 132. Job Inspections. The purpose of job inspections is to determine: (i) what functions need to be performed to achieve each organiution's objectives; (ii) the suitability of the organization's structure to facilitate performance of those functions; and (iii) whether or not the staffing pattern is appropriate for the effective and efficient performance of those functions within a suitable structure. That is a mammoth task if applied to all Government agencies. Government bas, in fact. attempted to undertake such analyses before; all such efforts having 26/ There are at least two schools of thought concerning the relationship between improved effectiveness and improved efficiencies. One school argues that effectiveness must be improved prior to addressing concerns for efficiency. The other school argues that an overriding concern for effectiveness can create conditions inimicable to any attempt to achieve efficiencies in either the short or longer tenns. This Report does not attempt to resolve that debate. Suffice it to say here that no actions should be taken justified solely on the grounds of improved effectiveness; any improvement efforts should carefully consider the possible negative effects they might have on improving efficiency over the longer tenn. 27 / This follows the survey of refonn components discussed in Barbara Nunberg and John Nellis, op.cit .. ; from which many of the examples have been borrowed. ~/ Louis de Merode, op.cit.. 43 foundered because of the time, energy, resources, and skills required. Nevertheless, some kind of effort of that kind is necessary to provide a basis for rational staffing decisions. From a practical perspective, therefore, Government should concentrate its efforts on only those offices performing high-priority functions within organizations operating in high-priority sub-sectors. Further, such analysis can be largely undertaken by consultants under the supervision of senior officials within those various organizations. The results of such analyses should be submitted, however, to Cabinet Office for review and decisions concerning the reorganization of Government agencies. Such an exercise, however, should not be undertaken as simply a n ~ effort. Rather, this function should also be systematiz.ed within an established institutional framework so that it can be repeated for each organization on a periodic basis; say every five or six years. 133. Organization and Method Studies. Organization and method studies consist of evaluations of the efficiency with which the same service is provided in a similar manner by more than one organization. By comparing the relative efficiencies of multiple organizations performing the same function, it is often possible to identify weaknesses, bottlenecks, and inhibiting procedures. This is, however, a technique of sufficient sophistication that it is not suggested here as a high-priority during the medium-term. Nevertheless, if and when serious doubts arise concerning the efficiency with which a particular service is being delivered, such a targeted study should be considered. 134. Ratio Analysis. Ratio analysis is another highly sophisticated technique which requires some understanding of the relationship between numbers of staff performing specific functions to some other variable ~ . medical officer/bed ratios, teacher/non-teacher staff/student ratios, or agricultural extension/farming population ratios). It has been argued that "ratio analysis is indicative rather than conclusive; it indicates needs for more specific policy reviews, and ultimately more detailed plans for recruitment, training, and staff development.• 29/ Nevertheless, it can also be used in the immediate term to identify obvious redundancies; as discussed earlier above (para. 117). 135. Training. The importance of the training function in maintaining an efficient and effective public service is well understood. Yet its role is too often overemphasi7.ed. Training is not sufficient to improve performance in the absence of the other facilitating requirements outlined earlier above (para. 101). In any event, even if other supporting requirements are met, an efficient training program must be systemi7.ed and linked directly to performance objectives. Ad hoc and/or supply driven training courses should be avoided. Training should be provided in response to demand by operational level managers who have identified specific skill deficiencies among staff. 136. Incentives. Rationalization of the wage, salary, and financial benefit structure must be the first priority given current conditions in Zambia; which means in the first instance decompressing personnel emolument scales (paras. 68 - 71). Government should make it widely known among public employees that this is its objective within five years (i.e., by no later than 1996). 137. Redistribution of Wage Bill Savings. The result of retrenchment calculations for 1991 and 1992 is a potential net savings of k598 million under option one and k549 million under option two. 30/ These levels 29/ Barbara Nunberg and John Nellis, m?£il.,. 30/ For the purposes of this report, the fonnula used for calculating savings from non-priority sub-sectors for redistnbution to staff perfonning key functions in high priority sub-sectors assumes a reduction in the current aggregate wage bill of 12 percent during FY1991. In nominal expenditure terms, such a reduction would equal a savings of about k81S million. That amount is disaggregated in Section m of both Tables IV.8 and IV.9 in tenns of wages and salaries at central and local government levels. The formula also posits that persoMel in the salary range S01-S06 would constitute the primary beneficiary group. The retrenchment scenario for the subsequent two (continued ...) 44 could be increased further through elimination of •ghost workers;• such •workers• might account for as much as 7 percent of current costs of the total wage bill (see para. 115 above), or another net savings of k400 million to kSOO million. In summary, after allowing for various costs associated with retrenchment (see para. 96), it can be expected that such a program could result in sufficient savings to allow expansion of the top civil servant scale (S01- S06 or equivalent) to a level sufficient to maintain their salary in real terms, through increases in nominal terms, during 1991 and 1992. That would, in tum, result in extending the upper levels of the compressed wage bill structure, if, as recommended, the nominal wage bill is held constant for the lower levels of public sector employment. 138. Non-Financial Incentives. Even though it is difficult to see beyond income as the first order priority, increased income for higher-level public servants is unlikely to be sufficient for improved performance. Such increases would only meet a fundamental minimum requirement. The other facilitating conditions identified earlier (para. 101) will also need to be in place. Therefore, it is not too early to begin identifying non-financial incentives appropriate to Zambian culture, as well as performance criteria and procedures to be applied to decisions concerning the distribution of such incentives to individual staff, once established. 139. As described above, non-financial incentives can take several forms in a variety of combinations; including culturally determined values (see para. 77). The point remains, however, that the purpose of any incentive system must be to improve performance. That being the case, although the criteria for evaluating personal performance and the process through which such evaluations are made should be sensitive to Zambian cultural concerns, it remains necessary that evaluation criteria must also directly linked both financial and non-financial incentive benefits directly to personal accomplishment of performance objectives. 140. Financial Management. Financial management is discussed in much more detail in Chapter V of this Report. Therefore, the discussion here is limited to two elements related directly to Public Service Reform: (i) wage bill budgeting and (ii) public employment expenditure and control. 141. Wage Bill Budgeting. This issue is linked to the relevance of the Establishment Register (ER) as a budgeting document (see para. 105 above). The ER procedure, involving the Ministry of Finance, the Personnel Division, and the Public Service Commission, was intended to clearly define required positions within Government and establish such positions within Government's staffing pattern as the basis for recruitment into the public service. Failures in that system should be addressed in two phases: (i) the enforcement of existing legislation regarding the ER 31/ and (ii) revision of the methodology for establishing positions within the ER so as to assess personnel needs on an effective management basis. 142. Expenditure Control and Monitoring. The 1989 Financial Report reveals that personnel expenditures exceeding budget allocations is a common practice. The MoF is little informed on the way salaries, wages, and benefits have been paid, on actual payments, and on unpaid salaries to be returned, since adherence to the financial management reporting system is deficient. Ministerial departments and provinces do not adequately report personnel expenditures to their headquarters, whose reporting to MoF is also inadequate. This failure to adhere to established procedures is particularly evident regarding COE and NJCE payment schedules and constitutes a serious problem with regard to the functioning of the system for maintaining accounts. 30/( ...continued) years would be based on the same principle as described here but with the wage bill ceiling defined as a given percentage of GDP, say about three percent. 31/ As an interim measure during the immediate tenn, all personnel managers should be instructed to comply with the ER and not to exceed agreed openings and such managers should be held accountable for not exceeding such ceilings. 45 143. The absence of such information at MoF is a key institutional issue, given its leading role for budget preparation and execution. The MoF's prerogatives to centralize information should be urgently strengthened through the establishment of an institutional reporting system from local units to ministries' headquarters and to the MoF, under the authority and coordination of the Payroll Section, MoF. Ministries and provinces must be obligated to report regularly on their recruitment and dismissal actions; especially regarding CDEs and NJCEs. In addition, MoF should clearly be directly involved in negotiating changes in salary, wage, and benefits scales. 144. In addition, the effort to get all non-parastatal Government employees (CSs and CDEs, including provinces) included in a computerized personnel information system should be completed as soon as possible (paras. 82 and 131). This system would need to be sustained by a training component to provide in-service training to all concerned public service personnel officers. 145. Finally, Controllers should have the power de {acto, rather than merely de jure to oppose any personnel expenditure committed or ordered in disregard of the financial regulations. Expenditure controls should be strengthened by: (i) enforcing the accountability of ministerial and provincial managers; and (ii) enforcing the power of the Senior Permanent Secretary, MoF, to apply sanctions for abuses. The AG's annual report should clearly highlight financial misbehavior and major offenses. Lonaer Term: Improved Efficiency 146. Almost all of the actions for improving the performance of the public sector initiated during the medium-term will need to be sustained over the longer term as well. Other actions deferred beyond the seven year medium-term cannot be expected to be identified in any detail; except to state that the transition from the medium- term to the longer term should be characterized by an increasing concern for efficiency beyond merely acceptable levels of effectiveness. Nevertheless, two longer term objectives can, and should, be identified at this early stage: (i) the contractina of significant levels of public sector production functions to the private sector and (ii) a major restructuring of personal emoluments from an over-reliance on benefits to an almost exclusive reliance on salaries and wages. 147. Public Sector Contracting To Private Sector. It was argued in Chapter II that the Government's responsibility for financing priority public sector goods and services did not necessarily require it to actually produce such goods and services. Responsibility for production could be contracted-out to private sector firms competing among themselves for the award of such contracts on a periodic basis. 148. Contracting responsibilities for production to the private sector would relieve Government of the responsibilities for negotiating terms and conditions of employment for a large number and variety of requirements. It would also relieve Government of the need to directly employ large contingents of often under-utilized CDEs and NJCEs. Contracting could also be arranged between Government and individuals;~. the •tengthman" system of road maintenance allows Government to contract with individuals for the performance of specified types of routine maintenance on designated sections of roads; with payment linked directly to performance. For other types of services, Government could contract with firms or groups of individuals~. watchmen, building maintenance, simple construction). However, such an approach implies; (i) a substantial change in current employment policies; (ii) changes in the mix of skills required of the staffing pattern; and (iii) changes in procurement regulations and practices so as to decentralize and speed-up the bidding and decision-making process while ensuring adequate review and transparency of that process. Nevertheless, even though such changes would require some reorganization and retraining of staff, that approach would provide Government the flexibility it requires to improve efficiencies in the provision of public goods and services at potentially lower costs. 149. Such an approach would solve three fundamental problems currently faced by Government: (i) the need to limit public sector employment; (ii) the need for more flexible mechanisms for staffing Government's responsibilities for production of public goods and services; and (ii) the inability to recruit and maintain highly qualified technical and managerial staff at the middle and higher levels. These three problems would be substantially 46 reduced by the simple expedient that full-time Government staff would no longer be required in excessively large numbers. . 150. Structure of Personal Emoluments. A rationalimtion of the structure of personal emoluments should emphasiz.e salaries and wages rather than allowances. The value of appropriate benefits should eventually be monetiz.ed and such monetary values should be incorporated into the wage and salary structure. It should be noted that the primary advantage in monetizing both direct and indirect costs of "income" to public employees is that it makes the costs of public employment more transparent than is the case when significant benefits are in the form of informal practices or provision "in-kind." In addition, such an approach would accomplish three other worthy objectives: (i) remuneration could be more directly linked to individual staff performance; (ii) administration of the wage bill would be substantially simplified and abuses of benefits should be reduced; and (ii) Government would be relieved of the need to manage and maintain a significant stock of houses and other unnecessary, non- producing, physical "assets." The first priority in this regard should be a Review of Government's policy regarding the provision of housing to staff with a view to eventually eliminating its ownership and maintenance of such housing (with the exception of special posts and locations; such as housing for customs officials at border posts and so forth). 47 V. PLANNING, BUDGETING. AND FINANCIAL MANAGEMENT 151. Planning and revenue generation, budgeting, and financial management are key functions of Government regardless of its role in the economy. These various financial functions are essential for transforming Government's decisions into action. 152. This Chapter describes and evaluates Zambia's public sector financial management system and presents a recommended strategy for improvement. In doing so, this report emphasizes budgeting and expenditure control as the primary mechanisms for financial management; as well as for integration and coordination. Other processes, such as strategic, program, and financial planning and resource mobilization (including taxation, debt management and foreign aid coordination) are understood as supporting the budget process. 153. The substance of Government's budgets in recent years is summarized in Chapter III;!/ while the impact of personal emoluments on the budget is emphasized in Chapter IV. Therefore, the focus of this Chapter is on the financial management system. A. Introduction 154. An effective financial management system has four broad purposes: (i) implementing expenditure controls to ensure that public resources are expended only in amounts and on purposes legitimately approved; (ii) promoting efficiency in resource use and ensuring that, for a particular program objective, minimum inputs are expended; (iii) promoting the effectiveness of governmental expenditures by ensuring optimum allocation of resources among programs and appropriately linking expenditures and objectives; and (iv) balancing governmental resources and expenditures over the long-term so as to ensure financial health and credit worthiness. 155. A complete financial system requires that all four purposes are met. The strategy outlined at the end of this Chapter is designed to produce a complete financial management system capable of accomplishing all four purposes in the medium-term. However, that strategy stresses the need for a phased approach, with initial emphasis on expenditure control. Therefore, a discussion of the institutional environment and organizational structure for planning and financial management, financial planning and other related issues is relegated to Volume II, Annexes II - VI. 156. Suffice it to say here that Government has attempted to implement most of the recommendations pr~ted in a 1983 World Bank report. 'l:l However, under pressure to respond to crisis situations, several elements of the recommended system have not been maintained. The major recommendations of that study and the current status of Government's actions in that regard are summariz.ed in Figure V.1. B. Issues 157. The fundamental recommendation in the 1983 Bank Report was that Government should establish a Program Planning and Budgeting System <PPBS}. Government did attempt to implement that recommendation, but subsequently abandoned it. The failure to sustain an integrated PPBS approach was due to: 1/ For more detailed discussion, refer to Zambia: Public Investment Program 1991 - 1994 (Lusaka: Government of Zambia, 1990) and Zambia: Public Expenditure Review (Washington, D.C.: The World Bank, 1992 [forthcoming]). 2/ Planning and Budgeting in Zambia (Washington, D.C.: The World Bank, 1983), 48 49 (a) The absence of an integrated MoF financial management system flowing from planning to auditing; (b) Lack of clear distinctions among three different types of planning functions 'J/ and between such functions and Budgeting; (c) Inappropriate assignment of responsibilities for such functions among NCDP, MoF, sectoral ministries, provincial administrations, and district councils; (d) Inadequate capacity to perform any of the various types of planning within the public sector, among all organizations involved and at all levels of Government; (e) The absence of medium-term financial planning; (t) The lack of realism in prepared budgets, ~/ particularly as regards actual expenditure intentions, resulting from weak costing and estimating techniques and the absence of actual prior year expenditure data during budget preparation and in the budget format; (g) Absence of effective budget ceilings, lack of basic and effective expenditure control, and non- adherence to audit findings; and (h) Insufficient numbers of qualified staff. 1S8. The fundamental issue is the limited capacity of the budgeting and financial management system to support the full range of four purposes identified above (para. 1S4). That lack of ~ity is a performance issue with reference to inadequate expenditure control; however, it is a systems design problem with reference to the three other purposes of efficiency, effectiveness, and financial viability and credit worthiness. Identifying the problem in that way, however, does not lead to an ability to solve it expeditiously. The deficiencies outlined above continue to plague the planning and budgeting system and it is unreasonable to expect that they will disappear, even with concerted effort, within the short-term future. Therefore, it is recommended here that establishing a PPBS system, while theoretically appealing, should be drop_ped as the focus for planning and budgeting system reform. Rather, such reform efforts should be targeted, at least for the short and medium-term future, on improving the budgeting and expenditure control functions. 1S9. Therefore, a discussion of the issues related primarily to the planning function ([b] - [d] above) are relegated to Volume II, Annex I. Of the remaining deficiencies, inadequate expenditure control is the most important. However, before discussing expenditure control in some detail, a summary discussion of the remaining deficiencies is provided immediately below. 'J/ The three types of planning are: (i) strategic macro-indicative planning; (ii) sectoral program planning; and (ill) project investment planning. Further, it is aJso important to consider various other activities required for the performance of each of those three types of planning: (i) data collection; (ii) data dissemination; (ill) determination of financial resources avai:able; (iv) allocation and budgeting of such resources in accordance with priorities; (v) preparation and dissemination of planning and budgeting guidelines to implementing agencies; and (vi) monitoring and evaluation of the extent to which macroeconomic policies have been achieved and the extent to which adherence to guidelines has occurred. ~/ A case in point is the PY1990 financing of secondary schools: i.e., the original budgetary provision was k30 million; during the course of the year an additional k2S million was provided as a Supplementary allocation; yet a total of k192 million was actually spent. 50 lntearation 160. As described in this Section, the organimtion of Government's current budgeting and financial management system is very complex. However, ironically, that organi:r.ational complexity is applied to the performance of a very limited set of functions. Even within that limited scope, an integrated system does not exist. As illustrated in Figure V.2, central government responsibilities for the major financial management functions are divided among several organizational units. That complexity is compounded by the roles of subnational levels of government (see Chapter VI). 161. For the system to operate effectively, the activities of the various organimtions involved in the process should be linked in a mutually reinforcing manner. The 1983 World Bank study 5,1 provided a comprehensive set of recommendations for the establishment of such a system. However, for the reasons described above (para. 157), the attempt to establish such a system has collapsed. More importantly, Government's attempt to implement the recommendations was premature, and remains premature, because of systemic weaknesses in the procedures for, and performance of, the expenditure control function (para. 158). 162. Staff responsible for performing essential functions within the system appear to be in a mild state of confusion. Such staff do not know what system they are operating. Thus, for example, Budget Office analysts do not know if they are supposed to empbasiu effectiveness, efficiency, or control and budget officers in spending agencies do not know if they are engaged in program budgeting, incremental budgeting, or something else. Participants may know what is done now, but they are uncertain as to what the purpose is and how they are expected to behave. Given that there is little common understanding as to what the current system is, it is not surprising that communication and links among the various organimtions are poor. For example, the MoF Budget Office, which is, in many respects, the center of the system acknowledges that two-way communications between it and four other units on which its performance depends are weak. Based on extensive interviews, conducted for the preparation of this report, the problem of inadequate inter-organimtional communications is not limited to MoF, but is common throughout Government. 163. An important consequence of attempts to establish a comprehensive system against the backdrop of poor inter-organimtional communications has been to divert attention away from establishing effective procedures for the performance of the most important function; i.e., expenditure control. Another consequence is that, in an environment characterized by over-committed managers and staff without systematic work programs, reliance is placed on crisis management. Thus, decision-makers defer those actions which they can, the decisions they take are ill-considered in the sense that they are not based on careful reflection, and, thus, decisions made are often not adhered to in practice. Therefore, priority objectives, which are sometimes established based on careful consideration, most often cannot be achieved. Thus, the attempt to be comprehensive in an environment characteriz.ed by limited capacity highlights the importance of prioritizing and phasing the process for improving system effectiveness. Bud&et Accuracy 164. The discussion presented in Volume Il, Annex V clearly demonstrates the inaccuracy of both revenue and expenditure estimates. Leaving aside the current psychology that accurate estimates are of only marginal operational value, another major problem is the inadequacy of required data for budgeting purposes. There is a shortage of timely and accurate data for both professional and management use. For example, many Government economists believe that production of professional macroeconomic analyses is hampered by the absence of valid data and the Tax Commissioner argues that he cannot set revenue targets because he does not have an adequate data base. Most importantly, perhaps, the Budget Office in MoF cites lack of valid data provided by reporting agencies as their biggest problem in the preparation of estimates and the monitoring of both revenue ~/ Planning and Budgeting in Zambia (Washington, D.C.: The World Bank, 1983). 51 Figure V.2 ORCNIZATION AND FUNCTl~S CHART: MAJOR FINANCIAL MANAGEMENT FUNCTIONS; CENTR~l._RESPONSIBILITIES Netlonel COMl11ioner llinl1tr1 Pen,ion Auditor for lllni1ter of Boerd Cenerel Developaent Pl1nnlng _______ _ Finance , Senior Pel'tlanent Senior Peraanent • Pen1ion • Finencia1 Secreter, Accounting Audit.ing • Computer Auditing Cen1u1 and l'tetl1tlc1 Adntlnl1t.rat.fon Office Director of Director of Pel'tlanent Control le, Senior lnveatllent EconOlllc Secretary of Under1ecretary Planning end Cooperetl on for Budget. Account.• for Re1Hrch Account, • Debt ll1nageaent • Budget Preparation • Recording • Internal Receipt, • Budget llanaa-nt. • Controll lng Auditing Accounting, • Reporting • Plyrol I Servicing Account.ing • Foreign Aid, Coordination, Donor Llel1on, Negotiation I c-l••loner Dat ■ Controller of of Proc111lng Cu1toa1 end Ta11e1 Manager E11cl1e • Col lectlng • Processing • Collecting • Accounting • Reporting • Accounting • Enforc-nt Enforcement. 52 collection and expenditure. Clearly, as a Headquarters staff agency, the Budget Office can only perform as well as the performance of those agencies on which it depends. 165. Thus, improving the effectiveness of the budgeting and accounting function will require significant activity in the line ministries, provincial administrations, and district councils. Improving the internal efficiency and effectiveness of the Budget Office alone can be expected to have only marginal effect in the absence of improvement of estimating and expenditure monitoring capacity in the operating agencies. Figure V.3 illustrates the relationship between MoF and other agencies for the performance of the budgeting and accounting function. 166. Staffing for performance of the budgeting and financial management function is inadequate. That situation is common knowledge and, therefore, that assessment in this report should come as no surprise. The more important point to be made, however, is that the inadequacy is not one of total numbers or the professional staffing pattern authorized in the Establishment Register. Rather, it is primarily a result of: (i) having insufficient staff with appropriate qualifications (for a discussion of incentives and their impact on recruitment, see Chapter IV); (ii) assigning staff without sufficient attention to the match between qualifications and responsibilities; (iii) the absence of on-the-job guidance to staff concerning their responsibilities in the context of overall system requirements; and (iv) lack of continuity in job assignments so as to benefit from accumulated experience. 167. Determining the adequacy of "capacity" and how such capacity can be transformed into adequate performance is a complex undertaking. §./ Nevertheless, even a cursory review of current capacity for the performance of planning and financial management functions within the Zambian Government must conclude that it is clearly inadequate. 168. To some extent, lack of apparent capacity is due to the demoralization of staff performing planning and financial functions. This is a problem throughout non-parastatal Government service and, therefore, is not unique to such staff. Leaving aside the issue of staff motivation, performance of the planning and financial management functions is severely constrained by: (i) inappropriate assignment of responsibilities; (ii) inadequate operating budget allocations; and (iii) inadequate staffing. 169. One obvious result is that competent staff are assigned heavy workloads characterized by discontinuous ad hoc assignments so that sufficient time is not available to perform basic responsibilities. At the same time, a corollary of such workloads is that many such staff do, indeed, perform those heavy workloads without any apparent rewards commensurate with such exceptional service. A determination of what motivates such staff might provide useful lessons applicable to the public service as a whole. 170. In any event, Table V.1 provides data on the number of authorized and actual professional staff assigned to budgeting and financial management functions in October 1990. Expenditure Control: Backw:ound 171. The absence of effective expenditure control is ironic precisely because Government's current budgeting system is designed essentially for control rather than to achieving the policy objectives of the expenditure program. The current budget format is organized by each organii.ation and specific categories of expenditure. That should provide a sufficient basis for controlling actual expenditures against specified allocations and an effective framework for accounting and auditing. However, in practice, that promise bas not been fulfilled. ~/ A discussion of the various elements which contribute to overall capacity is provided in Volume III, Annex VII. 53 Figure V.3 BUDGET PREPARATION PRQCF.SS ,, rontrol Al Approved Netional Ca1111ltt.ff I .,__- - - - - -.. •Budge t Assembly .. ~ • • Budget Speech and Draft Approved Budget CRZ Bucftet Budget • • • lliniatr y of .. Finance and National Comi•alon for i.. i-. Oevelopinent Planning Budget . Request . • Bucftet Cuidance Budget Request ... J. Ministries • Prov incl al1- ----- ----- ----- -- Counc11a 54 172. Although any budget is a working document and variations are to be expected over time, adherence to the general budget outline indicates the accuracy of the budget process itself, as well as the degree of control exercised over expenditure. Actual expenditure has consistently exceeded budgeted expenditure by an average of 9 percent during 1986-89. 173. The categories of expenditure which exceeded their budget by the widest margins were: (i) Recurrent Departmental Charges (RDCs); (ii) Grants and Payments; (iii) Pensions and Subsidies; and (iv) Personal Emoluments. 174. Recurrent Departmental Charges (RDCs). A review of initial budgets in recent years clearly suggest that: (i) estimates for RDCs were clearly inadequate; (ii) such inadequacies were sufficiently obvious as a percentage of total expenditure; (iii) the need for supplementary appropriations should have been anticipated; and, therefore, (iv) such supplementary appropriations were probably expected and relied on from the outset. 175. Subsidies. Subsidies, mostly for maize r.nd fertilizer, exceeded budget estimates, in part, due to: (i) variations in crop output during 1986-89; (ii) the failure of the marketing system; and (iii) the unforeseen need to import mai7.e and fertilizer in some years. 176. Personal Emoluments. Over-expenditure on personal emoluments is one of the causes underlying successive supplementary budget allocations during the 1985-89 period. Further questions about the accuracy of budget estimates are whether the inaccuracy of budget estimates occur in all economic sectors or whether it applies to some government agencies and ministries with particular force. According to Government's 1988 Financial Statement, supplementary allocations to personnel expenditure accounted for nearly ten percent of actual spending. Virtually all of these extra-budgetary allocations (18 percent of actual expenditure) occurred in non-economic ministries, including various judicial bodies of Government and the Ministry of Defence. The need for better control and accuracy of budget estimates is suggested by the fact that personal emoluments equalled 12 percent over initial budget estimates during the period 1986 - 1989; providing evidence to the excessive growth of public service employment during that period. 11 - 177. Capital. Constitutional, and Statutory Expenditure. These various expenditures (including loans and investments) and constitutional and statutory expenditures appear to be in line with initial budget allocations during the period 1986-1989. However, for capital expenditure, this is largely the effect of averaging overspending in some years and underspending in other years. Constitutional and statutory expenditures include debt service; accumulation of debt service arrears during 1986-89 helped to reduce expenditure to the level of the aggregate budget estimate. 11 Growth of public service employment is discussed in more detail in Chapter IV and Volume III, Annex VII. 55 178. Supplemental Authoriytions. The principal factor contributing to lack of expenditure discipline is the routine authorization of supplemental budget authorizations. Over the years, authorization of one or two supplemental budgets each year has become common practice. AB a consequence, budgeting and accounting personnel appear to place little emphasis on producing realistic expenditure (or revenue) estimates or upon preventing over-expenditure. The estimates are officially designated "Estimate" instead of some alternative term implying limitation. Currently, officials spend whatever is "needed," and then request either a supplemental budget provision or Treasury Authority to transfer funds from surplus to deficit accounts. That approach has been exacerbated by high inflation, so that budgeting and accounting officials now appear to believe that realistic management of estimating and budgetary control is not possible. 179. Based on the discussion in this Section, it is reasonable to conclude that inadequate performance of the expenditure control function is primarily the result of attitudes and behaviors, rather than system design. Recommendations in that regard will be a key focus of Section C. Prior to that discussion, however, this report turns to a review of accounting and auditing systems as key mechanisms for expenditure control. Accountin& and Auditin& 180. There are four primary mechanisms for enforcing expenditure control: (i) disbursing funds only within approved budget allocations, with subsequent supplemental authorizations provided only on an exceptional, and well justified, basis; (ii) accounting, within each spending agency, for actual disbursements, on a rolling basis, against specific expenditure items in the authoriz.ed budget; (iii) external auditing of accounts, at appropriate intervals, to determine if actual expenditures were consistent with authoriz.ed purposes; and (iv) sanctions for misappropriation of funds, as determined by such audits. Thus, effective accounting and auditing are fundamental instruments for maintaining expenditure control. 181. Accounting. The accounting functions of recording, controlling, and reporting Government revenues and expenditure are centered in the Controller's Office, MOF. That office is responsible for directing and controlling the entire accounting system. All accounting reports flow through the Controller's Office (including accounting for personal emoluments); such reports are then consolidated in a single Controller's report. Thus, the MoF Controller is responsible for supervising all Government accounts, based on formally established regulations, and for monitoring the flow of data on revenue and expenditure transactions to ensure that processing of financial data is timely and accurate. Figure V.4 illustrates the accounting process. 182. Actual accounting functions at the operational level are performed in the line ministries, provincial administrations, district councils, and other revenue collecting and spending agencies. Two significant exceptions to that rule are MoF's responsibility for payroll and pension accounting and NCDP's Loans and Investment Unit's responsibility for debt, including foreign aid, accounting. Nevertheless, all of these first line accounting agencies report such information to the MoF Controller. 183. Organization. Recently, the Controller's office was reorganiz.ed into seven sector offices, each specializing in specific functions ~ . refund of revenue) and/or specific agencies <Y:,, Copperbelt Province, Ministry of Commerce and Industry). These sector offices receive accounting reports from spending agencies, review them for timeliness, accuracy, proper coding, and adherence to regulations and prepare them for transmittal to the MoF Data Processing Unit (DPU); where the consolidated reports are prepared. This reorganimtion requires that accounting staff in the Controller's Office specializ.e. Such specialization is expected to improve the quality of their work. 184. Reorganimtion was required by Government's decentralization policy (see Chapter VI). An important element of decentralimtion policy is that Provincial Permanent Secretaries have become Financial Control Officers, with responsibility for maintaining consolidated provincial budget accounts (including accornts of all .Jistrict councils within their jurisdiction) and enforcing expenditure controls. Previous to the establishment of the decentralization program, financial reports from subnational levels were processed through the relevant sectoral 56 Figure V.4 THE ACCOUNTING PROCESS Con.o I i dated Accounts Periodi c Reports National Budget, Data Aa. .bly I I I CRZ Office ProceH1ng Budget. Unit. Budget Approved Account• Poating • Central Accoun ized ting..,____ ___ ___ ___ ___ ____ Unit. Interna l Audit Sect.Ion Flnanclal Regulat.lona Posted Guidance Account• Poat.ct Interna l Account• Budget. Audit.Ing Account.a .includ e•: • Mlni•t rl" Operating • Provlnclal Council• Unit.•• Operating • Pen•lon Board Unit•• • Payroll Accounting • Loan• and Inve•tlllenta Accounting 57 ministries. However, under the current system, such reports are consolidated at provincial level and are submitted by Provincial Permanent Secretaries directly to the MoF Controller's Office. 185. System. Provincial administrations, ministries, and other spending agencies are each responsible for maintaining revenue and expenditure ledgers; including commitment ledgers. This is the first point of control and provides a mechanism for rejecting transactions which would over-commit an account. When there is pressure for over-commitment, the system allows transfers of resources from accounts with potential surplus to accounts with potential deficits. Transfers within a major spending category may be approved by the responsible Financial Control Officer. Transfers between major expenditure accounts require the approval of the Budget Office, MoF. The role of provincial administrations and district councils is discussed more fully in Chapter VI. 186. The new organizational structure of the Controller's office responds to the new reporting system described above by creating specialized units to receive and process those reports. The flow of accounting data from the field directly to MoF, rather than through the sectoral ministries, is expected to accelerate the timeliness of the reporting process. Perhaps more important, the designation of Provincial Permanent Secretaries as Financial Control Officers exposes a new level of management to review and control. As a result, budget management and expenditure control might be improved once the new system is effectively implemented. 187. Commitment Ledgers. In the context described above, better maintenance of commitment ledgers is being emphasized. That emphasis is consistent with the 1983 World Bank report's recommendation that the practice of keeping commitment registers should be reinstated and monthly commitment reports should be submitted to MoF before funds are released. Unfortunately, to date, commitment ledgers are no yet maintained on an accurate and timely basis. 188. Accounting Formats. The accounting system is defined by a published set of Financial Regulations. The system operates as a double entry, cash system. It includes commitment accounting and reports revenue and expenditure transactions. A significant omission is that the system does not report cost data or performance data. With regard to accounting for loans to parastatals, NCDP's Loans and Investments Unit (LIU) follows commercial principles and, for its reports to the MoF Controller, reformats its data according to the general Government accounting format. 189. Auditig. Government's auditing system includes both internal and external audit functions. Internal auditing is the responsibility of a Unit within MoF's Office of the Senior Undersecretary of Accounts. That unit operates separately from the Controller's Office and provides a mechanism for maintaining an independent check upon operations of the accounting system within MoF. The external audit system is the responsibility of the Auditor General, a further check on accounting operations on a system-wide basis. 190. The current auditing system focusses on the limited objective of ensuring that spending agencies expend funds only in the amounts and for purposes authorized specifically in the budget and account for such expenditures as required by the Financial Regulations. That is an important function, as illustrated by the fact that the Report of the Auditor-General On The Accounts for the Financial Year ended 31st December. 1988 points out that kl78.8 million (2.14% of actual expenditure) for 1987 and k76 million (1.3 % of actual expenditure) for 1988 were unconstitutional. Another K2S0.6 million (about 3 % of total spending in 1988) was •contrary to the provisions of the Financial Regulations ...• • Thus, as currently performed, the Auditor-General does attempt to discover, and often does discover and report, specific financial irregularities. 191. Nevertheless, although a system is in place for financial compliance auditing, there are three major inadequacies in the design and operation of the auditing function: (i) weak financial discipline within the public service due to inadequate compensation (Chapter IV), overcommitted annual budgets (Volume II, Annex V), and access to supplementary budget authori7.ations (para. 178); (ii) lack of importance given to the Auditor-General's report by the legislature (see also paras. 208 and 209); and (iii) absence of performance auditing. Procedures for acting on the results of either internal or external audit reports are most often ignored. As only one example, of 58 301 recent internal audit reports, satisfactory responses were received from only 20 of the offending agencies (6.6%). There are two Government agencies assigned responsibility for pursuing corrupt Government employees: (i) the Anti-Corruption Commission, which is responsible for investigating and prosecuting minor infractions and (ii) the Special Investigation Team on Economics and Trade (SITET), which is responsible for major cases. The operations of these two agencies, however, are independent of the auditing functions of the Senior Undersecretary of Accounts and the Auditor General's Office. As regards performance auditing, there are simply no established criteria, indicators, nor procedures for assessing direct links between expenditures and the specific objectives to which they have been committed. 192. As designed, accounting and auditing systems are adequate for expenditure control and management (the first purpose of a financial management system); including sufficient provision for flexibility. Even in those limited terms, however, performance has been inadequate. Thus, as in the case of overall expenditure control described above, inadequate performance of accounting and auditing functions is primarily the result of inadequate skills and inappropriate attitudes and behaviors, rather than deficient system design. 193. Two efforts are currently underway to improve performance of the accounting function. In addition to the reorgani:zation of the Controller's Office within MoF, computers are being installed and ODA is funding a two-year training program to improve accounting throughout the GRZ. These efforts should be allowed to take effect, although MOF should monitor the ODA program to ensure that it is supportive of the long-term strategy presented in Chapter VII. 194. In any event, without cost and performance data, the accounting and auditing system is not designed to support the three purposes of financial management beyond expenditure control; i.e., the system cannot: (i) promote efficiency in resource use nor ensure that, for a particular program objective, minimum inputs are expended; (ii) promote the effectiveness of governmental expenditures by ensuring optimum allocation of resources among programs nor link expenditures directly to objectives; nor (iii) balance resources and expenditures over the long-term so as to ensure financial health and credit worthiness. c. Recommendations 195. Although a complete planning, budgeting, and financial management system requires that all four economic management purposes be accomplished, an improvement strategy requires priorities to be set and improvements phased. This is true for at least five reasons: (ii) the GRZ does not have the management or resource capability to mount a complete, simultaneous systems improvement effort across all four purposes concurrently; (ii) staffing requirements to accomplish improvement objectives require increasing levels of professional sophistication from establishing controls through promoting efficiency to promoting effectiveness of expenditure; (iii) current performance levels suggest that the GRZ does not yet have the staff and/or resources to operate a complete system, particularly at the higher levels, and that these resources will have to be developed over time; (iv) improved budget preparation and related expenditure controls is a prerequisite to achievement of all three of the other purposes; and (v) the brief review of current performance demonstrates clearly that such controls are not yet effective. Therefore, although the long-term objective is to produce a comprehensive financial system, the recommendations which follow are based on the awareness that the ERP requires effective expenditure control if it is to succeed, that such control does not yet exist, and that an urgent effort is required to achieve it. 196. Although linking budgeting to development planning should be an ultimate objective of Government over the longer term, the short-term priority should be improved budget formulation and expenditure control. The objective should be to ensure that financial policy-making and implementation is conducted in a transparent and explicit manner, rather than the accidental result of many, uncoordinated actions by operating agencies (as is now largely the case). Budgetary and accounting systems should be modified so that financial policy, as expressed in the budget, is proposed by MoF through established procedures and, once established, that compliance with such policies becomes, in actual practice, the norm. I 59 197. More specifically. Government should consider measures to improve the realism of proposed budgets, particularly revenue and cost estimates, and strengthen management of budget execution. Further measures to develop resource mobilization need also to be introduced. The various technical assistance efforts now underway (and planned) should contribute to those undertakings; although integration of such efforts will need to be carefully managed by Government. 198. Thus, the Government should: (a) Adopt an explicit Institutional Development Strategy for creation of an integrated strategic planning and financial management system realistically phased over the long-term; as follows: (i) Short-Tenn <1992 - 1994) - work on basic problems first, giving priority to operational improvements in budget preparation and expenditure control ~ . improve budget guidance) and initiate design of improved systems for revenue generation; (ii) Medium-Term {1995 - 1997) - (a) implement improved systems for mobili1.ation of resources; (b) expand budget preparation and expenditure control improvements into an overall, documented system; and (c) institutionali7.e the new system; (iii) Longer Term (1998 and Beyond) - Improve the link between planning and budgeting and institutionali7.e an integrated and complete system. (b) Consistent with (a) above, it is recommended that those specific implementation actions required by the PFP during the short-term should be distinguished from those actions required for the longer term improvement of the institutional capacity of key economic management organi1.ations so that confusion about each can be avoided. Short-Term {1992 - 1994); Budeet Preparation and Expenditure Control 199. In order to achieve effective expenditure control as the first priority objective, supporting actions will be necessary in each of the following areas: (i) revenue estimating; (ii) budget guidance; (iii) budget preparation; (iv) budget formatting and presentation; (v) operation of the Budget Office, MoF; (vi) expenditure monitoring; and (vii) sanctions for financial malfeasance. Thus, establishing budget preparation and expenditure control as the priority short-term objective requires improved performance in areas broader than those two functions themselves. Each of these supporting areas are discussed briefly below (paras. 200 - 209). 200. Revenue E,timating. A range of estimates should be produced with an optimistic, pessimistic, and realistic set of assumptions; ~ the inflation rate. Government should determine alternative policies for such factors as debt, based on each set of assumptions, and establish specific indicators which would trigger the pursuit of one alternative or another. Government would need to monitor performance of the economy and, as events unfold, adopt the proper policy. The results of such monitoring would then be used to provide a proper basis for preparation and consideration of supplemental budget requests. ii 201. Bud&et Guidance- In addition to those assumptions currently underlying budget guidance to spending agencies, the inflation rate assumptions should also be included. In addition, budget guidelines should ii Notwithstanding the proposal here, however, that the basis for revenue estimates ought to include an optimistic assumption (among othen), actual expansion of expenditures derived from such increased revenue, if it were to occur, should be limited according to established policies concerning the proper role of the public sector in, and the impact of public sector spending on, the economy. Thus, the ability to generate revenue does not necessarily mean that such revenue should be collected and/or spent under such conditions. 60 explicitly permit Provincial Permanent Secretaries to reallocate provincial budget items, originally agreed with MoF, across organizational boundaries ~ . from one Department to another Department) or from clearly specified non- priority sub-sectors to priority sub-sectors, but not from one sector to another. 'J.I Further, budget guidelines should require that proposals for capital projects be expressed in terms of aggregate financing required, not only foreign aid to be received, and require detailed recurrent cost estimates for each project nearing completion. 202. Finally, the implications of the staff retrenchment program (see Chapter IV), expressed in terms of financial and staffing targets, should also be clearly articulated for each ministry, provincial administration, district council, and other spending agencies. Budget submissions should be required to respond explicitly to retrenchment targets provided in the guidelines by including a list of all: (i) "ghosts;" (ii) persons retiring under normal conditions; (iii) persons accepting or being proposed for early retirement; and (iv) finally, those proposed for dismissal (and the reasons for dismissal in each case). Budget applications should also include an explicit statement about the impact of such retrenchment on the agencies mission and operations. By focussing on targets and criteria for decision-making, the retrenchment program can rely primarily on individual staff to accept voluntary departure and on operational supervisors to identify appropriate candidates in those agencies where such voluntary departures and normal retirements are not sufficient. Such an approach also largely obviates the need for a comprehensive census of all employees in all agencies prior to action being taken and requires only limited involvement of centralized decision-makers in the individual cases; except to monitor compliance with established criteria and process exit procedures. 203. Budftt Preparation. The emphasis here should be on the accuracy of expenditure estimates. For recurrent expenditures, incremental (or decremental) budgeting, based on prior year actual expenditures, should be explicitly promoted. Written statements justifying any proposed deviation from such extrapolations should be required. For annual budgeting purposes, capital budget proposals should include base cost estimates, refined by planning estimates, expressed in the context of a multi-year sector program. Perhaps most importantly, as a clear indication of substantive financial planning rather than routine financial extrapolations, operational managers within spending agencies should be held ultimately responsible for the preparation of proposed budgets; with the assistance of Budget Officers and accounting staff. Further, all budget submissions should be signed by the specific line manager(s) who will be responsible for the actual operation financed by specifically proposed budget items. "Line managers" should be understood in this context as department and unit Heads; not Permanent Secretaries. 204. Budget Docmnent. The Budget Document (i.e., the "Yellow Book") should include a column indicating actual expenditure for the previous year. The title of the last column should be changed from "Estimate" to "Ceiling" and the entries in that column should be treated as, in fact, ceilings. 10/ Although the organiution/objects class format should be retained as the most appropriate structure for control, it should be expanded to display organizational subdivisions at the Departmental, and perhaps Unit, level. 205. Budget Office. A specialized unit should be establish within the Budget Office, MoF, responsible solely for reforming, and managing, the budget system. Another specialized Unit should be responsible for preparing the annual Budget document. In addition, the Budget Office should amend, and then enforce, the budget cycle schedule; allowing sufficient time for planning, preparation, and review. Budget analysts should be organized into two-person teams assigned responsibility for specific groups of departments. However, supplemental responsibilities for object class reviews across departments should also be assigned to teams. Implementation of these recommendations would require a substantial increase in Budget Office staffing; but that should have a neutral 2/ This recommendation cannot be implemented fully with regard to donor financed projects without the legal agreement of such fmanciers. Thus, such agreement should be sought whenever practicable. 10/ Such a change is more than merely cosmetic in the context of the medium-tenn expenditure framework provided by successive Policy Framework Papers (PFPs) and Public Expenditure Reviews (PERs). · 61 effect on totaL staffing costs by retrenching low-level non-professional staff in the Department of Taxation and the Controller's Office. 206. ;Euenditure Monitorina and Enforcement. The preparation and distribution of monthly accounting reports is a key element for effective expenditure monitorin1. The capacity to prepare and issue such reports on a timely basis is not currently in place. The Data Processing Unit in MoF is currently three months behind in processing such accounting reports and, in any event, such reports are not widely distributed to .:elevant managers. As a first step, pending improvement in data processing and issuance of monthly accounting reports, emphasis should be placed on issuing the monthly Revenue and Expenditure Returns Report (prepared by the Budget Office). In that regard, the Report format should be modified so as to permit identification and easy reporting of likely over-expenditures to relevant managers. 207. Immediate emphasis should also be placed on redesigning the current format of the Commitment Ledger prepared by MoF's Controller and used by all accounting officers throughout Government. The current format is not designed to prevent over-expenditure. Rather, it is designed to record the justification for supplemental allocations to cover over-expenditure. Therefore, the format of the Commitment Ledger should be redesigned so as to emphasize avoidance of over-commitment and encourage compliance with initial budget allocations as ceilings. 208. In any event, the responsibilities of the MoF Controller, supported by the Senior Permanent Secretary, should be buttressed by the creation of a Public Accounts Committee in Parliament to whom reports should be submitted by the Auditor-General. In addition, the Public Accounts Committee should initiate its own scrutiny through the holding of public hearings focussed on both general patterns of abuse identified and specific instances of malfeasance as revealed in such reports. 209. Enforcement. Once such instruments are in place, the MoF Controller, supported by the Senior Permanent Secretary, should require responses from those agencies committing over-expenditure or misallocating resources which state what corrective action they intend to take and then monitor implementation of those actions. Agencies which do not report or respond to appropriate inquiries should be sanctioned by reductions in budgetary allocations proportional to the expenditures in question. Specific instances of malfeasance should be prosecuted vigorously in the law courts. Finally, public exposure of questionable behavior through hearings conducted by the proposed Parliamentary Committee on Public Accounts can serve as a deterrent to malpractice; although appropriate safeguards for the protection of managers accused of such malpractice would also be necessary. Revenue Generation 210. Until recently, the mobilization of domestic financial resources has not received sufficient attention. Those responsible for revenue policy and collection have been operating on a routine basis. Thus, as described above, the tax system has evolved incrementally and on an ad hoc basis. Therefore, reform of the tax system is an urgent reguirement. Nevertheless, notwithstanding the urgency of increasing Government revenue, it cannot realistically be accomplished without a carefully laid foundation. Therefore, during the initial short-term period, a comprehensive study should be conducted, at the earliest possible date, indicating: (i) alternative areas of possible taxation; (ii) areas which could most easily yield the highest returns in the shortest period of time; (iii) the structure of such taxes; and (iv) the design requirements for establishing appropriate systems (i.e., organizational structures, procedures, and capacity) for the actual generation of such revenue. Any such study should evaluate proposals for tax reform against criteria of: (i) equity; (ii) yield/adequacy; (iii) support for economic development priorities; and (iv) administrative ease. Medium-Term and Beyond 211. In Section B above, eight problems were identified as the most important affecting the performance of the planning, budgeting, and financial management functions. Of these, three elements have been addressed as 62 part of the recommended short-term program: (i) budget preparation; (ii) expenditure control; and (ii) resource mobilization. Over the longer term, it will be necessary to address broader system-wide requirements so as to ultimately accomplish all four of the economic management purposes identified in Section A. The most important among these longer term requirements relate to improved implementation of revenue generation efforts (based on the study recommended above) and improved performance of key planning functions. Thus, while immediate short- term requirements for planning performance can be met through technical assistance and direct donor support for preparation of PIPs and PERs, Government shouM have in view over the longer term the establishment of an effective and sustainable planning system • 212. Plannina Function.,. The fundamental requirement is that the planning system is rationalized so as to provide for the appropriate assignment of planning functions to appropriate organizations within the public sector and that mutually reinforcing links are established among those various organimtions. To accomplish th.at objective, changes will be required in the way Government is currently structured for the performance of planning functions, including: (i) creation of a Policy Formulation and Monitoring Unit; (ii) limiting the role and transforming the structure of NCDP; (iii) expanding the role of sectoral ministries and other government agencies for sectoral program and investment planning; and (iv) establishing strategic and sector level strategies and programs to provide a context for prioritizing and integrating the management of donor assistance. 213. Policy Formulation and Monitoring Unit <PFMU). Consideration should be given to the creation of a PFMU to: (i) formulate alternative macroeconomic scenarios as a basis for economic policy decision-making; (ii) disseminate information about economic projections, to both the public and private sectors, based on such policy decisions; and (iii) monitor performance of the economy in terms of Government priorities. Such a unit could be located within the Office of the President, Office of the Prime Minister, the MoF, Q! the University of Zambia Q! a specialized semi-autonomous research institution could be established for that purpose. 214. Each of the alternative locations of such a Unit entail advantages and disadvantages. If established within the Office of the President, such a unit could be expected to have access to the most senior level of political decision-making; however, it might also result in artificially avoiding the requirements of political negotiation in the Cabinet. If located with.in the Prime Minister's Office (PMO), the link to political decision-making with.in the Cabinet and, by extension, Parliament would be more firmly established; however, that advantage might be at the cost of premature visibility of the unit's analyses. If located within MoF. the unit might be able to avoid premature political influence on its analyses, but the likely political reaction to such analyses might th.en be deferred to the point that much effort would be expended for irrelevant purposes. Finally, if a specialized, semi-autonomous institution for macroeconomic research and analysis were established at either the University of Zambia or independently, it should be able to offer more attractive terms and conditions of service than could the regular public seivice, but might also be too far removed from decision-makers for whom such analyses is intended. In any event, if established, such a Unit should also be responsible for establishing the priorities of the Central Statistics Office (CSO). 215. National Commission for Development Planning <NCDP). Serious consideration should be given to transforming NCDP from a large Commission responsible for comprehensive central planning functions to a streamlined technical Department of either Cabinet Office or MoF. with responsibilities limited primarily to macroeconomic indicative planning; including a focus on: (i) inter-sectoral priorities; (ii) medium-term financing requirements; (iii) issuance of strategic planning guidelines; and (iv) review and monitoring of sectoral program \'>lanning to ensure conformity with macroeconomic investment priorities. In restructuring NCDP in th.is manner, special attention should be given to the importance of providing cross-sectoral economic data to the private, as well as public, sectors for purposes of their own strategic planning requirements. Statistics concerning consumer demand trends, supply projections for public goods and services, and prospects for exports, as welJ as analyses based on such statistics, would be useful to both the private and public sectors. 216. The numerous publications currently produced under the current system should be replaced by a single document entitled something like Policies, Programs, and Investments for Economic Development <PPIED). 63 That document should be completed in June of each year so as to foster coherence in the presentation of a three-year rolling public investment program (PIP). The proposed PPIED (or its equivalent) should include an assessment or evaluation of the economic situation in Zambia during the previous year. prospects for the economy. economic policies undertaken and their impact on ivestment program policies, and specific priorities for the next three years. 217. A fundamental feature of such restructuring would be that the successor to NCDP would not be responsible for. nor participate in, the preparation of either the capital or expenditure elements of the mmYl!l Government budget process; except for, perhaps, reviewing budget proposals to ensure adherence to macroeconomic priorities. 218. Sectoral Ministries and Other Operational Units. Sectoral ministries, other Government agencies, district councils and, perhaps, provincial administrations should have primary responsibility for strategic and program planning within the sectors of their responsibilities. including: (i) medium-term financing requirements; (ii) issuance of strategic sectoral planning guidelines; (iii) identification of project level activities for transforming program objectives into action; and (iv) review and monitoring of project proposals and implementation to ensure conformity with program planning priorities. 219. Thus, planning units, modelled on PD/MOA, should be established in all sectoral ministries and at the district level. Existing planning units within those ministries currently having them should be ~tructured and their functions more clearly defined. These planning units should be closely linked to statistics and data collection sections at the sector, provincial and district levels. Further, a planning format which addresses the needs of particular ministries and agencies should be developed. The experience of the Ministry of Agriculture should be reviewed and adapted to the needs of other sector agencies. 220. Although achievement of those objectives will require a long-term effort, that does not mean that identifying the process through which they can be achieved should be deferred indefinitely. Rather, in order to achieve objectives in the long-term, actions must be initiated in the short-term. Therefore, Government should establish a Financial Systems Reform Working Group. representing the key participants in any reformed system, and assign it responsibility for recommending a detailed phased action-program for achievement of the objectives outlined above over a five-year period. That Working Group should report directly to the Minister of Finance, with the Senior Permanent Secretary of Finance serving as Secretary. Donor financed Technical Assistance (TA) should be considered to assist the Working Group (see Chapter VII). 64 VI. DECENTRALIZATION 221. Alternative forms of public sector decentralii.ation can have important differential effects on economic performance at both macro or sector levels. Although it can be argued that some forms of decentralimtion are, at least theoretically, necessary for the achievement of economic growth with equity, it should also be noted that decentralizing economic management functions can also result in maintaining many historically rooted public sector inefficiencies. 222. Government's current official policy is to progressively devolve 1/ a wide range of responsibilities, currently performed by central Government ministries, to district councils. The objectives of that policy are, essentially, to: (i) improve management efficiency; (ii) improve financial performance; and (iii) increase popular participation in governmental decision-making. To those ends, the Local Government Act (1980), as amended, provides the legal basis for district councils to exercise substantial discretion in the performance of a wide range of responsibilities, which they are expected to finance largely by themselves. Nevertheless, as a Unitary syctem, the SS district councils are creatures of the national government and, therefore, any discretion exercised by them is based on grants of authority by Central Government. 223. However, none of the three objectives specified above, with the possible partial exception of improved channels for popular expression of demands for Government services, have been, or are likely to be, achieved under current circumstances. Systemic inefficiencies in public sector performance have been dispersed through decentralization, rather than improved, and deficit financing of expenditure at the district level is having a significantly negative affect on Government's fiscal stabilization program. District councils in Zambia have historically: (i) established parastatals, monopoli7.ed productive sectors and/or distorted the terms of private sector participation and competition; (ii) engaged in deficit financing of their expenditures; (iii) received preferential treatment by financial institutions and borrowed a disproportionate share of locally available credit; (iv) employed excess staff with inappropriate qualifications; and (vi) attempted to implement their responsibilities through inappropriate institutional structures and inefficient procedures. 224. Thus, under current conditions in Zambia, changing the structure of the public sector through decentralization will not likely result in changes in macroeconomic policy behavior that could foster greater efficiency. Unless substantially new approaches are taken to the structure and objectives of Government's decentralization program, increased participation in public sector investment and service delivery decisions is likely to increase demand pressures on government resources; to say nothing of the likelihood that decentralization will increase, rather than decrease, aggregate public sector employment. 225. With those cautions in mind, this Chapter discusses the following: (i) current organizational structure; (ii) assignment of responsibilities for, and capacity to perform, the key provision functions of planning, budgeting, and financial management within that structure; and (iii) current employment policy and practice at subnational levels. The discussion of those factors is followed by a summary of major issues: (i) central government control; (ii) deficit financing; and (iii) economic policies. The Chapter concludes with specific recommendations for reform. 1/ Devolution involves the transfer of discretionary responsibilities for a range of sub-sectoral activity to subnational levels of government, such as a district council. For example, if district level education officers are responsible directly to a District Council, perfonnance of the education functions assigned to that officer have been devolved by the central government to the Council. If, on the other hand, the Officer is assigned to a subnational level, but remains responsible directly to the central government Ministry, authority has been deconcentrated (see Footnote 4). 65 A. Current System 226. Government is organii.ed in three tiers: (i) the national level Ci&,, central government); (ii) provincial administration; and (iii) district councils. 2/ The structure and performance of the key planning, budgeting, financial management, and public sector employment functions at the central government level has been the primary subject of the preceding Chapters. The focus here is on the second and third tiers of that structure. Structure 227. Provincial Administration. The nine provincial administrations have recently been created as subunits of the national government. As indicated in Chapter V, the Officer responsible for governmental activities at the provincial level is UNIP's Member of the Central Committee (MCC) assigned for that purpose in each province. That position is roughly equivalent to that of a Provincial Governor in other countries. To each Provincial MCC is assigned a Provincial Permanent Secretary from the regular Civil Service, who manages the various Department Heads assigned to the provincial level by each of the sector ministries. 228. Provincial administrations were created for the dual purpose of integrating (i) the activities of UNIP and Government at subnational levels and (ii) sectoral activities into a coordinated subnational regional program responsive to local requirements and conditions. With those objectives in mind, Provincial Permanent Secretaries: (i) finalize provincial budget requests, including the allocation of all sector specific budgetary requirements within the province; (ii) authorize budget transfers within major expenditure categories to respond to local needs and priorities (however, transfers among major categories require approval of the MOF Budget Office); and (iii) are responsible for the administration of personnel support systems and day-to-day management of all Government employees assigned to the provincial level (although such employees are supervised regarding technical matters by their respective sectoral ministries). Thus, the Provincial Permanent Secretary reports to the UNIP MCC for the province, 'J/ is responsible for preparing the provincial budget and annual operating plan, and directs and controls those resources allocated to the Province by the central government for the accomplishment of the provincial plan. 229. It should be noted that provincial administrations do not have legislative powers nor do they have any authority to raise their own revenue to support initiatives undertaken on their own behalf. Rather, provincial administrations are decoocentrated extensions of the Central Government. ~/ 230. District Councils. By contrast with provincial administrations, district councils are, in principle, independent local governments. Such councils are elected and, according to the 1980 Act (as amended), they have authority to: (i) prepare their own budgets; (ii) raise revenue from specified sources; (iii) enter into legally 'l,/ As indicated in Volume U, Annex I, the FNDP outlined a strategy and established objectives to revitalize and streamline the planning function from NCDP to sectoral ministries and provincial, district, and village planning units or committees. As suggested therein, it was contemplated that village administrations would also be officially established through some sort of integration of the Chieftaincy system with more formal governmental administration. However, no such action has yet been taken. 'J_/ The decision to move toward a multi-party political system has led to consideration by Government to separate UNIP and the Government at provincial and district levels again. ~/ Deconcentration involves the assignment of authority to sub-units within a single hierarchically structured organization; for example, assigning responsibility for certain decisions to district level Education Officers who nevertheless are responsible directly to the Ministry of Education at the central government level, rather than to a district council. 66 enforceable contracts; (iv) own and operate commercial enterprises; (v) borrow from lending institutions; (vi) legislate local ordinances; and (vii) determine what services they will provide. Council employees responsible for implementing the decisions of its elected members are not subject to the authority of the national government. 231. Nevertheless, in actual operational terms, the autonomy and discretion provided by the 1980 Act to district councils is more apparent than real. Thus, the exercise of discretion by district councils is inhibited both legally and operationally. Legal inhibitions derive from the provision in the Act that the exercise of any of the discretionary authority provided district councils is subject to the approval of the •Minister Responsible" (i.e., which, under current organizational arrangements within the GRZ, is the Prime Minister). As discussed further below (see para. 253), that exceptional authority to approve the actions of district councils, as delegated to the Decentralization Division of the Prime Minister's Office (PMO), has been exercised routinely; to the point that such councils exercise very little actual discretion. Functions 232. fl'JD!!iJl&. Figure VI.1 illustrates the link between national and subnational planning systems. A notable feature of the planning system is that provincial and district planning activities are linked to the central government planning function through separate channels. 233. Provincial Planning. At the provincial level, Provincial Planning Units (PPUs) are responsible for consolidating the plans developed by the various deconcentrated sectoral departments of the ministries into integrated provincial plans. Such plans, once prepared, are submitted to the NCDP. These PPUs are assigned responsibility for a wide range of tasks, including: (i) preparation of the provincial budget (recurrent and investment); (ii) discussions regarding district council budget proposals (see para. 242 below); (iii) monitoring project implementation within the province; (iv) collecting and analyzing data required for effective planning; and (v) coordinating donor financed activities at provincial level. Although such responsibilities are wide-ranging, staffing and operational support of PPUs is substantially inadequate. PPU staff lack technical skills and are provided only limited training opportunities. Office space and equipment are maintained at minimal levels. 234. In response to some of these inadeqW&Cies, various donor agencies have been providing technical assistance in support of PPUs. Figure VI.2 summariz.es such assistance, which excludes only Lusaka and Copperbelt provinces. GRZ officials believe those two provinces were excluded because donors were interested primarily in rural development with reference to the provincial level and viewed Lusaka and Copperbelt provinces as essentially urban; although NCDP and the provincial officials in those two provinces have argued that most of their constituents are rural. 235. In any event, although the PPUs in all but Lusaka and Copperbelt provinces appear to be reasonably well equipped as a result of donor assistance, the procurement and maintenance of such resources are highly dependent on the continued presence of expatriate technical assistance staff. Further, as Figure Vl.2 clearly shows, coverage of support for improved planning is limited to agriculture or livestock development in three of the provinces (42.8%). Finally, ODA is seriously reconsidering its support for development of planning capacity at the provincial level because its experience suggests that such assistance cannot be effective within the current institutional system. 236. Lusaka province provides a suggestive example of what might be expected of PPUs in the absence of foreign donor agency support. Thus Lusaka's PPU has only two of its eight authoriz.ed professional positions filled (25%); and even then, serious questions can be raised as to whether a staff of eight would be adequate. The two planning officers in place concentrate on budget preparation during the budget cycle. The remainder of their time is occupied with monitoring of urban service delivery and ad hoc resolution of implementation crises. Such staff have little, if any, time to concentrate on either program or project investment planning at the provincial level. 67 Figure VI.I: LINKING SUBNATIONAL AND NATIONAL PLANNING SYSfEMS Senior Pennanent Sec:retar1 Ministry of Finance NCDP lfOF NCDP Econ. Planning Budget Division Econ. Cooperation Cabinet Office Sectoral Oec:entrallzatlon MlnlatrlN Dhlalon Central Oovernaent Econ. a Technical Coordination Reglonal NCOP Dept. NCDP Dlatrlct I Council Sectoral Provlncial I Planning M!nlatrlN At f'rovlncea Plann1ng Units I Unl\a I Foreign ........................ I Funding Iii 68 237. Another aspect of inadequate PPU planning capacity is the ambiguity concerning the role of the provincial level in the planning process and the status of PPU staff. As pointed out in Volume II, Annex I, the FNDP's recommendation that an Act be legislated providing legal status for planning entities at national, provincial, district, and village levels has never been enacted. Thus, PPUs exist as an administrative creation of the central government at provincial levels. Although supposedly preparing integrated provincial plans, they, in fact, simply coordinate the largely discrete planning efforts of sectoral departments whose staff are seconded to the province by central government ministries. Such provincial plans are, in tum, submitted by the PPUs to the Regional Planning Department of NCDP. 238. District council plans are forwarded directly to the Decentraliution Department of the Prime Minister's Office (PMO) for review and approval. However, copies are, in most, but not all provinces, routinely sent to the respective PPU at the same time. Nevertheless, it is not clear to the PPU staff, or to relevant district council staff, whether or not they have any authority to approve or disapprove such plans. Discussions among district council representatives, PPU staff, and representatives of the Decentralization Division do occur at meetings at provincial headquarters; but, again, PPU staff view their role in such meetings as anomalous. 239. Further complicating the situation is the fact that PPU staff are, themselves, seconded to the provinces by NCDP. As is the case with all other provincial staff, they are supposedly responsible to their central government agency for technical support; to the Provincial UNIP/MCC for policy direction; and to the Provincial Permanent Secretary for managerial and administrative support. Such a system would be complicated enough if characteriz.ed by explicit and clear guidelines concerning the respective roles of these three sources of leadership and supervision. However, the current situation does not include such guidelines. PPU staff complain that they do not know for whom they really work and to whom they are actually responsible. In particular, they complain that, having assigned them to a province, NCDP no longer takes responsibility for their career development <£:&, promotion and training) and that such facilities are not available to the provincial administration. 240. District Planning. With the exception of five district councils in Luapula Province supported by FINNIDA technical assistance (Figure VI.2), there are no district council planning units. Efforts initiated in 1981 to establish such capacity in Central and Luapula provinces were abandoned in 1983 due to lack of adequate resources. The recommendation in the ~NDP that such units (or committees) be established at district and village levels have not been implemented. Thus, the administrative structure of district councils does not provide for any institutionalized planning capacity. The Financial Secretary is normally an accountant by profession and the Development Secretary, most often an engineer, is basically responsible for civil works. There are no personnel with technical skills in planning or project preparation employed by district councils specifically to perform planning functions. The result is that many district council investments, such as those in Ndola, Mufalira aod Kalulushi districts, are initiated without any effective economic and financial preparation and appraisal ,,r substantive operational planning. The result of such investments attest to that failure. 241. The exception to the above summary description is the five district councils assisted by FINNIDA in Luapula province. In those districts, planning units have been established. These five councils are linked directly to the Economic and Technical Coordination Department, NCDP for financing approved district plans. The other SO district councils have no such planning units and do not have any financial or technical links with NCDP. 242. Budgeting and Financial Management. All operations at the IW£vincial level are financed by the central government through the Ministry of Finance. The one innovation resulting from Government's 69 decentralmtion policy is that expenditures at the provincial level are not included in the various ministerial budgets, but are rather provided to the provinces directly. Thus, the Provincial Permanent Secretary. rather than the various ministerial Permanent Secretaries, serves as the Financial Controlling Officer for such funds. S/he is responsible for presenting the provincial budget proposal directly to the Ministry of Finance and is the primary official responsible for defending that proposal at budget negotiation sessions. Through that mechanism, some marginal discretionary authority is available to the provincial administration to initiate investment proposals and, later, to transfer authorized funds from one budget line item to another within sectors. 243. However, in actual practice, the establishment of provincial budgets has represented not much more than the integration of disaggregated sectoral ministry budgets, rather than an integrated provincial budget. That is due in large part to the reliance for sectoral budget proposals on the staff of technical departments at the provincial level who are largely responsive to direction from sectoral ministries at the national level. Combined with the inadequate capacity of PPUs and the absence of any power to generate its own revenue, provincial administrations remain instruments of the central government and are responsive to it. Thus, although funds for expenditures at the provincial level which, under the previous system, would have been included in ministerial budgets are now presented in the form of provincial budgets, the basic change is in budget presentation and accounting, not in financial planning or discretionary financial management. 244. District Level. District councils, as legally separate governmental entities, are not part of the national financial management system. However, as is the case in many developing countries,~/ the primary source of finance for district councils is transfers from central government. Although the percentage of locally generated revenue, in the form of taxes, licenses, and users' fees, varies by district, it constitutes a very low share of total district council financing. Thus, in actual practice, central government finances council activities in several ways; including providing loans for capital construction and assistance in obtaining external donor funding for capital projects. 245. The formal procedures for requesting and providing financial assistance by central government to district councils is adequate. However, the problem of inadequate adherence to such procedures, identified in Chapter IV for the central government itself, also applies to the operations of district councils and relations between councils and the Ministry of Finance. Primary among these problems are large end•year supplementary appropriations reflecting inadequate expenditure control and lax fiscal discipline. Thus, although district councils contribute only marginally to aggregate public sector revenue, they do impact significantly on aggregate expenditure. The problem of deficit financing at district levels is discussed in Section B below. Therefore, the discussion of budgetary transfen: from central government to district councils is limited here to a description of procedures. 246. A flow chart depicting the budget preparation and approval process at the national level is provided in Volume Il, Annex V; Figure AN.1 illustrates that process at provincial and district levels. In summary, tentative annual expenditure budgets are prepared by district councils and submitted through the Decentrali:mti.on Division, PMO to the MoF during the period prior to August of each year. District Councils are, by law, required to prepare balanced budgets; any budget not in balance is automatically rejected by the Decentralimtion Division. Such budget proposals, including items to be financed locally, must be approved by the Decentrali:mti.on Division and, in some cases, are often also reviewed by provincial administrations (see paras. 246). That procedure effectively allows only one month for district councils to respond to changes required by the Decentrali:mti.on Division and submit their revised budget proposals for incorporation in the National Budget document. However, as difficult as the prescribed schedule is, actual practice creates a substantially worse situation, as illustrated by the case of the District Council which claimed that actual notification of its approved budget was not received for the duration of a full fiscal year. 5.,/ Jerry M. Silvennan, Public Sector Decentraliz.ation: Economic Policy Refonn and Sector Investment Programs, Africa Technical Department, Public Sector Management Division Study No. 1 (Washington, D.C.: The World Bank, 1990). 70 247. Although district councils are required to maintain accounts of actual expenditure in terms of approved budgets and to report such information to the MoF Controller, central government monitoring of local financial management is almost impossible. Two key factors account for that difficulty: (i) district councils do not adhere to the requirement that accounts be submitted to MoF and the reporting by most such councils is several years in arrears and (ii) there is almost no transparency in terms of the purposes to which central government transfers are to be allocated at the local level. In regard to the latter issue, MoF transfers to all district councils are aggregated in the national Budget under the single category of Grants and Other Payments (see Chapter V). No disaggregation of that item is presented either by district or specific purpose ~ . personnel, petrol, office supplies and equipment, vehicles). Thus, for example, the MoF assumes that 60 percent of such transfers are allocated by councils to personal emoluments; but it does not know, nor does it attempt to determine, if that is, in fact, the case. Employment 248. Information concerning aggregate numbers, distribution, and costs of district council employment is provided in Chapter IV. As described in that Chapter, 13 percent of total non-parastatal public sector employment is at district level. Local Administration Service (LAS) employees, who constitute the district equivalent of regular Civil Servants at the national level, account for 8 percent of public service employment; while about 20 percent of •temporary" public service employees are at that level Ci&., National Joint Council Employees [NJCEs]). 249. Although each district is responsible for its own payroll and payment procedures, most Jocal employees are, in fact, financed by the central government. As indicated in Chapters Ill and V, such financing appears in the national Budget under the category Grants and Other Payments. Further, although distinguishing salaries, wages, and benefits from other expenditures within block grants is impossible, it is clear that MoF's assumption of 60 percent for personal emoluments is too low a percentage (Chapter IV). Districts have diverted significant amounts of non-wage jl'aDts to personnel emoluments, as they are simply in great need of cash for that purpose. At the same time, district councils are clearly overstaffed. 250. However, bec.ause district councils rely on manual payment lists as their primary means for monitoring personnel size and distribution and managing expenditures on personnel, precise information on sir.e and composition, either permanent or contractual, is difficult to obtain. Central government ministries have little control or knowledge of actual employment and personnel expenditures at subnational levels. 251. Estimates of district council empfayment are discussed in Chapter IV. Suffice it to reiterate here that permanent LAS staff at district level equal 8,275 (a 26% increase from 1983) and supposedly temporary NJCE are estimated at 18,772 (about a 3S0'% - 400% increase over 1983 levels). In that context, it is worth remembering that perhaps as much as 35 percent of NICE are engaged in local government owned and managed industrial and commercial ventures. Recent aggregate wage ·nm increases, as described in Chapter IV, have exerted devastating effects on district councils. It is estimated that, if cunent trends continue, total wage biU costs for district council employees will likely reach k2.4 billion by end 1991. B. ~ 252. In the short-term, three major issues will need to be resolved if Zambia's decentralization program is to result in a positive contribution to achievement of the dual objectives of fiscal stabilization and improved efficiency and effectiveness of the public service. These three issues are: (i) excessive central government control; (ii) deficit financing; and (iii) inappropriate economic policies. 71 Central Government Control • 253. The provision of the Local Government Act which subjects all decisions taken by district councils to intervention by the Minister responsible •at the discretion of the Minister• has been employed by central government authorities so as to, in fact, require approval from the Decentralimion Division of practically all local decisions. The essence of the problem is that implementation policies at central level and the economic policies and actual implementation capacities of district council administrations do not, and cannot as currently constructed, contribute to achievement of the program's objectives. More specifically: (a) The successive ministers responsible for decentralization (currently the Prime Minister) and their staff (currently in the Decentralization Division) have transformed, by their behavior, the Local Government Act's assignment of substantial discretionary authority to district councils into almost complete central government control over all council decisions; (b) Under cunent conditions, District Councils' have almost no control over their expenditures and have little capacity to generate necessary increases in revenue; (c) The Decentralization Division issues the financial regulations under which the district councils operate; (d) The Decentralization Division must approve district council requests for central government grants prior to submission to MoF, issues guidance for budget preparation indicating the expected amount of such grants, and the Prime Minister must approve each district council's budget submission (see Figure VI.3); (e) The Prime Minister must approve any change in revenue measures proposed by district councils; (t) The Decentralization Division negotiates labor contracts which establish the terms and conditions of service for district council employees; (g) Provincial level local government officers and Local Government Auditors employed by the Decentralization Division respectively, are required to inspect council operations quarterly and audit council accounts (although actual performance of that responsibility is lax); (h) The prime Minister has general authority to exercise all powers of district councils; and (i) The President appoints the district governors who serve as chairpersons of the councils. 254. The issue here is not that the central government should exercise no influence on the activities of district councils; quite the contrary. Experience in many countries strongly suggests that effective local government performance requires strong central government performance of mutually reinforcing functions. It should be expected that the center will retain responsibilities for those functions for which it has the comparative advantage; although even in that case central ministries might usefully deconcentrate responsibilities for specified functions to their own personnel at field levels. At the same time, some specific functions in selected subsectors are best devolved to local governments (see paras. 283 - 284). 72 Figure VI.3 BUDGET SYSTEM AND PROCEDURE: FLOW CHART AT PROVINCIAL AND DISTRICT LEVELS ' ...................................... Senior PeM1111nent Secreter, UOF (Budget)• NCOP I........................................... ' • • • • • • • • • • • • • • •= • • • • • • • D • • • • Sectoral Decentral1zatlon • • U1nlatr1H Office • • (HQe) •••••••••••••• •••••••••••••• •••••••••••••• ••• a I (Cabinet) I • • . • • • • • Interectio n • • . • ·- . . • Pel'llenent S.Cretar1 (Prov1nc1al) - Sectoral llln1etr1•• . • In Province• . • . • . • Interaction • Dletrlct Prov1nc1al Counclle Plann1ng Interaction Unite Note: ••• lndlcetee flow of lnetructlone on budget guldellnee and celllnge. Indlcetee Interaction and coneultatlon during budget preperetlon. = Indlc•te• flow of budget eubaileelon• to (UOF end NCDP). • 73 25S. Such an approach requires both strong links and shared res,ponsibilities among both central and local governments. That, in tum, requires judgements about which functions, in which subsectors, ought to be delegated ~/ to which organizations or deconcentrated or devolved to which level of government. The performance of each and every function at district level should not be the objective of Zambia's decentraliution program. Thus, some functions are clearly national in scope and responsibilities for them ought to be retained by the central government. With that understanding in mind, an outline of criteria for decisions concerning appropriate assignment of functions to local governments can be developed; as suggested in Section C below. Suffice it to iudicate here that current actual practice in Zambia does not adequately balance the roles of central and local governments. 256. The negative consequences of currently excessive intervention and control of local governments by central government is illustrated by two common situations: (a) When, in order to balance their budgets, district councils request the Minister to approve increases in property taxes (i.e., •rates•), such increases are seldom approved (or only in small proportion to the request) on the grounds that increased revenue should be sought from more efficient collection of rates from all obligated taxpayers rather than increased collection from the small percentage of persons who actually pay such rates; and (b) Although, by law, district councils are supposed to have the authority to determine their own staffing structures, levels, and terms and conditions of service (the primary reason for abolishing the Local Government Service Commission), in practice, the Decentraliution Division bas established uniform staffing patterns for councils and the relevant trade unions negotiate a uniform contract directly with the Decentralwtion Division, which is then uniformly applied to employees of all councils. 257. The result of such practices is that district councils, because of lack of capacity to efficiently and effectively collect rates combined with little control of their primary expenditure item (i.e., their wage bill) <ire, under current economic conditions, forced into deficit financing. Deficit Financin& 258. Although district council budgets are required to balance revenue and expenditure, such balancing is, under current conditions, virtually impossible; in part for reasons summarized above. Thus, almost all budgets prepared and submitted by district councils each year are admittedly fictional; the common practice is to attempt to realistically estimate expenditure and create whatever revenue estimates are necessary to balance such estimated expenditure. The important question which arises from such practices is: how do district councils finance such deficits? The answer to that involves the use of a variety of financing mechanisms, including the following: (a) District councils have the legal authority to borrow and have done so in the past (however, banks are no longer prepared to lend money to councils, that sort of financing has dried-up, but many councils are accumulating arrears because of non-payment on such previous loans); ~/ Delegation involves the assignment of authority to autonomou11 (or largely autonomous) organizations responsible for a single activity or limited range of related activities; for example, assigning responsibility for the performance of specific activities to a parastatal, university, or water district. The essential point here is that such organizations are essentially independent for all operational, although perhaps not strategic, decisions affecting its responsibilities. 74 (b) District councils accumulate overdrafts on their cash accounts with various banks (however, that source is also drying-up because of tighter controls on overdrafts by the Central Banlc); (c) District councils collect various license fees, 10 percent of which they are entitled to retain and 90 percent of which is supposed to be forwarded to the central government Treasury, however the practice has been common for the past several years for district councils to retain the entire 100 percent of such fees (this source is also likely to dry-up as the Ministry of Finance has indicated that it intends to begin deducting the 90 percent of the estimated value of such fees from the grants provided by the central government to district councils); (d) District councils delay, or stop altogether, payments to suppliers of goods and services (however, this method is also drying-up as suppliers have begun to refuse to provide goods and services to councils without cash payment in advance); (e) District councils retain the amount deducted from employee salaries as contributions to the Superannuation Fund (pensions) and do not pay their own obligatory matching contributions to the Fund (however, this too is likely to dry-up as the Ministry of Finance has indicated that it intends to begin deducting the estimated value of such contributions from the grants provided by the central government to district councils); and, finally, when all of these various sources of deficit financing dry-up, (f) District councils delay payment of wages, salaries, and benefits to their employees. 259. That these various deficit-financing mechanisms have almost all been shut-down is clearly demonstrated by the spate of newspaper reports, during 1990, of local government employees not being paid wages, salaries, and/or benefits for several months in various parts of the country. Economic Policies 260. An additional source of unbalanced district council revenues and expenditures is the various commercial and industrial ventures operated hy district councils. Such ventures are ill-advised for at least four reasons: (i) almost all of them produce financial losses; (ii) such activity is a replication of past public sector economic policies at the central government level which directly contradict the principle underlying the current Economic Recovery Program that the public sector should withdraw from such economic activities in favor of the private sector; (iii) the district councils do not have the technical or managerial capacity to operate such ventures efficiently or effectively; and (iv) the current terms and conditions of service provided council employees working in such ventures are significantly better than the equivalent benefits provided employees of private sector competitors. 261. The latter point is particularly important because it can be expected to have a direct impact on the public service reform program. In summary, employees of councils who work in commercial ventures (£&, the operation of bars and taverns, retail shops, hotels and guest houses, council farms) receive the same wages, salaries, and benefits as all other local government employees and can be dismissed or retired only under the terms, and in conformity with, the procedures established for local government employees. This is a significant issue, in part because it is estimated that perhaps as much as 3S percent of local government employment consists of workers in these various enterprises. 262. The result of the practices outlined above is that District Councils suffer from the worst of both devolved and deconcentrated systems: (i) they are assigned r:sponsibilities that they have no ability to carry-out from the perspective of both policy and institutional capacity and (ii), for all practical purposes, they are dependent on direction with respect to both technical and financial matters from central government ministries. Thus, it is 75 imperative that Zambia's decentralization policy is revisited and that it is addressed in terms of changes required to conform to the overall macroeconomic reform strategy currently in place. c. Recommendations 263. Government should give serious attention to resolving fundamental policy issues concerning the appropriate role and responsibilities of provincial and district level authorities because of the significant impact they can have on the implementation of the economic reform program and on the performance of necessary Government functions. Specifically, the Government should clarify its intentions concerning: (i) the extent to which district councils, as public sector entities, are expected to adhere to Zambia's current fiscal stabilization objectives and principles underlying the Economic Reform Program; (ii) the primary functions that district councils should be expected to perform in which specific subsectors; and (iii) the manner in which central and provincial government will be organized so as to foster mutual support and complementarity among the three tiers of government. Therefore, recommendations concerning decentralization will be presented under two headings: (i) Fiscal Stabilization and (ii) Organizational Arrangements. Fiscal Stabili:zation 264. Several actions could be undertaken by Government to reduce the negative impact of district council activities on the aggregate public sector deficit. Such actions encompass the following areas: (i) district entelprises; (ii) expenditure management; (iii) debt management; (iv) the wage bill; and (v) revenue generation. 265. District Enterprises. Commercial entelprises operated by district councils are a severe drain on local public sector budgets. Few of these entelprises generate Slllplus n,venue and all place a substantial management and logistical burden on district government. More importantly, the maintenance of such entelprises within the public sector at whatever level of Government is contrary to the underlying principle, inherent in the economic reform program, that the role of the public sector will be transformed to providing an enabling environment for private sector investment and operations in the commercial sector. 266. Therefore, at a minimum, district councils should no longer be permitted to allocate any financing from central government grants to the establishment or operations and maintenance of such entelprises. Government should also consider establishing guidelines for district councils directed at divestiture of all such entelprises over a specified time period. As an element of such a policy, the assets of divested entelprises ~ . buildings and equipment) could be leased to employees dismissed as part of the retrenchment program to facilitate their transition to the private sector. 267. Expenditure Mapaaement. As in the case of financial management at the national level (see Chapter V), there are two problematic areas at the district level: (i) transparency of budget presentation and (ii) effective accounting and auditing. 268. Trans_parency of Central Government Transfers. The Ministry of Finance (MoF) should establish a budget format which would disaggregate the Grants and Other Payments item by district and, within each district grant, by line item allocations for which such grants are made. The same recommendations for enhancing financial management within ministries at the central government level should be applied to actual expenditures financed by grants to district councils (refer to Chapter V). Such a policy would provide a basis for monitoring expenditure at district levels in terms of impact on aggregate public sector financing. 269. Accounting and Auditing. An urgent requirement is the establishment of an institutional capacity within district councils to implement a priori expenditure control. Thus, the position of internal controller should be filled in all district administrations and, in those cases in which such positions are filled, capacity should be 76 improved. The accountability of district managers should be enforced through the strengthening of the Auditor General's Office and the application of sanctions based on bis/her reports (see Chapter V). 270. In addition, the larger district councils currently have the authority to conduct their own audits; including the use of external auditors. That authority should be extended to all district councils. If under such a system, annual audits are not submitted by councils, the Decentralization Division should hire external auditors to complete them on behalf of delinquent councils; deducting the cost of such audits from the following year's MoF grant. The Auditor-General's staff could then selectively review district council audits for accuracy and disqualify auditing firms demonstrating a pattern of inaccuracy from an authorimd list of such private sector firms. 271. Finally, gross financial malfeasance of district officers, as uncovered by audits, should result in disciplinary measures by district councils. If that does not occur, or if councils do not comply with financial management regulations, the Auditor-General should be authorimd to sue delinquent managers in the law courts and/or the MoF should withhold subsequent grants from recalcitrant district councils. 272. Debt Management. According to the Local Government Act of 1980, district councils have the authority to obtain loans from domestic sources and many have, over the years, accumulated significant debt. However, under current conditions, many councils no longer have the ability to service that debt. Default might soon become a common occurrence. It is essential that a review of outstanding district council debt be undertaken and, if a serious problem is found to exist in that regard, a program for addressing current and future debt issues should be established. 273. District Wye Bill. This Section is limited to recommendations for non-parastatal public sector staff retrenchment which are specific to the district level and should be understood in the context of the overall program presented in detail in Chapter IV. 274. Based on a clear identification of overall program requirements, the first step is to reduce central government transfers to district councils by an amount equal to approximately 25 percent of their estimated personnel costs, with the obligation of councils to establish a retrenchment program under the same terms as those for central government employees. 275. As in the case of ~!ral government agencies, requests for central government grants to district councils should include explicit retrenchment targets provided in guidelines prepared by MoF, in close consultation with the Cabinet Office and the Decentralization Division (PMO). Such targets should include a list of all: (i) "ghosts;" (ii) persons retiring under normal conditions; (iii) persons accepting or being proposed for early retirement; and (iv) finally, those proposed for dismissal (and the reasons for dismissal in each case). Budget applications should also include an explicit statement about the impact of such retrenchment on the Council• s mission and operations. Future recruitment of NJCEs should be drastically reduced; if not baited entirely, until such time as districts are largely self-financing. 276. Reyenue Generation. Revenue generation at district levels is woefully inadequate. Little incentive currently exists for district councils to improve the effectiveness of local revenue collection because there remains a sense that the central government will come to the financial rescue when n ~ . Three significant changes must occur if that situation is to be improved. Thus, district councils must: (i) be held to the same strict standard of expenditure ceilings and control to which Government is now committed to holding central government ministries; (ii) have access to adequate sources of revenue; and (iii) improve their technical ability to design, establish, and implement revenue systems. 277. Although district councils are authorimd to levy and collect property taxes, their capacity to assess such taxes and to collect against such assessments is severely limited. Further, it is not known whether property taxes would provide a sufficient revenue base for the more rural districts. Therefore, the comprehensive review of the national tax structure recommended in Chapter IV should include explicit attention to the resource 77 requirements of district councils; with a reassignment of one or more tax sources from central to local levels as one possible result. To undertake such a review and assist district councils to improve their revenue generating capacity under any new public sector revenue system, technical assistance should be provided to both central and district levels of government. Organizational Arran&ements 278. Determination of whether or not specific organiz.ational arrangements are appropriate is contingent on their role and purpose; in that sense form follows function. Therefore, if Government is serious about pursuing some form of decentralfaation in the context of the current economic reform program, it needs to clarify its approach by identifying those specific functions within specific subsectors which ought to be: (i) privatized; (ii) devolved to district councils; and/or (iii) delegated. if any, to specialized institutions, such as parastatals, teaching or research institutions, and/or "special districts"~. water districts). All other functions, by sub-sector, would then, by definition, be reserved to the central government. As for the organimtion of central government and provincial administrations, decisions would also need to be made concerning which specific functions reserved to the central government ought to be deconcentrated within specific subsectors and which ought to be performed at the central government level. 279. As each of the decisions outlined above are made on an evolving basis over time, appropriate systems and procedures can be designed and established to improve the financial and operational efficiency and effectiveness of district councils (and determine the appropriate role, if any, of provincial level administration). In order to facilitate appropriate decision-making in the context of that approach, it will be necessary for Government to establish specific criteria to guide such decisions. The remainder of this Section suggests some criteria which might usefully be considered in that regard. 280. Privatization. Decisions concerning the overall role of the public sector provide boundaries for contingent decisions concerning what functions ought to be devolved, delegated, and/or deconcentrated within the public sector itself. A basic principle inherent in Zambia's economic reform program is that the role of the public sector in the economy will be transformed from the primary producer of goods and services to: (i) the provider (but not necessarily producer) of public goods and services and (ii) the regulator of appropriate private sector economic behavior (see Chapter m. Thus, it is expected that the public sector will focus its attention on providing an enabling environment for the private sector to become the primary engine for economic growth. In broad terms, the essential functions which Government must perform relate to the provision of public goods and services. As presented in previous chapters, such functions include: (i) activities necessary for decisions to provide a public good or service (including assessment of its technical feasibility and economic viability); (ii) the specification of construction and/or operational requirements; (iii) an investment and/or recurrent financing plan; (iv) the generation of finance; (v) the contracting for and supervision of production of the good or service (which might be produced by the private sector under contract to the public sector); and (vi) evaluation of the provision and production of such goods and services for enhanced future planning and decision-making. The point has also been made that the list of functions which Government must necessarily perform to fulfill its provision function do not include the actual production of public goods and services. 281. These general principles, as applied to the role of the public sector at the central government level, are no less applicable to the role of government at the district level. Indeed, it has been shown above (paras. 260 - 262) that the expansive role undertaken in the commercial sector have had similarly negative impact on the financial condition of district councils. Therefore, consideration should be given to establishing a policy on district councils' divestiture of their commercial and industrial ventures. Priority ought to be given to divesting (i.e., liquidating or privatizing) those enteJFrises which are currently operating in deficit, followed by those whose overall inefficiencies represent a managerial, if not financial, chain on district council resources. In the latter case, priority should be given to the divestiture of commercial operations in direct competition with the private sector ~ . guest houses, bars and restaurants, furniture manufacturing. and district farms). To be effective, such a policy will require that 78 responsibility for providing guidelines and monitoring progress toward divestiture objectives is assigned to some specific organi7.ation; perhaps in the Decentralization Division of the PMO. 282. Public Sector; As.,ianment of Functions. With the divestiture of responsibility for operating commercial enterprises and relief from the need to produce a vast array of public goods and services, district governments would be able to focus on political decision-making, planning, and financing rather than those of physical infrastructure construction or management of actual service delivery. The former skills are precisely those in which local authorities share a mutually supporting advantage with national governments. Where the focus of political decisions are on national integration issues, central governments have the comparative advantage; where the focus is on location-specific issues, local authorities have the comparative advantage. Where the focus of planning is on macroeconomic priority-setting and inter-sectoral integration, central governments have the comparative advantage; where the focus is on location-specific investments and service delivery within such national guidelines, local authorities have the comparative advantage. In the case of financing, comparative advantage is largely determined by the sources of such financing. 283. Devolution. Criteria for determining which functions, within the appropriate scope of the public sector, ought to be devolved to district councils should include, but not necessarily be limited to, the following: (a) Those operations and functions for which a high degree of location specific variation is required; (b) Those operations and functions for which quick response times are required; (c) Those operations and functions for which economies of scale are relatively small; (d) Those operations and functions which are labor, rather than technical and capital, intensive; (e) Those operations and functions in which inter-level dependencies among levels (central, provincial, and district) are relatively small; (f) Those operations and functions which require close integration of activities across sectors, but which, at the same time, are limited geographically; and (g) Those operations and functions where local participation is desirable and/or required. 284. In that context, district councils should have primary responsibility for strategic and program planning within the subsectors of their responsibilities, including medium-term financing requirements, issuance of strategic sectoral planning guidelines, identification of project level activities for transforming program objectives into action, and review and monitoring of project proposals and implementation to ensure conformity with program planning priorities. 285. Unfortunately, the decentraliz.ation system in Zambia largely ignores such criteria for determining the assignment of responsibilities among levels of government. Changing the allocation of responsibilities so as to benefit from the comparative advantages of the various levels of government and the private sector should result in the enhanced performance of local governments. If that were accomplished, much of the criticism of local government inefficiencies might be resolved. Thus, the improvement of district administration sufficient to meet the more limited scope of its responsibilities would not require any massive institutional development program at subnational levels. Nevertheless, to effectively perform even these more limited responsibilities, it will be necessary to establish District Planning Units (DPUs), define their functions and roles clearly, and provide them with trained personnel and technical staff. 286. Finally, if substantive, rather than formalistic, devolution of specific functions is to operate effectively, it is absolutely necessary that district councils become largely self-financing with respect to 79 implementation of the decisions they, themselves, make. Without the ability of district councils to largely finance their own decisions, central government will need to protect itself against potential financial exposure which might result from unwise local decisions. Substantive devolution cannot be expected to occur under such circumstances. However, such a rule does not mean that district councils need to be self-financing with regard to every activity they undertake. Indeed, to the extent that central government mandates responsibilities for specific functions in specific subsectors to district councils without any discretionary choice about such participation, central governments should finance such activities. In the latter case, the issue is not the financing of devolved authority, but rather the financing of deconcentrated authority (see para. 288). 287. Deleiiation. Delegation of authority in Zambia is exercised primarily with regard to parastatals, universities, training institutions~. the National Institute of Public Administration [NIPA]), and hospitals. With the exception of delegation to parastatals (which is addressed in other World Bank reports), such limited use of delegation is probably appropriate. The basic criteria to be applied to any consideration of delegating authority to specializ.ed organizations is either that the service provided is either highly speciali7.ed or that the geographic area which maximiz.es service efficiencies is ecologically different from the established administrative boundaries of the State. Thus, for example, the efficiency of water delivery for either irrigation or drinking purposes might be enhanced if the same organization integrated the management of the infrastructure from the watershed to delivery points. It should be noted, however, that if such specialized agencies are created, they must have the ability to finance their operations through either special taxing authority or the levy of adequate users• fees. 288. Deconcentration. The fundamental criteria for choosing to deconcentrate responsibility within specific subsectors, rather than devolve such responsibility, is that the requirements for uniformity in application outweigh the advantages of loca: discretion. In such cases, additional criteria to be considered should include, but not necessarily be limited to, (i) those operations and functions for which some significant, but not major, degree of location specific adaptations are required to generally applicable policies and implementation specifications and (ii) those operations and functions for which span of control factors can have a significant impact on the effectiveness of decisions, again within the context of generally applicable policies and implementation specifications. With specific reference to deconcentrating authority to intermediate levels of administration, such as the provincial level, two &dditional criteria are usefully considered: (i) those operations and functions in which inter-level dependencies among levels (central, provincial, and district) are relatively greater than appropriate for district levels but smaller than necessary at central level and (ii) those operations and functions which require some integration of activities across sectors, but which, at the same time, are limited geographically (although not as limited as those of a district). Assuming Government's commitment to maintain provincial administrations as a deconcentrated arm of the central government, there is an obvious need to develop appropriate criteria for assigning functions to both the provincial and district levels so as to clarify the various ambiguities currently existing in the system. Illustrative Conclusion 289. Applying the criteria outlined above, it might reasonably be decided, for example, that in the primary education sub-sector, district councils would be responsible for: (i) selecting sites for primary schools; (ii) determining the specifications for construction and geographic coverage of each school; (iii) financing the basic construction of such schools; and (iv) hiring, promoting and dismissing teachers and school administrators. The central government Ministry of Education might reasonably be responsible for: (i) establishing basic curriculum standards; (ii) establishing professional qualifications for teachers and school administrators; (iii) financing basic curriculum materials; and (iv) setting examination standards or, at certain stages of the education process, examination content. Monitoring adherence of district council schools to those standards appropriately set by the central government should also be the responsibility of the Ministry of Education, but such monitoring might appropriately be deconcentrated to the Ministry's professional staff assigned to provincial or district levels. This example should be viewed as illustrative, rather than prescriptive. The point to be made is that the assignment of responsibilities should not be defined in terms of an entire sub-sector, but rather in terms of specific functions within the sub-sector. Having done so, mutually reinforcing institutional arrangements among levels can be determined. 80 VIl. RECOMMENDATION: AN INTEGRATED STRATEGY 290. As stated in Chapter I, the purpose of this report is to assist the Government of Zambia to formulate a medium-term strategy for rationalizing its performance of economic, financial, and personnel management functions in the context of its overall macroeconomic reform program. In that context, the discussions in previous chapters have included several specific recommendations for action in the short, medium, and longer terms. Those specific recommendations need not be repeated here. Rather, this Chapter focusses on the need to integrate these various recommendations into a sufficiently comprehensive and unified Program which can be managed by Government and supported in a consistent and mutually reinforcing manner by donor agencies. Particular stress is placed here on the need for: (i) program integration; (ii) appropriate phasing; (iii) new approaches to Technical Assistance (TA) and training; (iv) improved coordination of international donor financing by Government; and (v) a structure for effectively managing the overall Public Sector Management Reform Program (PSMRP). The Chapter concludes with a brief discussion of assumptions, risks, and proposed responses to such risks. A. Prom:am Intmation 291. A careful reading of Chapters Il - VI demonstrates the mutual contingencies among planning, financial management, and personnel management functions. Thus, for example, if reform of public employment is to support Government's economic policy of growth with equity, it will need to be considered in terms of both fiscal stabilization on the one hand and, on the other hand, improved operational performance of the public sector's necessary functions. At a minimum, it is clear that without effective expenditure control, no program of staff retrenchment in Zambia is possible. Further, to improve Government's performance of its operational responsibilities, streamlining the structure, procedures, and staffing of the public sector is necessary; even if staff retrenchment was not required to meet more limited fiscal stabilization requirements. That, in tum, requires an ability to monitor the si7.e and distribution of public employment in terms of Government's planned priorities over time. An important element of an effective personnel information system is an effective, computerized, payroll system. Such a system can serve not only to monitor expenditure, but also to produce key information on the size and distribution of personnel in an easily manipulable manner. 292. Therefore, effectiveness of any Public Sector Management Reform Program (PSMRP) will need to integrate key elements among the planning, financial management, and personnel management functions. That, in turn, requires a clear articulation of Government's policy in that regard. Such a policy should focus on three interrelated objectives: (i) design and initiation of a comprehensive medium-term program for achieving the dual objectives of fiscal stabilization and improved performance in the context of the overall goal of economic growth with equity; (ii) establishment of specific limits on public expenditure based on clear cross-sectoral priorities established in terms of its Economic Recovery Program (ERP); and (iii) effective improvement of its performance in the high-priority sub-sectors established in the context of its overall strategy and financial constraints. 293. In particular, it must be stressed here that appropriate reforms of the public service are contingent on the broader restructuring of the public sector implicit in the requirements of both Government's ERP and its policy of ·1~tralization (Chapter VI). In that regard, the focus should be on the entire public sector at all levels, not just central government organizations. Given the demonstrated impact of local government economic, financial, and employment behavior on public sector performance, Government should give serious attention to resolving fundamental policy issues concerning the appropriate roles and responsibilities of central, provincial, and district level authorities and the relationships among them. 81 B. Phasin& 294. Notwithstanding the interrelated nature of these vario•IS economic management functions, not all of the actions required can, or should be, undertaken simultaneously. An improvement strategy, to be successful, requires priorities to be set and improvements phased. That is tnae for at least six reasons: (i) no organization, including Government, is capable of undertaking all of the actions required simultaneously; (ii) stomng requirements to accomplish improvement objectives require increasing levels of professional sophistication; (iii) current performance levels suggest that the GRZ does not yet have the staff and/or resources to operate a complete system and such resources will have to be developed over time; (i\') some actions are contingent on the completion of other actions; and (v) the fact that some actions require longer periods of time to accomplish than other actions; and (vi) the brief review of current performance demonstrates clearly that the initial building blocks for other actions are not yet effective or, in some cases, in place. Therefore, although the long-term objective is to comprehensively improve public sector performance in the context of its changing role, the recommendations provided in this report are based on the awareness that any such program must be appropriately phased over time. The discussion in this Section will be limited to identifying contingent relationships (i.e., •building-blocks•) and the need to differentiate among actions which can be accomplished in the short-term from those requiring a longer term perspective. Buildin1-Blocks 295. Under current economic and financial conditions in Zambia, three related element., provide the foundation for the effective accomplishment of PSMRP objectives: (i) a clear articulation of Government's policy, including clear specification of objectives and priorities for reform of the public sector; (ii) fiscal stabiliution and redistribution of available financing so as to support high priority development objectives and functions; and (iii) control of expenditure so as to ensure that actual disbursements reflect budgetary decision-making. That foundation is of particular importance to any sustainable progress toward public service reform during the initial three-year period. 296. Government Poli4:y. As argued throughout this report, effective implementation of the recommended PSM~ requires an explicit and £1grlx articulated Policy Statement(s) which sets forth the framework guiding Government's reform effort. The constituent elements of that policy framework have been described in various parts of this Report and need not be repeated here. Suffice it to say that too many government officials are ignorant of the overall policy framework which ought to guide their own operational decision-making. Examples of inconsistencies between operational decision-making and current policy commitments abound; many of which have been reported in this paper. Two clear eumples suffice to illustrate the point: (i) Personnel Division's lack of understandin~ tltat Government's policy is no longer to continue compressing the personal emoluments scale, but is now predics•~ on attempts to decompress such scales (Chapter IV), and (ii) continuing investment by some district councils in the creation of commercial enterprises with the expectation that return on such investment will serve as a primary source of local government ,:g·venue (Chapter VI). 297. As discussed in Chapter II, Government's economic policy objective, its redefined role in the economy for the achievement of that objective, and the limited resources available to it for the performance of its appropriate roles, requires decisions as to its investment and operational priorities. Determination of economic development priorities requires a focus along two dimensions: (i) the actual financing and production of priority public goods and services and (ii) the functions which Government needs to perform so as to provide such public goods and services. Improving performance within high priority sub-sectors is the test of any program to improve the public service. 298. The provision of high priority public goods and services can be achieved only through the performance of key economic management functions. Those which are necessary to such provision are, therefore, high priority. This report has identified five such functions as high priority: (i) decisions concerning a public guxi 82 or service (including assessment of its technical feasibility and economic vi9.bility); (ii) specification of requirements; (iii) planning to meet financing requirements and actual generation of such finance; (v) contracting for and supervision of production of the good or service; and (vi) evaluation of the provision and production of such goods and services for enhanced future planning and decision-making. Co:..nbining consideration of high priority sub- sectors with the priority functions which must be performed within them provides a rational set of criteria for operational policy decision-making in the context of an economic and financial development strategy. 299. In order to effectively implement a program according to these criteria, Government should, at a minimum, prepare and circulate among its own officiaJr. policy documents which explain the key criteria for ect'nomic decision-making and institutional reform inherent in its own ERP; includin& a discussion of the changing i:ole of the public sector and identification of cross-sectoral priorities within the scope of such changes. Once such guidelines are disseminated, decisions concemini key actions which need to be implemented at the operational level can be taken in a consistent manner by officials at that level. 300. Fiscal Stabilization and Reallocation of Expenditure. Fiscal stabilization is essential if any significant improvement of Government's operational performance is to occur. This paper has documented the effect of Government's deficit financing on the economy as a whole and has shown that, even with such deficit financing, sufficient resources are not available for the actual performance of either recurrent or investment activities. Thus, increases of staf/ have been accompanied by an almost absolute reduction in operating funds. Priorities need to be clearly established for redistributing scarce resources so tht high priority operations are emphasized and can be implemented. Among those high priority operations is the reform of public sector management itself, so as to improve Government's capacity to plan and manage its financial and human resources. In the absence of such reallocation of resources so as to concentrate on high priority operations, each and every Government function will remain under-funded and sub-optimization will continue. Extraordinary requirements, such as the PSMRP, will wither away due to lack of sustenance. 301. Expenditure Control. Effective control over actual expenditures is an absolute prerequisite to achieving fiscal stabilization in the short-term and improving public service performance over the longer term. Without financial discipline, policies, plans, and budgets have little, if any, impact on the actual use of resources. At best, high priority initiatives will succumb to procedural inertia; at worst, resources will continue to be maldistributed in the interests of those with actual expenditure authority and/or physical access to such resources. This is particularly the case with reference to public service r;;form. For example, initial targets for stah retrenchment should be established in fiscal terms, rather than in terms of specific numbers of personnel or positions to be reduced. However, such an approach cannot be implemented without effective expenditure control. 302. Thus, effective implementation of the PSMRP's medium-term objective to improve the efficiency and effectiveness of the public sector requires a well-established foundation consisting of the three elements identified above. Short vs. Lonrier Tenn 303. Some actions can be undertaken in the short-term because Government either has sufficient current capacity in the broader institutional sense (i.e., appropriate organizational, procedural, and staff elements are in place) and/or because such actions are sufficiently discrete so as to enable the employment externally of skilled personnel who can serve as substitutes for inadequate internal staff capacity. In the former case, improved performance is most likely to depend on changes in operational priorities, management attention, and changes in staff behavior; all achievable with sufficient political commitment. In the latter case, where current capacity is inadequate, the common response is to seek donor financed technical assistance for the direct performance of critical activities in the short-term while initiating a program to develop an internalized sustainable capacity over the longer term. I 83 304. In the specific case of Zam?>ia, this paper has assessed Government's current capacity to articulate an integrated policy fra,11,>•.-.-~1'k and substantially improve expenditure control as adequate to meet short-term requirements. Thl.o, 1t is feasib:o. to meet such requirements in the short-term. However, with regard to the other key economic management functions, which have been described and assessed in detail elsewhere in this report, Governm~t•s current capacity is weak. Sustainable responses to those requirements necessitate a longer term institutional development approach; although interim assistance can be provided to meet other short-term requirements. C. Technical A;isistance and Trainini= Technical Assistance ITA} 305. S1gillficant amounts of TA have been provided, .'er many years to improw Govemment'R ability to plan, yet performance remains extremely weak. That is partly due to inappropriate design of TA effoi ;. That criticism can also be applied to TA in support of improving the other functions discussed in this Report. 306. Different types of TA are required for the performance of short-term priority functions on the one had and, on the other hand, development of sustainable institutionalized capacity over the medium-term. In the short-term, an integrated team of macroeconor-ists should be assigned to NCDP and at least one sector economist should be assigned w each of the high priority sector ministries to actually r,erform priority planning functions (as determined by the PIP). With regard to such assistance within NCDP, the structure of current TA financed by various donors through separate and discrete projects should be phased out in favor of an integrated, multi-donor financed, TA group similar to that employed in Kenya. 307. For medium-term institutional development purposes, the role of, and approach to, TA should be significantly different than that for the short-term performance of specific planning functions. As in the case of training (see para. 311), TA provided for the purpose of developing sustainable capacity for performance of Government's essential functions cannot be expected to have any significant impact under current conditions. Unless terms and conditions of service improve and systemic rationalimtion and support of the planning, financial, and personnel management functions are established, consultants are unlikely to accomplish their objectives. 308. Developing institutionalized capacity for planning, or the performance of Government's functions requires a long-term focus; although progressive improvement should be observable soon after commencement of the process. For any such process to be successful, the role of TA personnel must be understood primarily in terms of •coaching• and •mobilizing" on-the-job experience of m!!m composed of Government staff, rather than on individual Government staff counterparts. Indicators of progress must focus on the output of those Government teams rather than on specific plan documents, whether produced by TA personnel or Government. !/ Such an approach requires qualitatively different behavior on the part of TA personnel than that required for the performance of a specific function to meet a critical short-term requirement. It also requires a different relationship with those Government staff with whom TA advisors/consultants work. Key to TA in support of institutional development is that responsibility for specific outputs, other than improved capacity itself, remains with Government staff and is not transferred to the advisor/consultant. 309. With that approach in mind, the following principles should be applied to the provision of TA for improving Government's planning capacity over the medium-term: 1./ See Jerry M. Silvennan, Technical Assistance and Aid Agency Staff: Alternative Techniques for Greater Effectiveness, Technical Paper Number 28 (Washington, D.C.: The World Bank. 1984). 84 (a) The emphasis should be placed, in the initial phases, on system design requirements ~ . clarifying different purposes and types of plWllllg, appropriate assignment of responsibilities for each type, and mutually reinforcing procedural links among organi7.ations); (b) Responsibility for producing the design of such systems should be assigned to GRZ working teams organiz.ed around specific system design issues; (c) TA personnel should be assigned to assist the work of the specific GRZ working teams organized as above (b) by participating, but not independently producing (i) work plans and schedules, (ii) reviewing the q~ity of work produced by teams according to the schedule, and (iii) suggesting changes in such work output. 310. Those principles are based on experience which suggests that adult professionals learn best by undertaking actual activities, making mistakes, and having facilities available to quickly correct such mistakes. There are at least two implications of such an approach. The !i!§! implication is that TA personnel employed to assist in the medium-term capacity-building effort should be explicitly distingu,shed from those employed to directly perform discrete tasks in the short-term. Assigning the dual objective of short-term production of data or documents and medium-term institutional development is likely to result in attention to the discrete short-term objective driving- out the more diffuse medium-term objective. The second implication is that TA personnel should not normally serve on a long-term residential basis in order to avoid the tendency to "take over" the substantive work from GRZ staff. Rather they should return to Zambia on a regularly scheduled basis to: (i) review the work undertaken under the terms of previously established quarterly work plans; (ii) instruct GRZ planning staff on bow to correct technical faults in such work; and (iii) assist in the formulation of the next quarterly work plan. Successive visits of ten days or so every three or four months should be sufficient to provide such assistance. 311. Training. Training by itself will not achieve improved performance if the systems within which staff must operate and incentives for staff are inappropriate. With that qualification in mind, the first priority, and immediate, training requirement is for "accountants" (actually bookkeepers) within ministries and district councils. Longer term training should focus on basic budget preparation skills and on-the-job experience related to structured interaction with TA personnel. D. Donor Coordination 312. Donor financed assistance to Zambia has historically been uncoordinated by Government. Various donors have initiated project proposals without little or no reference to Government's own overall development plans (such as the FNDP). That behavior has been due, in large part, to the inadequacy of such plans as strategic planning documents and the absence of Government leadership in specifying cross-sectoral priorities, sectoral programs derived from such priorities, and specific project investments linked to transforming such priorities into action. Some progress has been made in resolving such uncoordinated donor initiatives through the establishment of a Consultative Group (CG) for Zambia as a fo1um in which Government can present a programmatic approach to the financing of Zambia's economic development. However, much remains to be accomplished before effective coordination of donor financing by Government can become a reality. 313. Leaving responsibility for coordination to the Donors themselves is not an appropriate, or sustainable, option. In an environment charactem.ed by scarce resources and the burden of external debt, Zambia must assume effective control over the objectives towards which external financing is directed. Acceptance of any and all donor financed initiatives can no longer be justified by the prevailing inclination to obtain as much foreign aid as possible (see Volume II, Annex IV). 314. Government currently attempts to coordinate donor financing through a recently created Economic Cooperation and Debt Management (ECDM) Department in NCDP and by assigning to the Minister of Finance and Planning sole authority to approve foreign financed Joans and credits. Structural arrangements to ensure 85 Government control of Slrh financing, however, is not sufficient. It is also necessary for Government to assess all proposals for donor financed assistance in terms of there contribution to achievement of the Country's priority objectives; whether requiring repayment (i.e., loans and credits) or not (i.e., grants) or whether initiated within Government itself or by donors. 31S. This recommendation is particularly relevant to grant financing because Government bas largely abdicated responsibility for ensuring conformity of 8'.ICh assistance to development objectives. Apparently on the assumption that grants do not incur debt, government officials within sectoral agencies can be authorized by their own ministries to officially accept grant aid. Indeed, district councils have been known to accept grants in the form of physical assets, such as vehicles, without the approval of any central government agency. The problem here is that grants do incur recurrent financial costs to Government. Such costs occur in both the immediate short-term and over the longer term. In the short-term, local counterpart financing is almost always required for personnel, facilities, equipment and supplies, and other o&M costs. Such costs contribute to Government's current deficit financing and, therefore, are met, at least in part, through domestic borrowing. In addition, grant financing allocated to non-priority projects can distort investment and implemen~tion priorities. 316. In response to such issues, this Report makes two specific recommendations for implementation during the short-term: (a) Meetings of the CG should increasingly focus on sectoral program strategies disaggregated into specific •projc:,ct modules• presented as requests to donors for financing; and lb) More immediately, grant aid should be approved by Government based on the same criteria and through the same procedures as loans and credits (i.e., the authority of the Minister of Finance and Planning for approving foreign financed loans and credits on behalf of the Government should be extended to authority to approve grants). Thus, proposals for donor financing should be approved only if they can be justified in terms of clearly articulated program objectives in high priority sectors which are, in tum, justified in terms of an established macroeconomic investment and operational strategy. 317. For the specific purpose of effective implementation of the PSMRP itself, those more general recommendations can be effected through Government: (i) preparing an initial presentation of an integrated PSMRP at a special meeting of the Consultative Group (CG) focussed exclusively on that program; (ii) insisting that all proposals for donor financed support for improvement of the planning, financial management, and personnel management functions, whether initiated from within Government or by individual donors, be submitted through, and approved by, the National Economic Monitoring and Implementation Committee (NEMIC; see Volume II, Annex II); and (iii) that all such assistance, once approved, be managed under the supervision of the single senior official assigned responsibility for the integrated management of the program (see Section E immediately below). E. Intmated Manaaement of the Reform Effort 318. Because the objectives of short-term fiscal stabilization, medium-term economic growth through the private sector, and the need to improve the efficiency and effectiveness of Government's role in providing an enabling environment are so inextricably linked, it is necessary for the Government to integrate responsibility for operational decisions regarding the various elements of its economic reform program. That is not now the situation. Rather, operational decisions concerning the short-term fiscal impact of Government performance is vested in the Ministry of Finance; responsibility for addressing longer term impact of performance on economic growth is vested primarily in NCDP; staffing and improved efficiency of the national level public service is vffled in Cabinet Office; and staffing and improved efficiency of district councils is vffled in the Decentralization Division in the Office of the Prime Minister. 86 319. Under such arrangements, it is natural that government officers focus on their own particular mandates, resulting in uncoordinated and, sometimes, contradictory actions. Thus, those responsible for improving the performance of public employees can be expected to focus their attention on improving incentives while those concerned with immediate fiscal stabili:r.ation requirements are more likely to focus on the need to restrict such incentives because they increase the Government's financial deficit. 320. Resolving such potential contradictions while achieving necessary improvements in public sector performance in support of economic growth with equity !Yllt fiscal stabili:r.ation requires both: (i) integrated management and (ii) mobilu.ation of participation and support from among a critical mass of stakeholders. 321. The~fore, NEMIC should be assigned responsibility for overall coordination of the PSMRP (see para. 317 above). However, it is well understood among institutional development professionals that a program cannot be managed by committee. Nor can a program be managed solely by one organi:r.ation when implementation requires actions by multiple organi:r.ations. Therefore, it will be necessary for Gov~mment to appoint an official with sufficiently high rank (i.e., no less than a Senior Permanent Secretary) to manage the Program on NEMIC's behalf. Such an assignment should be long-term and full-time. Nevertheless, such an assignment does not mean that the appointed official should be relied on to perform all day-to-day management functions. Rather, such direct management responsibilities should be dispersed to specific managers within those various organiz.ations responsible for implementing specific program components. Nevertheless, ultimate responsibility should remain with the one designated senior official. Further, once such an official is in place, s/he should rely on a participatory process involving a wide-spectrum of stakeholders in necessary studies and implementation actions. 322. The recommended process would include assigning responsibilities to ad hoc working groups; each of which would: (i) focus on a specific issue affecting priority requirements of reform (see recommendations in Chapters IV - VI); (ii) be responsible for collecting and analyzing data; and (iii) preparing recommendations for action in the context of the Government's overall policy framework. Those working groups should not, in all cases, be limited only to government officials. For example, where issues of retrenchment are addressed, relevant labor union officials should also be involved. Such a recommendation is not idealistic. Avoiding negotiations among such stakeholders during the analysis and design phases of a program only leads, more often than not, to the failure of implementation efforts later. Nevertheless, the process recommended here requires that all participants are fully aware and understand the specific nature of policy constraints and the reasons for such constraints. 323. The approach recommended here also requires that the working groups are held responsible for the work assigned to them. Although such groups should, when necessary, be assisted by specialimd TA consultants/ advisors (see para. 309 above) in order to benefit from experience elsewhere (particularly in \frica), 2/ such external persormel should not be responsible themselves for the work. F. Assumptions and Risks Assumptions 324. The most important and obvious assumption is that Government's officially articulated policy for reforming the public sector, although not well advertised within Zambia itself, represents substantial commitment. The history of p~~io-:1s reform attempts, however, does not augur well for effective implemeritation of any PSM reform effort. At the very beginning of this Report (Chapter I) it was argued that "[several] factors have influenced 1/ The lessons learned from such experiencr. are often counter to what "common sense• might suggest and Government should place a high premium on avoiding mistakes maJe elsewhere. See, for example, Lcuis de Merode, Civil Service Pay and Employment Reform in Africa: Selected Implementation Experiences, Institutional Development and Management Division, Africa Technical Department, Division Study Paper No. 2 (Washington, D.C.: The World Bank, June 1991). 87 prior failures and, therefore, it would be seriously misleading to ~ that such failures have been due solely to lack of Government commitment and/or duplicity on the part of those responsible for implementing past reform efforts." Rather, it was argued, it would be more productive to assume that "the primary causes of failure have been: (i) absence of an appropriate and well-understood policy context; (ii) lack of sharply focussed priorities and systematic phasing of program implementation; and (iii) too much faith in technical remedies which have ignored political and social realities.• Thus, the primary assumption underlying this Report is that the approach recommended here adequately takes into account the various reasons for those prior failures and provides an effective response to them. 325. More specifically, it is also assumed that Government: (i) will be able and willing to meet its obligations to the international donor group currently financing itj,! Economic Recovery Program; (ii) was fully aware of the implications for the changing role of the public sector inherent in the ERP and, therefore, that its commitment to that program will be sustained over the medium and longer term without regard to the emergency nature of current external balance of payments financing; and (iii) is concerned with improving public service performance, and all that entails, beyond its immediate concern for the fiscal stabilization objectives of such a program. Finally, it is also assumed that, if all of the above is correct, that any changes in Government which might occur over time, given Zambia's recent return to a multi-party electoral system, will not result in significant discontimities in the implementation of an effective strategy once underway. 326. The obvious risks are essentially the obverse of the assumptions identified above. Nevertheless, there are also other risks beyond the control of Government. Among such risks are: (i) lack of response on the part of the international donor community (especially those capable of providing grant aid); (ii) lack of cooperation by other Zambian stakeholders (such as the relevant labor unions) in spite of Government's best efforts to take a broadly participatory approach to design and implementation of specific elements of the program; and (iii) the strong possibility that implementation of a balanced approach to achieving the dual objectives of the program will be unbalanced because of external pressures to meet potentially unrealistic fiscal stabilization targets at any cost. Response to Risks 327. To the extent that the primary risks are that Government and/or the international donor community might not be prepared to sustain their commitment to the full implications of effectively implementing the program recommended in these pages over the long-term, there is little that the program itself can do to counter that risk. Nevertheless, the approach underlying the recommended strategy and action-program is self-consciously directed at minimizing such risks. Thus, as outlined in Chapter Il, the approach taken in this Report empbasi7.e8 the need to: (i) tightly focus the reform program a limited number of achievable priority objectives; (ii) phase program implementation in a systematic manner, focusing initially on those objectives easiest to achieve, both technically and politically and defer eventual initiation of the most difficult actions until accomplishment is practicable; (iii) develop broad-based understanding and support for the program through effective articulation, communication, and participation among both domestic and external stakeholders; and (iv) design a clearly articulated process for developing such understanding and support. I lffM3QIEN'f I Page 1 of 1 ., BICI PUOUTY SOI-SICTOU AID J'DIIC!lORS Bip priority euhaectora tentatively identified to date. subject to further clarification, ares Agriculture -- (i) retearch; (ii) agricultural and livestock extension. to small farmera; (iii) rural feeder road construction and maintenance; and (iv) credit to •mall scale farmer cooperatives for rural maize ~roceseing and maize aud fertilizer etor.tge; Education -- Ci) primary education; (ii) secon.aary education; and (iii) technical and vocational training (while all these activitie• are important. becau1e primary education ha• been hiatorically underfunded in relation to technical and higher education. that imbalance 1hould be redre11ed even within these priority areas); · Health-· (i) aupply of essential drug,, laboratory and dental material• and (ii) rehabilitation and uinteuance of health faciliti••; Trmsport -- (i) rehabilitation and maintenance of roads •••ential to agricultural and copper production and marketing and (ii) repair and extention of essential feeder road• (see Agrt"culture above). · VJ.thin high priority 1ubaector1 and the nati011&l planning and financial agencies(~, Kiniatry of Finance, NCDP, Central lank, and Cabinet Office), the following hip prlorlt7 functicm1 are also tentatively identified to date, aubject to further clarificaticmt (1) economic and financial policy formulation and analysis': (ii) planning and budgeting; (iii) financial accounting and management: (iv) domestic revenue generation; (v) aid coordination and debt management; and (vi) manpower developm&nt and personnel administration • ..
Группа Всемирного банка · Pre-2003 Economic or Sector Report
Zambia - Public Sector Management Review (Vol. 1 of 2) : Executive Summary and Main Report
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Pre-2003 Economic or Sector Report
Страна
Замбия
Источник
Всемирный банк