Guinea - Public Expenditure Review

Document of THE WORLD BANK FOR OFFICIAL USE ONLY 4"CONFIDENTIAL Report No. 15147-GUI REPUBLIC OF GUINEA Public Expenditure Review January 19, 1996 (Revised) Country Operations Division Western Africa Department Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS Currency Unit Guinean francs (GNF) USS1.00 GNF 998 (January 18, 1996) GNF I million US$ 1,002 (January 18, 1996) SYSTEM OF WEIGHTS AND MEASURE: METRIC Metric US Equivalent I meter (m) 3.28 feet (ft) 1 kilometer (km) 0.62 miles (mi) I square kilometer (km2) = 0.39 square mile (sq mi) I hectare (ha) 2.47 acres (a) I metric ton (t) 2,205 pounds (lb) 1 kilogram (kg) 2.2046 pounds (lb) FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS ABEDA Arab Bank for Economic Development in Africa ACGP Administration and Control of Large Projects ADB African Development Bank ADF African Development Fund AGETIP Agence d'ex6cution de travaux d'int&r8t public (Public Works and Employment) BCPA Bureau de coordination de la politique agricole (Bureau of Agricultural Policy Coordination) BND Budget national de d6veloppement (National Development Budget: is the sum of GNF counterpart funds of external financing in foreign currency, usually by the EDF and USAID (including PIA80) and domestically financed investment, including contributions to FINEX and 100% domesticallyfmanced projects) BOP Balance of payments CCEF Comitd de coordination 6conomique et financiere (Economic and Financial Coordination Committee) CERAC-SANTE Conseil d'6ducation de recherche et animation communautaire - Santd (Guinean NGO concerned with health matters) CfD Caisse frangaise de d6veloppement CG Consultative Group CIDA Canadian Agency for International Development CNSS Caisse nationale de s6curitd sociale (Social Security Office) DAAF Division des affaires administratives et financires (Division for Administrative and Financial Affairs) DDI Division des d6penses d'investissement (Division for Investment Expenditures) DEM Division des 6tudes macro6conomiques (Division for Macroconomic Studies) DGTC Direction des grands travaux (Cameroun) (National Directorate for Public Works - Cameroon) DGTx Direction des grands travaux (C6te d'Ivoire) (National Directorate for Public Works - COte d'Ivoire) DNA Direction nationale de l'agriculture (National Directorate of Agriculture) DNB Direction nationale du budget (National Directorate of the Budget) DNCI Direction nationale pour la coop6ration internationale (National Directorate for International Cooperation) DNE Direction nationale de l'6conomie (National Directorate of Economics) DNFF Direction nationale des for8ts et de la faune (National Directorate for Flora and Fauna) DNGR Direction nationale du gdnie rural (National Directorate of Rural Engineering) DNIP Direction nationale des investissements publics (National Directorate for Public Investments) DNPE Direction nationale du Plan et de l'dconomie (National Directorate of Plan and Economy) DNT Direction nationale du Trdsor (National Directorate of the Treasury) DP Division des pr6visions (Forecasting Division) ECOWAS Economic Community of West African States EDF European Development Fund EDI Economic Development Institute EDR equalizing discount rate EIB European Investment Bank ENELGUI Energie 6lectrique de Guin6e (National Electricity Company) ESAF Enhanced Structural Adjustment Facility EU European Union FAC Fonds d'aide et de coop6ration (French Agency for Cooperation) FAPA Fermes agro-pastorales d'arrondissement (production farms) FINEX Investment budget entirely financed by donors. It represents an average of 80% to 85% of the total investment budget. FAO Food and Agricultural Organization of the United Nations GER gross enrollment rate GFFC gross formation of fixed capital GNF Guinean Franc IDA International Development Association IDB Islamic Development Bank IMF International Monetary Fund IRDP integrated rural development projects IFAD International Fund for Agricultural Development IRAG Institut de recherche agronomique de Guin&e (Institute of Agronomic Research of Guinea) JSI John Snow, Incorporated KfW Kreditanstalt flar Wiederaufbau (German development aid agency) LADP Letter of Agricultural Development Policy MAEF Ministre de I'agriculture, de l'61evage, des eaux et forets (Ministry of Agriculture, Livestock and Forests) MEPUFP Minist&re de 'ducation pr6universitaire et de la formation professionnelle (Ministry of Pre-University Education and Professional Training) MESRC Minist&re de l'6ducation sup6rieure, de la recherche scientifique et de la culture (Ministry of Higher Education, Scientific Research and Culture) MOF Ministry of Finance MOH Ministry of Health MOP Ministry of Plan NER net enrollment rate NGO non-governmental organization O&M operation and maintenance OGUIB Office guinden du bois (Wood Office) OPEC Organization of Petroleum Exporting Countries ORS oral rehydration salt PASE Programme d'ajustement sectoriel dans 1'6ducation (Program of Sectoral Adjustment in Education) PDR Projet de d6veloppement rural (rural development project) PE public enterprise PER Public Expenditure Review PFP Policy Framework Paper PHC primary health care PIP Public Investment Program PNIR Projet national pour l'infrastructure rural (National Rural Infrastructure Project) PRSE Projet de restructuration du secteur de 1'61evage (Livestock Sector Rehabilitation Project) RC Relance Caf6 SEATS Family Planning Service Expansion and Technical Support (USAID project managed by John Snow, International, Arlington, VA, USA) SFD Saudi Fund for Development SNAPE Service national d'am6nagement des points d'eau (National Rural Water Supply Service) SNPRV Service national de promotion rurale et de vulgarisation agricole (National Directorate of Agricultural Extension) SNSA Service national des statistiques agricoles (Agricultural Statistics Office - MAEF) SOGETRAG Soci6t6 g6n6rale des transports de Guin6e (National Transport Company) SOGUIPAH Soci6t6 guindenne palmier i huile et d'hv&6as (National Palm Oil and Rubber Tree Company) SOTELGUI Soci6t6 de t6ldcommunication guin6enne (National Telecommunications Company) SPA Special Program for Africa TBA traditional birth attendant TOFE Tableau des op6rations financi&res de l'Etat UNCDF United Nations Capital Development Fund UNDP United Nations Development Programme UNICEF United Nations Children's Fund USAID U.S. Agency for International Development VET Vocational Education and Training WHO World Health Organization FOREWORD This report is based on the findings of a multi-donor mission and a Bank mission that visited Guinea respectively in February and July 1995. Bank staff on the mission were Ms. Pascale Kervyn (Task Manager), Ms. Carmen Martinez (Sr. Country Officer), Mr. Robert Franco (Sr. Country Economist and Mission Leader), Mr. Mathurin Gbetibouo (Agricultural Economist), and Ms. Christine Soulier (Research Analyst). Other mission members were Mr. Christian Girier (Consultant, FAC) and Ms. Florence Pasnick (Health Economist, EU Consultant). Mr. Frangois Laporte (Lead Economist) participated in the mission for a few days. The Government Team that worked with the mission was led by Mr. Alhassane Balde, Secretary General of the Ministry of Finance and President of the PER Coordination Committee. Other Bank staff who made specific contributions to the substance of this report are Mr. Sergiu Luculescu, Mr. Prabat Jha (health), and Ms. Julia Dayton (health and education); Ms. Catherine Laurent (education); Mr. Mohamed Ben-Senia (agriculture); Ms. H616ne Albouy (Operations Analyst); and Ms. Yasemin Yucelik (Research Analyst). Helpful comments were also made by Mr. Hasan Tuluy and Mr. Alain Thery. Mr. Roger Key and Mr. Robert Myers served as Peer Reviewers. Mr. Sanjay Pradhan (PRDPE) was Lead Advisor. Mr. Jean- Louis Sarbib is the Department Director, Mr. Frangois Laporte the Lead Economist, and Mr. Birger Fredriksen the managing Division Chief. Secretarial support was provided by Mr. Mather Pfeiffenberger, Ms. Atossa Shafa, and Ms. Song Li Ting Fong. PRELIMINARY NOTE This first and belated Public Expenditure Review (PER) exercise for Guinea was prepared mainly by World Bank staff with assistance from the French Ministry of Cooperation, the European Union, and the UNDP Resident Mission in Conakry. (The latter was particularly helpful in contributing to the budgetary projections by sector.) At this stage, the study consists of an analysis of the problems confronting the budgetary and expenditure processes in Guinea, and does not go into the details of the policies that would be necessary to enhance the efficiency of expenditure allocations. Unfortunately, for the preparation of the diagnosis it did not prove possible to have the desired collaboration with the Guinean authorities, given that the PER Committee that the Ministry of Finance had organized did not materialize. As such, the PER reflects the Bank's analyses and positions. However, given the controversial nature of the topic, the authorities, who are intimately involved in budgetary procedures, may question in some cases the data and the analyses itself. This is to be expected; nevertheless, the Bank staff hope that this document will serve as the basis for open and frank discussions with the Government, which would improve its accuracy, quality, and objectivity. Ultimately, the PER is meant to become a regular home- bred exercise allowing for the thorough review of all sectors in several iterations. CONTENTS EX E C U TIV E SU M M A R Y ...........................................................................................................i IN T R O D U C T IO N ..........................................................................................................................1 CHAPTER I: THE MACROECONOMIC ENVIRONMENT .................................................2 A . IS S U E S ..........................................................................................................................2 B. THE REFORMS: ATTEMPTS TO IMPROVE THE ECONOMIC AND FISCAL FR A M E W O R K ............................................................................................................3 A N onviable Fiscal Fram ew ork...............................................................................4 CHAPTER II: BUDGETARY TRENDS AND ISSUES ............................................................8 A. ISSUES .....8 B. INCREASING IMBALANCES BETWEEN O&M AND WAGE RELATED EX PEN D ITU R E S.........................................................................................................8 C. THE STEEP DECLINE IN CAPITAL EXPENDITURES AND THE RECURREN T COST CRISIS...................................................................................10 Government's Allocation Preferences...................................................................12 D. DONORS' RESPONSES TO PUBLIC EXPENDITURE ISSUES............................13 The Rationale for Domestic Contributions to Donor-Supported Projects.............13 W hat A re the O ptions?...........................................................................................15 CHAPTER HI: PUBLIC EXPENDITURES ON HEALTH, EDUCATION, AND AGRICULTURE AND INTERSECTORAL ALLOCATIONS..............................................16 A . IN TR O D U CTIO N .......................................................................................................16 B . K E Y ISSU E S ......................................................................................................6........16 C. FUNCTIONAL COMPOSITION OF EXPENDITURES: DISPARITIES BY SE C T O R ....................................................................................................................16 D. PUBLIC EXPENDITURES ON HEALTH......................................................E1.......18 K ey Issues............................................................................................................. 8 Health Status and Health Sector Policy ..........................................................oy.....19 Organization and Performance of the Health Sector..............................................19 What Does The Government Spend on the Health Sector?................................20 Economic Composition of Budget and Expenditures............................................20 Functional Composition of Expenditures.............................................................21 Who Uses Public Health Services?......................................................................22 The Equity of Public Health Expenditures: Do They Benefit The Poor?.............23 How Well Targeted Are Public Health Expenditures?..........................................23 The Burden of Health Care Expenditures: On the Government or on H ouseholds?................................................................................................25 Cost Recovery and the Burden of Household Spending for Public Health Services on the Poor.............................................................................................25 Policy O ptions........................................................................................................26 E. PUBLIC EXPENDITURES IN THE EDUCATION SECTOR ..................................27 K ey Issues..............................................................................................................27 Education Outcomes: Enrollment and Completion Rates ....................................28 Higher Education: Quality and Cost-Effectiveness ..............................................29 An Assessment of the Government's Education Strategy .....................................29 Cost Implications of the Government's GER Strategy..........................................30 What Does the Government Spend on Education?................................................30 Equity of Public Expenditures on Education Among the Population....................32 Contribution of Households to Financing Public Education .................................34 The Burden of Education Expenditures on Households........................................34 Policy O ptions........................................................................................................34 F. PUBLIC EXPENDITURES IN AGRICULTURE.......................................................35 Introduction ............................................................................................................35 M ain Issues ............................................................................................................35 C urrent O bjectives .................................................................................................36 Sectoral Performance and Contribution to Overall Growth...................................37 Sectoral Performance and State of Public Expenditures in Agriculture................38 Current State of Public Expenditures in Agriculture .............................................38 Overview and Trends of Public Expenditures in Agriculture................................38 Economic Composition of Public Expenditures in Agriculture ............................39 Functional Composition of Public Expenditures in Agriculture............................41 Policy O ptions........................................................................................................43 G. INTERSECTORAL ALLOCATIONS ........................................................................44 Costing Sectoral Strategies....................................................................................45 CHAPTER IV: THE INSTITUTIONAL ENVIRONMENT AND THE BUDGETARY P R O C E S S .....................................................................................................................................47 A. THE CURRENT INSTITUTIONAL STRUCTURE: NUMEROUS AND OVERLAPPING FUNCTIONS ................................................................................48 B. CHECKS AND BALANCES: ACCOUNTABILITY, TECHNICAL EXPERTISE, AND MONITORING CAPACITY ...................................................49 C. THE BUDGET PREPARATION PROCESS..............................................................50 The Budget Process and Cycle ..............................................................................51 Macroeconomic and Sectoral Strategies................................................................52 Preparation of the Current Budget.........................................................................53 Preparation of the Investment Budget....................................................................53 D. BUDGET IMPLEMENTATION AND MONITORING ............................................54 The Time-Consuming Budget Implementation Process........................................54 The Current Budget Implementation .........................................................56. PIP Budget Implementation: "Les 6tapes du Tour de Guin6e" ................56 Incentives for the Line Ministries to Improve Their Management Skills..............57 E. CURRENT BUDGET PROBLEMS WITH CIVIL SERVICE MANAGEMENT A N D PEN SIONS.................................................. .............. 58 F. BUDGET MANAGEMENT AND INFORMATION SYSTEMS...............................59 CHAPTER V: POLICY AND STRATEGIC OPTIONS FOR INSTITUTIONAL REFORMS AND BUDGETARY PROCESS ............................................................................61 A . IS SU E S ........................................................................................................................61 B. WHY SHOULD THE GOVERNMENT RAISE REVENUE? ...................................62 The Importance of Raising Revenue......................................................................62 The Importance of Managing Mining Revenue Uncertainty .................................64 C. PRIVATE SECTOR DEVELOPMENT AND DIVERSIFICATION.........................64 How Could the Government Strengthen an Enabling Environment for Private Sector Investm ent?...............................................................................................65 D. THE ROLE OF THE PUBLIC SECTOR....................................................................66 E. EXPENDITURE ALLOCATION: HOW CAN THE GOVERNMENT SOLVE THE RECURRENT COST CRISIS? ........................................................................ 68 Capital Expenditures are Unsustainable ................................................................68 First Option: M obilize Revenue............................................................................ 68 Second Option: Cut Capital Expenditures ............................................................69 Reallocation of Capital Expenditures ....................................................................69 Reallocation of Recurrent Expenditures Towards O&M.......................................70 F. CRITERIA FOR INTERSECTORAL ALLOCATIONS: WHAT ARE THE O P T IONS ? .................................................................................................................7 1 G. POLICY AND STRATEGIC OPTIONS FOR INSTITUTIONAL REFORMS AND THE BUDGETARY PROCESS ......................................................................72 Institutional R eform ...............................................................................................73 C hecks and B alances..............................................................................................73 B udget Planning .....................................................................................................73 C ivil Service R eform ..............................................................................................74 Restructuring Pension Program s............................................................................75 B udget Im plem entation..........................................................................................75 C onclusion .............................................................................................................75 B IBL O G R A PH Y ....................................................................................................................... 76 A N N E X E S ................................................................................................................................... 79 Annex 1 Selected Indicators for the Guinean Economy and Public Expenditures.................... 81 Annex 1 Table 1.1 GDP in Constant 1989 Prices, 1988-1994 ........................................ 81 Annex 1 Table 1.2 GDP in Current Market Prices, 1988-1994....................................... 83 Annex 1 Table 1.3 Selected Indicators, 1988-1994......................................................... 84 Annex 1 Table 1.4 Savings and Investment Balances, 1988-1994.................................. 85 Annex 1 Table 1.5 Consumer Price Index, 1988-1994.................................................... 86 Annex 1 Table 1.6 Central Government Revenue, 1988-1994........................................ 87 Annex 1 Table 1.7 Central Government Expenditure, 1988-1994.................................. 88 Annex 1 Table 1.8 Central Government Financial Operations, 1988-1994.................... 89 Annex 1 Table 1.9 M onetary Survey, 1988-1994............................................................ 90 Annex 1 Table 1.10 Balance of Payments, 1988-1994...................................................... 91 Annex 1 Table 1.11 External Debt Service, 1988-1994.................................................... 92 Annex 1 Table 1.12 Sectoral Composition of Investment Budget (constant prices) and Investment Expenditures (current prices), 1988-1994................... 93 Annex 1 Table 1.13 Sectoral Composition of Investment Expenditures, 1988-1994 (constant 1989 pric s) .................................................................... 94 Annex 1 Table 1.14 Sectoral Composition of Current Budget, 1988-1994 (constant 1989 prices) .................................................................... 95 Annex 1 Table 1.15 Sectoral Composition of Current Expenditures, 1988-1994 (constant 1989 prices) .................................................................... 96 Annex I Table 1.16 Sectoral Composition of Total Budget, 1988-1994 (current prices) ............................................................................... 97 Annex 1 Table 1.17 Sectoral Composition of Total Budget, 1988-1994 (constant prices) ............................................................................. 98 Annex I Table 1.18 Sectoral Composition of Total Expenditures (current prices)..........99 Annex 1 Table 1.19 Tax-GDP Ratio in Selected SPA Countries (excluding grants)...... 100 Annex 2 Assumptions and Analysis of the Unsustainable Scenario.......................................... 101 Annex 2 Table 2.1 Terms of Trade Projections: 1994-2000 ........................................ 105 Annex 2 Table 2.2 Government Financial Operations Projections: 1994-2000........... 105 Annex 2 Table 2.3 Financing Simulations..................................................................... 106 Annex 2 Table 2.4 Principal Scenario Assumptions, 1995-2000.................................. 106 Annex 2 Table 2.5 Central Government Revenue Projections and Expenditures, 1994-2000 ..................................................................................... 107 Annex 2 Table 2.6 Mining-Based Revenue Projections, 1994-2000 ............................ 108 Annex 2 Table 2.7 External Debt Projections, 1994-2000............................................ 108 Annex 3A Calculation of Required Recurrent Expenditures and Assessment of the R C oefficien ................................................................................................................... 09 Annex 3a Table 3.1 The R Coefficient by Sector............................................................ 10 Annex 3a Table 3.2 Recurrent Cost Estimates: the PER vs. a Road Project ................. 12 Annex 3a Table 3.3 Two Methods of Recurrent Cost Calculation ................................. 113 Annex 3a Table 3.4 R Coefficient: 7% ........................................................................... 115 Annex 3a Table 3.5 Increase in the Share of Capital Expenditures in the PIP (R Coeficient: 14% )................................................................... 115 Annex 3a Table 3.6 Variations in the Sector Weight of the R Coefficient..................... 117 Annex 3 Calculation of Recurrent Costs...................................................................................... 118 Annex 3 Table 3.1 Recurrent Costs Generated by Capital Expenditures: PFP Case ... 118 Annex 3 Table 3.2 Recurrent Costs Generated by Capital Expenditures: Case IA and C ase I B .................................................................................. 119 Annex 3 Table 3.3 Recurrent Costs Generated by Capital Expenditures: Case 2A..... 121 Annex 4 Financing Requirements for Periodic and Routine Road Maintenance, 1995-2000 ...................................................................................................................... 122 Annex 5 Costing of Priority Programs in Health Strategy for Year 2000................................ 124 Annex 5 Table 5.1 Costing of 7 Priority Programs of Health Strategy by Year 2000.. 124 Annex 5 Table 5.2 Expenditures by Level of Care........................................................ 125 Annex 6 Expenditure Incidence Analysis on the Poor in the Health Sector............................. 126 Annex 6 Table 6.1 Health Sector Budget Allocations: 1988-1995 .............................. 127 Annex 6 Figure 6.1 Health Facility Use for Illness by Quintile: 1994.......................... 128 Annex 6 Table 6.2 Rate of Illness and Rate of Seeking Treatment............................... 129 Annex 6 Table 6.3 Utilization of Outpatient Public Health Care Facilities for Preventive and Curative Services, by Quintile ............................ 129 Annex 6 Table 6.4 Recurrent Cost Per Visit for Public Health Care by Region: 1994 .............................................................................................. 130 Annex 6 Table 6.5 Per Capita Health Public Subsidy: 1994........................................ 131 Annex 6 Table 6.6 Cost Ratios per Inpatient/Outpatient Visit, Results from Monou Prefectural H ospital: 1994 .......................................................... 131 Annex 6 Figure 6.2 Distribution of Public Health Expenditures: 1994......................... 132 Annex 6 Table 6.7 Self-Reported Costs of Public and Private Health Care Services Per Consultation: 1994................................................................ 133 Annex 6 Table 6.8 Self-Reported Costs of Public and Private Health Care Services Per Consultation: 1994................................................................ 133 Annex 7 Costing of the Government's Expenditures in Education........................................... 134 Annex 7 Table 7.1 Gross Primary Enrollment Rates, By Region: 1992-93 and 1993-94 ................................................................... 134 Annex-7 Table 7.2 Costing of Government's Strategy for Primary Education by Year 2000: Project Scenario ....................................................... 137 Annex 7 Table 7.3 Costing of Government's Strategy for Primary Education by Year 2000: Government Scenario............................................... 141 Annex 7 Table 7.4 Education Budget: Economic Composition................................... 145 Annex 7 Table 7.5 Expenditures for Education: Economic Composition.................... 146 Annex 8 Poverty Incidence Analysis in Education...................................................................... 147 Annex 8 Table 8.1 Per Student Public Education Expenditures ................................... 148 Annex 8 Table 8.2 Results of Alternative Methods of Incidence Analysis................... 149 Annex 8 Figure 8.1 Per Capita Distribution of Public Expenditures on Education by R egion .......................................................................................... 150 Annex 8 Figure 8.2 Comparison of Relative and Absolute Distribution of Public Expenditures on Education: 1994............................................... 51 Annex 8 Table 8.3 Per Student Household Expenditures on Public Primary and Secondary Education: 1993-1994............................................... 51 Annex 8 Table 8.4 The Burden of Household Spending on Public Education ............. I52 A nnex 9 A griculture....................................................................................................................... 153 Annex 9 Figure 9.1 Agricultural GDP Growth (1988-92).............................................. 155 Annex 9 Figure 9.2 Index of Selected Performance Indicators and Public Expenditures in iAgriculture......................................................... 156 Annex 9 Figure 9.3 Public Investment Program in Rural Development, 1991-94......... 156 Annex 9 Table 9.1 Rural Roads Rehabilitated, 1990-1995........................................... 157 Annex 9 Table 9.2 Public Expenditure in Rural Infrastructure by Source of Funding and Category of Expenditure, 1990-96........................................ 158 Annex 9 Figure 9.4 Shares of Public Expenditures in Rural Infrastructure by Source of Funding and M ajor Categories..................................................... 158 Annex 9 Table 9.3 Domestic and Foreign Financing of Agricultural Extension, 1990-94 ....................................................................................... I60 Annex 9 Table 9.4 Functional Composition of Investment Expenditures in Livestock, 1992-94 ....................................................................................... 160 Annex 9 Table 9.5 Economic Composition of Public Expenditures in Livestock Services ....... ................................................................................ 16 1 Annex 9 Table 9.6 Ministry of Agriculture, Livestock and Forests: Organizational Flow -C hart ................................................................................... 162 Annex 9 Table 9.7 Volume of Foreign Assistance to Guinean Agriculture, 1995........ 163 Annex 10 The Impact of the Garafiri/Kaleta Project on Public Expenditures........................ 164 Annex 10 Table 10.1 Impact of the Garafiri/Kaleta Project on Public Expenditures ....... 166 Annex 11 The Institutional Environment and the Budgetary Process...................................... 167 Annex 12 The Administration and Control of Large Projects (ACGP) ................................... 171 EXECUTIVE SUMMARY I. This is the first Public Expenditure Review (PER) for Guinea. The report was prepared by the Western Africa Department, with the support of several donors. A counterpart team in the Government assisted the Bank in collecting data and participated in the discussions on the PER preliminary conclusions. Over time, it is hoped that the PER would become a regular home-bred exercise and a tool for monitoring public resource management and managing dialogue with donors. Background 2. Well endowed with rich soil, abundant rainfall, and a large array of minerals, Guinea has the bounty of natural resources which should facilitate achievement of economic prosperity. But the country has inherited the legacy of 25 years of mismanagement by the previous regime, setting back the clock on its economic and social development. A decade ago, the new regime undertook sweeping economic reforms, an effort supported vigorously by the donor community. Public investment increased by almost 200% in real terms between 1986 and 1990. Reforms were implemented in several key areas -- most notably the liberalization of the monetary and financial sectors -- and yielded impressive results: the progressive control of inflation and overall improvement in economic performance. 3. However, reforms came short of securing the complete transition to a market economy. Public sector involvement in productive and commercial activities remains substantial (around 10% of the public investment budget) and a sizable part of economic activity and employment is still provided by the public sector. In addition, reforms had little success in spurring private sector development. As a result, the economy still relies excessively on bauxite, which accounted for 80% of exports, 20% of GDP, and 30% of fiscal revenue in 1994. 4. Public management remained hampered by an opaque institutional environment, insufficient accountability and transparency in the budget process, cumbersome procedures and poor incentive structures. Inefficiencies in public expenditure allocations have prevented meaningful alleviation of the burden of poverty. As a result, poverty and malnutrition remain prevalent. An estimated 48% of the population still lives on less than US$1 daily, and one child in four dies before reaching 5 years of age. The Revenue Crisis Has Led to Unsustainable Macroeconomic Imbalances 5. Guinea suffered from a severe terms of trade shock as the price of bauxite declined by 33% between 1990 and 1994. Because the country was insufficiently prepared for the revenue shortfall, fiscal management deteriorated rapidly and precipitated a severe economic crisis. The formal private sector shrunk, and private investment outside of mining fell to 7.4% of non-mining GDP in 1994 from 9.2% in 1988. Total investment (outside of mining) in Guinea is now at 15% of nonmining GDP (against 16.6% in SSA in 1993).1 These levels of investment threaten the economy's ability to renew its capital stock. 6. Because of its limited tax culture, Guinea's tax effort in the nonmining sector could not outweigh the decline of mining revenue. Consequently, overall revenue fell from 15.8% of GDP in Excluding Nigeria and South Africa. 1990 to 10.4% in 1994. However, Guinea's revenue performance is at odds with its revenue mobilization potential and is far short of the median revenue performance of SPA2 countries between 1991-1993 (14.6%), and in particular of other African countries which also depend on mining. 7. Expenditure adjustment from 24.6% of GDP in 1990 to 17.5% in 1994 has not been enough to prevent a progressive increase in arrears (to 15% of revenue in 1994); external borrowing to- support expenditures has significantly increased debt service on external and domestic debt (including interest and amortization) in the past two years, as a percentage of exports (FOB), from 32% in 1992 to 42% in 1994 before rescheduling; this large debt service burden is now comparable to the Sub-Saharan Africa average of 43% in 1993 (before.rescheduling). Current levels of public expenditures are now far lower in Guinea than in Sub-Saharan Africa (30% of GDP in 1993) and are inadequate to deliver services and promote development. 8. Thus today, Guinea faces the substantial challenge of enhancing domestic revenue mobilization and achieving fiscal viability in the face of likely restrictions in external assistance for the foreseeable future, the fragility of the debt situation, and the cyclical nature of mining proceeds. In that context, the 11% real growth in domestic revenue that the Government is committed to in the context of the Policy Framework Paper? would represent a radical departure from trends in the past five years. If PFP revenue targets were missed, one of two scenarios could happen, neither one desirable. First, if expenditures do not adjust to the shortfall in revenue, unsustainable budgetary deficits would trigger a spiral of recessionary effects on the economy, with new accumulation of arrears, the loss of the public sector's borrowing capacity on concessional terms, and further disincentives to private investment. Second, if expenditures do adjust so as to sustain the deficit, their growth rate would be limited to 2.3%. These levels would be insufficient to maintain minimum public expenditure allocations for social and other high-priority sectors needed to permit the economy to function. Uneven Expenditure Adjustment Has Crowded Out Operation and Maintenance Expenditures 9. The revenue shortfall, associated with shifts in expenditure allocations, has led to a "recurrent cost crisis", primarily reflected in the 43% decline (in real terms) of the share of O&M expenditures in the current budget, from 36% of total current expenditures in 1990 to 24% in 1994. The administration's ability to deliver services to the population is now seriously impaired. The most disturbing examples are hospitals without materials or drugs, as witnessed, for example, in hospitals in Kindia or Dalaba (per capita allocation for drugs is about one fourth the average in Sub- Saharan Africa). Another consequence is the decline in the return on investment, as recurrent cost requirements generated by investments are not met. For instance, the deterioration of the road network caused by inadequate funding for road maintenance (and aggravated by Guinea's dramatic climatic conditions) has prompted donors to undertake vast rehabilitation programs. Between 1988 and 1993, they spent about US$32.5 million (more than GNF 31 billion) to rehabilitate around 430 kilometers under the Transport Sector Project. This amount is exorbitant when compared with the estimated GNF 19 billion necessary for routine and periodic maintenance of the entire 18,000- kilometer network in 1994. 2 The SPA or "Special Program for Africa" is an assistance fund set up in the early 1980s to finance balance of payments support on concessional terms to the poorest African countries. The Policy Framework Paper (PFP) is the medium-term macroeconomic framework (1995-1998) for Government policies, encompassing balanced monetary, fiscal, external, and real sector accounts. It is prepared jointly by the Government, the IMF, and the World Bank. 11 10. Four factors contribute to chronic O&M underfunding. The first is the unsustainable rise in the wage bill and pensions. Wages and transfers rose from 29% of revenue in 1990 to 51% in 1994. Employment and promotion measures as well as a generous pension system raise serious concern about sustainability of continuing present trends. 11. A second factor is the Government's preference for allocating its own resources to capital expenditures rather than to contributions to externally-financed projects, resulting in delays in the implementation of projects, and thus hampering their economic impact. In 1994, a further revenue . decline gave a new urgency to this prioritization issue, since total BND disbursements would have been insufficient to cover contributions to FINEX. 12. A third is the increasing external financing of O&M as donors have understandably given priority to ensuring maximum impact of the projects they are financing and continuity of service to their stakeholders' welfare. In the short run, this has alleviated implementation problems, but has also encouraged further O&M underfunding and underspending by the Government. In the longer term, the build-up of unsustainable O&M financing by donors constitutes a danger for the viability of investments. 13. A fourth factor is the separate preparation of current and investment budgets by the Ministries of Finance and Plan, without mutual consultation. The complete absence of analytical links between capital and recurrent expenditures has obscured the fact that the Government is trying to do too much with its available resources. Capital Expenditures Are Unsustainable 14. The.sad irony is that because O&M financing has been gravely neglected in the past, the current level of capital expenditures, already abysmally low, has also become unsustainable. Recurrent cost requirements (estimated at 14% of capital expenditures) generated by investments are incompatible with the macroeconomic constraints on expenditure growth (6.9% in the PFP). The implication is that capital expenditure projections in the PFP are unsustainable. The "recurrent cost" gap would end up at a minimum of 23% of requirements in year 2000 (figure 1). 15. Thus far, public capital expenditures have borne the brunt of expenditure adjustment, declining from 8.5% of GDP in 1990 to 4% in 19944, well below levels in Sub-Saharan countries. Therefore, a serious question pertains to whether further cuts in public capital spending would bring domestic investment to the point at which the depletion of capital would jeopardize developmental perspectives for the long term. Indeed, this risk argues strongly for refocusing Government's role on priority sectors and shedding wasteful and nonpriority projects. To alleviate the recurrent cost crisis and minimize capital expenditure cuts, medium-term priority should be given to a reallocation within recurrent expenditures, through a wage bill cut and a reform of the pensions system. Ministry of Plan data. iii Figure 1 Actual capital and recurrent expenditures (1989-1994); PFP and IMF projections for capital and recurrent expenditures (1995-2000); Recurrent expenditure requirements (in constant GNF; 1989=100) 400. 350 ap 300 200- - 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 - capital expenditures -U-recurrent expenditure requirements -0-recurrent actual Inter- and Intrasectoral Composition of Expenditures Reveals Persistently Low Allocations to Basic Services and Substantial Allocations to Tertiary Programs and in Productive Sectors 16. The salient feature of functional expenditures is the very low level of resources devoted to the social sectors, at an average per capita of US$3.1 for health and US$8.5 for education between 1988 and 1994 in 1989 US dollars. These allocations are about a third of an average of 21 Sub- Saharan African countries in 1993. There is also a consistent bias in favor of agriculture at an average of 17% of total expenditures, which far outweighs allocations in Sub-Saharan Africa (6%). Similarly, infrastructure has been allocated large expenditures, reflecting efforts to rehabilitate roads and other infrastructure after 1985. 17. To some extent, health indicators in Guinea, among the worst in the world, reflect inadequate allocations to the sector. But more meaningful are the increased allocations to tertiary health care at the expense of primary care, and particularly preventive care (from 34% of health expenditures in 1991 to 22% in 1994). Expenditures on education nearly doubled in real terms between 1988 and 1991. However, since then, they have declined by 26% and were at 14% of total expenditures in 1994. Notwithstanding significant increases in gross primary school enrollment (GER) rates since 1989 -- from 29% in 1989 to 40% in 1994 -- primary education has not received its proper share of allocations, declining from 50% of the total education budget in 1991 to 42% in 1993. 18. The functional composition of social expenditures raises equity and poverty alleviation issues. Tertiary programs benefit almost exclusively the wealthiest portion of the population. Hence, expenditure allocations reflect the most unequal distribution of health and education expenditures among eight Sub-Saharan African countries. For example, in education, the poorest quintile receives 5% of the subsidy and the wealthiest receives 44%. In CMte d'Ivoire, these allocations are, respectively, 10% and 37%; in Ghana, they are 16% and 37%. In the case of health, the introduction of cost recovery for primary health care has further increased social inequality. In fact, not only does the poorest quintile pay more than the wealthiest for a visit to a public health unit (including costs of transport and wait period), but public health care is relatively more expensive for the poor than private care is for the rich. iv 19. Agriculture is the second most important and the most promising productive sector of the economy after mining. It achieved a commendable 4% growth rate during the period 1988-94, at the expense, however, of environmental and economic sustainability. While the share of agriculture in the budget is high, the effectiveness of expenditures and investment programs in agriculture appears low and the sustainability of some programs is uncertain. An analysis of the effectiveness of the whole agricultural sector expenditure program is in order, with a view to better determining the respective roles of the public and private sectors and core public services. This entails data and analytical skills that are lacking today at the Ministry of Agriculture (MAEF). Agriculture is perhaps the most symptomatic of the recurrent cost crisis facing Guinea's public sector: 98% of the MAEF's 1994 budget was devoted to salaries. Financing of Strategies in Priority Programs Would Demand Difficult Decisions on Resource Allocations 20. Scant allocations to basic services and the large and inefficient absorption of resources by agriculture argue for improving the efficiency of inter- and intrasectoral allocations towards cost- effective programs and investments that yield high returns. A critical condition for making knowledgeable intersectoral choices is the existence of realistic and affordable long-term strategies in key sectors of the economy and the compatibility of these strategies with the macroeconomic framework. 21. To illustrate this issue, the PER evaluated the cost of road maintenance requirements and the Government's strategies in primary education and basic health by year 2000. The results indicate that current expenditures in these sectors would absorb- 32% of total current expenditures -- more than twice their 1994 share in current expenditures -- if revenue growth meets PFP targets (the "tax effort" scenario). If poor revenue performance limits expenditure growth to 2.3% annually (see para. 8), current expenditures for meeting the sectoral objectives would amount to 45% of the sustainable level of current expenditures in year 2000. This would represent three times the 1994 allocation to these sectors. These projections are presented in figure 2. 22. Regardless of revenue performance, it is clear that the pursuit of the Government's objectives in each of these three sectors would require a major reallocation of resources towards these sectors. Such a reallocation would require clarification and common understanding within the Government and the donor community about Guinea's long-term development strategy, poverty alleviation objectives, and the contribution of the sectors towards these objectives. 23. If the Government decides to finance its strategies in basic health and primary education and in road network maintenance, it would have to generate savings in other expenditures. The principles guiding this process would be those with which the Government adopted the liberalization option in 1985: the withdrawal of the public sector from productive and commercial activities, and allocations of public resources to basic services when market failure prevents private sector activity (for example, basic education and health). In addition, savings should be sought by rationalizing the number of Government's agencies and staff in some ministries. Resources should be reallocated within sectors as well. For example, in education, savings (amounting to GNF 1.5 to 2.0 billion annually) could be generated by rationalizing higher education. v A Complex Institutional Environment 24. The institutional landscape consists of a Figure 2 large number of institutions with overlapping Costs of sectoral priorities mandates. Long-time tensions between the Ministry of Plan and the Ministry of Finance for 1994 - Current expenditures public resource control, particularly the excluding debt service investment budget, have manifested themselves 85% in an increasing number of intervening %controllers in the expenditure process, and in significant delays in procurement (up to one year for important bids). To improve disbursement 2% performance on the PIP, the new Agency for Education Control of Large Projects (ACGP) in 1994 was D Road maintenance (incl. rural roads) granted the broad mandate of controlling or * Others participating in all stages of projects above US$1 million and supervising the privatization of 2000 - Scenario 1: No tax efffort public enterprises. Although its mandate raises 21% some questions with regard to accountability and capacity building in the line agencies, its 10 performance will need to be assessed. 56% 25. Limited accountability has proven to be a 9% key constraint against the efficiency of ary Education institutions and processes. Thus, the two main BRoad maintenance (incl. rural roads) control agencies, the Auditor General and the EOthers Financial Control, have had little success in fostering accountability, as illustrated by the 2000 - Scenario 2: Tax effort estimated GNF 50 billion in unpaid fees that the 16% Treasury owes to the Chambre des Comptes due to delays in providing timely accounts. 10% However, it is hoped that the creation of a Legislative Assembly in 1995 will result in 67% promoting transparency and more effective sanctions. @Primary Education DHealth R 0 maintenance (i. rural roads) The Budget Process Is Not Conducive to Efficient Public Resource Management 26. The first formal Government budget was prepared in Guinea in 1988, thirty years after independence. Inefficiencies in the budget process have limited the budget's role as a tool for economic management. The country lacks a credible detailed medium-term macroeconomic framework: although the PFP provides for balanced macroeconomic accounts in a three-year time horizon, the Government's sectoral policies have not been quantified, nor has their consistency been tested against the expenditure framework, as discussed above. In that context, sectoral strategies cannot be budgeted. Coupled with the absence of well-established levels of sectoral spending and a clear analytical link between capital and recurrent expenditures, the preparation of the budget is an open-ended process. vi 27. The absence of a credible short-term macroeconomic framework to start the budget preparation cycle is due to the dearth of key information, such as comprehensive data on the previous year's budget performance, particularly for the PIP budget. As a result, significant changes in the budget draft -- several months after budget negotiations have been completed -- lead to discretionary reallocations. 28. Budget proposals are not subject to binding ceilings. The current budget proposals are prepared by the line ministries, which for the most part do not possess adequate tools for assessing their needs; budget proposals are commonly based on profiles from previous years, to which arbitrary requirements are added. For example, the 1995 current budget proposal by the Ministry of Agriculture justified some allocation requests based on "the reintegration of 150 civil servants from the Livestock Ministry" and "a fairly large number of promotions." 29. In theory, the investment budget preparation rests on a three-year rolling investment program. Because the MOP and the line ministries lack expertise in sectoral strategy and project evaluation, the criteria for including projects in the PIP are largely reduced to availability of external financing for FINEX. As a result, public investment programs reflect primarily the perceptions of donors about needs and preferences. The MOP's weakness becomes critical when marginal or politically motivated projects are proposed in the PIP. 30. The implementation of the budget is a long and painful process (the implementation of the investment budget takes up to forty steps); several institutional entities, some duplicating others, compete for the commitment of expenditures, the latest one being the ACGP. This process has created substantial delays in payments, with implications for the implementation of projects, and has promoted rent-seeking behavior. In addition to these delays, disbursements patterns result in altering allocation priorities. Efficient monitoring of budget implementation is also hampered by the absence of a comprehensive, integrated budget management system. 31. Concentrating decision-making at the level of the central agencies gives sectoral ministries and other agencies little incentive to improve their budget management skills. Final decisions about the budget are made by a handful of individuals in the central agencies. Line agencies have no control over and usually very little information on committing or disbursing their expenditures, and their autonomy has been eroded further by the 1995 imposition of quarterly ceilings on all budget items. They have limited capacity to supervise resource use in their sectoral investment. For example, only two inspection missions have reportedly been carried out by the Ministry of Health since 1990. The Challenges Ahead 32. To address the constraints on economic and social development and bring the efforts of the last ten years to fruition, Guinea faces an ambitious reform agenda. Guinea's first challenge in order to secure balanced growth will be to achieve fiscal viability through significant and durable revenue mobilization efforts. 33. A second challenge is to lift constraints on the diversification and growth of the economy by providing the private sector with a predictable and stable incentive framework. In the context of a liberalized economic system, the role of the public sector needs to be geared towards provision of basic services and maintenance of public infrastructure. In particular, the public sector needs to withdraw from activities where the private sector has a comparative advantage, including activities in the productive and commercial sector. vii 34. Indeed, the budgetary analysis provides evidence that the public sector is involved in more activities than it can sustain financially, leading to poor-quality service and a low return on investment. Rigorous attention to recurrent costs issues, notably through joint preparation of the current and investment budgets, would clarify the Government's financial constraints and the need for defining the public sector's role in the economy. 35. Yet another challenge is to enhance considerably the efficiency of public resource management, including rationalizing allocation decisions, in order to address priority sector needs. Securing accountability and transparency of public expenditure management will be critical. The sectoral and intersectoral analysis indicates that current levels of resources are insufficient to meet the basic needs of the population, maintain the existing infrastructure, and promote a more efficient civil service. In particular, basic services must receive considerably more resources if Guinea wants to reduce poverty and malnutrition in a significant and sustained manner. Hence, raising additional revenue is a sine qua non for adequate public services provision and development. In addition, even if revenue mobilization is increased to the levels projected in the PFP, inter- and intrasectoral choices will have to be reoriented towards efficiency-based allocations of public expenditures, including adequate endowments to O&M. Refocusing the public sector's activities away from productive and commercial activities will be necessary to enhance the quality of services to key sectors. Government and donors will need to enhance coordination and understanding on and commitment to overall development objectives and key priorities. 36. Most importantly, the public sector will need to alleviate constraints on a time-consuming budgetary process and a complex institutional environment. In that context, the roles and accountability of institutions and individuals must be- greatly enhanced. As the public sector delineates its functions, public resource management will need to be geared towards objectives and outcomes. 37. Removing constraints on the efficiency of public expenditure management and improving the institutional environment will require difficult choices from the Government. However, only in- depth reforms towards improving public resource management will bear results and hasten the development process. A favorable outlook predicted for the bauxite sector offers a renewed opportunity to consolidate previous gains and deepen reforms. As demonstrated by experience in the past five years, insufficient preparation for a downturn in revenue can start a downward spiral in fiscal management, with severe setbacks to economic and social progress. 38. Provided the Government succeeds in enhancing the efficiency of its resource management and developing a predictable and consistent framework for the development of the private sector, Guinea could capitalize on past efforts and reap the benefits of its enormous potential thereby lifting its population from poverty. viii INTRODUCTION I. The PER process is innovative in Guinea. The participatory approach has been selected to promote the Government's ownership of a reform agenda in public resource management. The foundation for this collaboration with the Government has been laid during the successive missions. In early 1996, a number of workshops will be held with EDI's assistance. The participants, who will be technicians from the relevant ministries, will discuss the PER diagnosis and resulting policy options. The purpose of the January/February 1996 mission is to discuss this preliminary diagnosis and, in the context of the workshops, exchange views on the policies required to render more efficient the budgetary process in general and expenditure allocations in particular. 2. This PER will be an important part of the macroeconomic dialogue with Guinea, as issues addressed in this report call for in-depth reforms in the management of public resources and the role of the Government in the economy. The adoption of an action plan resulting from a consensus on the PER diagnosis is viewed as crucial for the deepening of macroeconomic reforms in Guinea. To that effect, implementation of the action plan will be an important component of the next adjustment operation in Guinea. 3. Guinea has come a long way since the beginning of the new regime in 1985. The efforts required by the drastic reforms after 1985 should not be minimized and, in many key areas, reforms have shown spectacular results. Such is the case in the monetary and financial sector, but also in the real sectors of the economy. However, over the last five years, the pace of reforms has slowed down while, at the same time, Guinea was starting to suffer from the decline of international prices of-bauxite. As a result, a number of key constraints still hinder durable improvement in economic growth and poverty alleviation. As Guinea embarks on the second decade of a market economy system, it has become crucial to renew adjustment-led growth. In that context, considerably enhanced public resource management will be key for the success of Guinea's refoirms. 4. Key among the challenges to be faced in public resource management are institutional weaknesses and inefficiencies affecting the budgetary process, notably poor accountability. Other constraints include weakness in macroeconomic and sector policy management and deficient allocation process and decisions. The array of issues to be addressed under a public expenditure review is very wide. Therefore, the scope of this first PER will be focused primarily on the key sectors and on institutional and process issues which are believed to be the most critical in the management of public expenditures. The analysis of the macroeconomic situation will be reviewed in chapter 1. In addition, the macroeconomic consequences of uncontrolled expenditures will be assessed and a level of expenditures compatible with sustainable deficits will be determined. Budgetary analysis in chapter 2 will review the imbalances affecting the economic composition of the budget. The inter- and intrasectoral analysis in chapter 3 will review expenditures in agriculture, health, and education, and the incidence of health and education policies on the poor. Intersectoral policy choices and allocation criteria will also be discussed. Process and institutional issues will be discussed in chapter 4. Policy and strategic options for addressing the main constraints on efficient resource management are presented in chapter 5. 1. THE MACROECONOMIC ENVIRONMENT Methodological remark 1. Accounting of current expenditures. The Government uses the macroeconomic framework developed in conLunction with the IMF. Accounting of current expenditures in the PER is based on data as presented in the TOFE.' 2. Accounting of investment expenditures. Two reference sources provide data on inmestment expenditures -- the TOFE and figures from the Ministry of Plan (MOP). There are significant data discrepancies between these two sources, explained by three main factors- The first and the most important is the existence of several projects (grants) managed directly by donors that do not appear in the Public Investment Program (PrP) but are recorded in the Balance of Payments and thus the TOFE. The second is that TOFE expenditures are calculated on a commitment basis, while MOP figures are calculated on a disbursement basis. Finally, coordination and information sharing between MOP and MOF is generally poor; thus, for example, the ministries use different figures for the exchange rate. For the discussion on expenditure trends in the macroeconomic section (chapter 1), data from the TOFE are used because they reflect the level of actual investment expenditures more accurately and in a manner consistent with the macroeconomic accounts. Data disaggregated by sector and by ministry are provided only by MOP, hence they serve as the basis for the discussion on PIP implementation and allocations (chapter 2 and beyond). But the rehability of these data is subject to uncertainty. A. ISSUES 1.1 In 1985, Guinea implemented a series of reforms that significantly improved the country's economic performance. But the reforms did not prove deep enough to support more sustained economic growth, foster the development and diversification of private sector activity, or recenter the public sector's activity towards basic services, notably by addressing its deteriorating social structure, particularly in health. As the Government had not developed a viable fiscal framework, it was insufficiently prepared for the brutal decline in the prices of bauxite and alumina in 1990. As a result, as activity in the nonmining sector has declined after 1990, the country has suffered a progressive shortfall in revenue, which has been exacerbated by excessive growth in domestic consumption and a sharp decline in public and private investment. Between 1990 and 1994, total revenue dropped by 25% in real terms. At 10.4% of GDP, Guinea's tax-GDP ratio is a little more than half the SSA average of 19.4% (1993). After 1990, the Government did pursue an expenditure adjustment policy to address its growing resource gap, but it was not enough to halt its declining primary surplus, growing arrears, and mounting debt burden. In short, Guinea's fiscal performance has deteriorated markedly during the past five years. In the context of the PFP negotiated in July/October 1995, the Government has developed a strategy for enhancing revenue mobilization. These targets are key to Guinea's fiscal viability, as uncertainty weighs on external financing, and economic and fiscal performance remain prey to fluctuations in Tableau des Opirations Financiares de l'Etat (TOFE) is an accounting framework for central Government expenditures developed and used by the IMF. 2 international prices of bauxite. The rationale for enhancing revenue mobilization and policies towards achieving this objectives are discussed in chapter 5. B. THE REFORMS: ATTEMPTS TO IMPROVE THE ECONOMIC AND FISCAL FRAMEWORK2 1.2 In 1985, the Second Republic undertook sweeping reforms to remove the main distortions in the economy and obstacles to private sector development. The Government embraced economic liberalization and reduced its intervention in economic activity. These reforms achieved spectacular successes in the following areas: * Exchange rate. Progressive liberalization of the exchange rate starting in 1985 and adoption of an interbank foreign exchange system in 1994. * Monetary policy. Control of inflation and adoption of indirect money management instruments, notably a treasury bill market. * Pricing. Liberalization of consumer and producer prices, except cotton. * Trade policy. The removal of quantitative restrictions and export licensing requirements; simplification of the structure of import duties. * Fiscal management. Introduction of a PIP in 1986 and adoption of the first budget in 1988. Development of a framework for the management of public resources. *- Public sector downsizing. Disengagement from some productive activities (e.g., marketing of agricultural products) and privatization or liquidation of over 100 public enterprises. * Financial sector. Restructuring of the Central Bank; gradual replacement of six insolvent state-owned banks with privately managed deposit-money banks. * Regulatory framework. The adoption of investment, labor, mining, and land tenure codes; improvement in the legal and judicial system. Revision of the investment code in July 1995 towards rationalization of import duty exemptions. 1.3 However, the country continued to rely too heavily on bauxite for economic growth and budget revenue, and the reforms were not deep enough to meet four critical objectives: * promotion of economic development through a viable fiscal environment that would enhance public resource management. This will be discussed in the current chapter. * refocusing the role of the public sector in the development process, with more efficient allocations of public resources towards basic services and away from 2 For a complete discussion on fiscal performance and macroeconomic stabilization in Guinea, see the Country Economic Memorandum (CEM), World Bank, October 1995 (particularly chapters 1 and 2). 3 activities in the productive sectors of the economy. This issue will be addressed primarily in chapters 2 and 3. * creation of an institutional environment enabling the public service to recenter its activities and improve the quality of public services. This issue is discussed in chapter 4. * support to private sector development and diversification in the nonmining sector through a stable and predictable regulatory framework. This issue will be discussed in chapter 5, together with an outline of a strategy for the development of the private sector. As a result, the sudden terms of trade shock precipitated the fiscal and economic crisis, with a progressive shortfall of public resources and a decline in private sector activity (and subsequent shrinking of the fiscal base). A Nonviable Fiscal Framework 1.4 Insufficient expansion of the nonmining revenue base. Guinea had made considerable strides in developing a viable fiscal base as attested by the 80% revenue increase between 1988 and 1990. In particular, nonmining revenue grew by almost 90% during the period, albeit from a low basis. However, durable improvement in revenue mobilization was hampered by the absence of a taxation culture, in which heavy dependency on indirect taxation stifled the effectiveness of tax and customs administration, and an excessive reliance on proceeds from bauxite, which still provided about 67% of Government's revenue in 1990. 1.5 Consequently, as the terms of trade deteriorated by 33% between 1990 and 1994, total budgetary revenue declined in real terms from GNF 222.4 billion (15.5% of GDP) in 1989 to GNF 178.4 billion (10.4% of GDP) in 1994. Because the economic structure remained undiversified and the tax effort insufficient, the increase in nonmining revenue, by 12.7% annually, could not outweigh the 13.8% annual decline in mining revenue (see figure 1.1). 1.6 Guinea has now a tax-GDP ratio lower than most SPA countries in the region and much lower than SSA average of 19.4% (1993). Paradoxically, Guinea has perhaps one of the highest potentials for increasing its revenue. For example, compared with Mauritania -- a country that also relies heavily on mining revenue -- Guinea's tax-GDP ratio is not favorable. During 1991-93, the ratio for Mauritania was 22.4, compared with 13.3 for Guinea; 1994-96 projections indicate an increased ratio of 25.5 for Mauritania, but a declining ratio of 12.8 for Guinea. (Annex 1 table 1.19 also draws these comparisons for other SPA countries.) 1.7 Rapid increase in expenditures followed by a drastic but insufficient expenditure adjustment. In 1985, the new Government promptly tackled the formidable challenge of rebuilding the devastated economy. Between 1985 and 1990, expenditures grew at a rapid 4 pace, sustained by high revenue growth, massive investment and budgetary support from the donor community. As a result, total expenditures grew by 9% between 1988 and 1990 in real terms and investment expenditures by 22%. At the same time, increasing external borrowing swelled the external debt stock (long-term publicly guaranteed outstanding and disbursed) from US$1.8 billion in 1986 to US$2.3 billion in 1991. 1.8 Rapid growth in expenditures brought the deficit to GNF 165 billion in 1990, from GNF 114 billion in 1988. More significantly, the deficit excluding mining revenue stayed at around 19.5% of GDP between 1988-1990, and in 1990, the Government again began to accumulate external arrears. 1.9 After 1990, nominal expenditure growth (on a commitment basis) remained positive, but slowed significantly until reaching a standstill in 1994, due to the sharp drop in investment expenditures. In real terms, however, total expenditures fell by 18.5% between 1990 and 1994, and total expenditures as a share of GDP declined from a peak of 24.6% of GDP in 1990 to 17.5% in 1994, a figure that is now significantly lower than the SSA average of 30% (1993). As a result of these trends, the public deficit decreased from 8.7% of GDP in 1989 to 7.2% in 1994. Figure 1.1 1.10 Expendi- Mining and Nonmining Revenue: 1990-1994 ture adjustment triggered a spi- ral of adverse 16% - 14%- consequences. a 12% On the mone- U 10%- 8%_ ~tary side, in 6%- order to meet 2% the benchmarks 1988 1989 1990 1991 1992 1993 1994 agreed to under UMining revenue NNon-Mining revenue D Total the ESAF, credit extended to the Government by the Central Bank for financing the expenditure gap resulted in delays in counterpart fund disbursements, hence considerably slowing down the implementation of projects. Poor public investment disbursement performance led to a progressive squeeze on the availability of external assistance (particularly since 1992) which further hampered PIP implementation. Dwindling Government revenue increasingly hampered debt service. Finally, expenditure adjustment was insufficient to prevent the raise in arrears which contributed to threaten private sector (including financial sector) activity and investment. Public investment expenditures increased from US$131 million in 1986 to US$304 million in 1990 in nominal terms. 5 Figure 1.2 1.11 Availability of External assistance as percentage of GDP: 1988-1994 external grants. Although grants rose from GNF 45 14% billion in 1988 to GNF 61.1 12% billion in 1994 (fluctuating a 10%- 0 8% around 3.7% of GDP), total a6%- external assistance (grants 4__ __ __ -6 n19 fgr .) 2% - and loans) dropped from 0% __ 12.3% of GDP in 1989 to oo M a~ c4 co 00 0 7.6% in 1994 (figure 1.2), 0 0 M 0reflecting notably difficulties SGrants Wloans Ototal in investment implementa- tion. 1.12 The expanding debt. Despite the decline in the debt stock as a percentage of GDP from 83.2 % of GDP in 1989 to 80.7% in 1994, the amount of debt outstanding since 1989 has increased. This reflects several cumulative factors, including high expenditure inflows associated with resources for the PIP, relatively small amortization payments, and the capitalization of external arrears of principal and interest during successive Paris Club reschedulings. Figure 1.3 1.13 Relative to Debt service (a) due as a percentage of exports of goods (FOB), exports, the burden of revenue and current expenditures: 1988-1994 external debt service due (a) includes interest and amortization, excluding debt relief. (in interest and principal, excluding debt relief) 90% has increased rapidly in 80% - the past two years, from 70% 60% - 32% of exports of goods 50% in 1992 to 42% in 1994 40%- (before rescheduling) 30% 20% (figure 1.3). Similarly, 10% debt service due rose 0% Lfrom 49% of domestic 1988 1989 1990 1991 1992 1993 1994 ______________________________revenue in 1989 to 55% E Debt service as % of exports of goods in 1994 and from 58% of N Debt service as % of GOG revenue O Debt service as % of GOG current expenditures current expenditures in 1989 to 61% in 1994. Reflecting this situation, net public capital movements became negative in 1994 as the increase in amortization due (from US$97.9 million in 1993 to US$146.9 million in 1994) outweighed public borrowing from abroad. 6 1.14 Domestic and external arrears accumula-tion. It has been estimated that the stock of arrears at the end of 1994 stood at more than GNF 65 billion (2% of GDP). 4 Accumulation of arrears has increased the expense of projects to the Government, as entrepreneurs incorporate interest charges into their costs. Arrears also hamper the Government's credibil- ity, which is inhibiting foreign investment. 1.15 The decline in private sector activity. As a result of the above, private sector activity has declined during the past six years while the increasing tax burden on a shrinking tax base has contributed to further discouraging private sector development. This is reflected in the decline in private investment (except for mining) from 9.2% of nonmining GDP in 1988 to 7.4% in 1994. Coupled with the sharp decline in public investment, total investment (except for mining) in Guinea fell from 18.8% on nonmining GDP in 1990 to 14.9% in 1994, below the SSA average of 16.6% in 1993.s These levels of investment threaten the economy's ability to renew its capital stock. This amount does not capture the arrears accumulated prior to 1991, for which little information exists. Including Nigeria and South Africa. 7 2. BUDGETARY TRENDS AND ISSUES Methodological remarks I. Composinon of the m-stetbdget Guinea' 1995 investment budget (PIP) comprises 75% of capita expenditures, 10%o of current expenditures and 1500 of tchnical assistance. In the: presentchapter, it will be assumed that all technical assistance is of a recurrent ine inex 3 contains a sensitivity analysis on the classification of technical assistance. 2. Capital and investment expenditures. Capital expenditures designate the share of investment expenditures (PIP) devoted to gross capital formation (750o of investment expenditures. given the. above-mentioned assumption on technical assistance). Annex 3 discusses the distinction between economic and accounting definitions of capital expenditures. 3. Current and total current expenditures. Total current expenditures is the sum of current expenditures (found in the current budget) and current expenditures found in the investment budget (about 25% of investment expenditures, given the assumption on the nature of technical assistance). 4. iRecurrent expenditures;designate salaries and O&M(aswell as the recurrent costs of the admiistration for the management of investments) generated bN capital expenditures. In Guinea, recurrent expenditures are found in the current budget and the investment budget. See annex 3 for details on the distinction betmeen types of current expenditures in the budget. 5. The Budget Nanonal de Diveloppement (BND) is the sum of: (i) GNF counterpart funds of external financing in foreign currency. usually bN the FED and USAID (including PL480), and (ii) domestically-financed investment. including (a) contributions to FINEX: and (b) 1000-o domesncally-financed projects. The FINEX is the investment budget financed by donors. It represents an average of 8000 to 85% of the total in%estment budget. A. ISSUES 2.1 The deteriorating terms of trade since 1990 have undermined Guinea's fiscal performance. Cuts in current expenditures have been operated at the expense of O&M expenditures, which have declined by 43% during 1990-94 while wages and pensions have increased in relative terms. As a result, the Government's ability to deliver services to the population is seriously impaired. Although public capital expenditures have fallen to very low levels (4% of GDP in 19941), their impact is eroded, as recurrent costs they generate cannot be financed. Unsustainability of capital expenditures has been worsened by the Government's uneven commitments to sectoral investments. Policy options towards resolving the recurrent cost crisis are discussed in chapter 5. In response to the fiscal situation, donors have increased shares of external financing in projects and loosened the enforcement of credit agreements, which raises concerns about sustainability. B. INCREASING IMBALANCES BETWEEN O&M AND WAGE RELATED EXPENDITURES 2.2 Current expenditures have declined by 18% in real terms between 1990 and 1994. However, the brunt of the cut was born by O&M expenditures which declined by 43% over the same period (table 2.1). As a result, the Government's ability to deliver services to the population is clearly hampered. The most disturbing examples are hospitals without materials (in, for example, Kindia, MOP data. 8 Dalaba, and Labe) or drugs, and education centers without teaching materials (students in Guinea have an average of less than one textbook each). The Government is also having difficulty in meeting daily living necessities. For example, in 1995, the initial budgetary allocation to electricity was cut in half, corresponding in some ministries to a reported 20% of consumption needs. 2.3 As the O&M share of current expenditures declined, "nondiscretionary" expenditures (the . wage bill, transfers, and external debt service) became increasingly rigid. Although the wage bill declined in real terms by 8% between 1992 and 1994 (table 2.1), in relative terms, it increased from 41% to 50% of total current expenditures between 1990 and 1994. The increases were due to the progressive reversal of the drastic civil service reduction measures of 1987 (such as the near doubling of salaries in 1992). Two recent measures are also likely to increase wage expenditure requirements in the longer term. The first is the advancement of a large number of civil servants to the highest wage category (category A). The second measure is the 1994 reorganization of the Government, which increased the number of line ministries and autonomous Secretariats from 16 to 28. The corresponding personnel reassignments and/or increases and their additional drain on the budget have not yet been fully measured. Table 2.1 Economic composition of total expenditures (in constant 1989 GNF millions) 198 19 1991 1992 1993 1994 Mayease Total Current Expenditures (1) 26,37 233.291 232472 220.1 2.1 189.828 219050 Goods and services 162,841 178,986 175,614 171,200 161,525 141,505 165,279 o/w Personnel 88,772 96,461 99,442 108,214 107,165 94,961 99,169 o/w Operations & maintenance 74,069 82,526 76,172 62,986 54,360 46,545 66,110 Transfers 27,592 14,414 19,606 19,582 17,438 20,094 19,788 Capital - - - - - - 0 Others 45,944 39,890 37,251 29,372 23,213 28,229 33,983 o/w debt service 45,944 39,890 37,251 29,372 23,213 28,229 33,983 Capital Expenditures (2) 120.455 121631 10 _ 114461 113.001 65.73 107314 Total 356,833 354,922 341,077 334,615 315,177 255,560 326,364 Total (excl. debt service) 310,889 315,031 303,825 305,242 291,964 227,332 292,380 Memo Items Recurrent/capital ratio 1.96 1.92 2.14 1.92 1.79 2.89 2.10 Shares of GDP (%) Total Expenditures 25.8% 24.7% 22.7% 21.8% 19.9% 15.4% 18.9% o/w Current 17.1% 16.2% 15.5% 14.3% 12.8% 11.5% 12.7% Capital 8.7% 8.5% 7.2% 7.4% 7.1% 4.0% 6.2% (1) including current expenditures in investment budget; (2) data from MOP. 2.4 Transfer expenditures also rose during the period, from 6.2% of revenue in 1990 to 10.6% in 1994 and from 6% of total current expenditures in 1990 to 10.5% in 1994 (see table 2.1). The increase was due to the growth in pensions and scholarships. Entitlements under the public pension program have been increasing exponentially; in 1995, they will reach an estimated GNF 21.6 billion, or 13% of the wage bill, covering some 34,800 beneficiaries. Weaknesses with the pension system have already surfaced. In 1994, the Government accumulated arrears in pension payments. Although the arrears were included in the 1995 budget and the pensions paid, overall allocations for 9 pensions have been underestimated, meaning that arrears will continue to accumulate. (Chapter 4 discusses the institutional perspectives associated with wages and pensions.) 2.5 As a result, the ratio between O&M and salaries in Guinea, although higher than the SSA average in 1990, has now fallen below the SSA average (1985-1989). In fact, it is now lower than in any region of the world (table 2.2). Given expenditures on goods and services of GNF 142 billion in 1994 (table 2.1), maintaining the ratio between O&M and wages constant at its 1990 level (46%- 54%), would have brought O&M expenditures to GNF 65 billion in 1994, an allocation which would have required a 20% cut in the wage bill. However, even if O&M expenditures had amounted to GNF 65 billion in 1994, it is unclear that this allocation would have been sufficient to ensure adequate delivery of services and functioning of the administration. For example, in the health sector, the SSA allocation to drugs is about $0.45 per capita2 and in Guinea, it has been estimated that drugs should amount to about 50% of non-wage current expenditures. Complying with these standards would have brought total O&M expenditures in the health sector to about GNF 6 billion in 1994. Similarly, in the road sector, O&M requirements for routine road maintenance were estimated at GNF 6 billion for 1994, and periodical maintenance requirements at GNF 19 billion.3 These estimates show that adequate O&M funding only in these two sectors would have consumed almost half of the GNF 65 billion O&M allocation. Table 2.2 Shares of salaries and O&M (a) in expenditures on goods and services (1985-1989) East .Asia South Asia Latin Middle Sub-Saharan Guinea America East/Norih Africa Africa 1990 1994 Goods and Serices 100O? 10O0. 10090 1006* 100V 0 1', 100% Wages and salaries 63% 36% 66% 57% 60% 54% 67% Other goods and services 37% 64% 34% 43% 40% 46% 33% (a) O&M are represented by "other goods and services" which in some countries may include other types of expenditures. Further disaggregation was not possible on basis of available data. C. THE STEEP DECLINE IN CAPITAL EXPENDITURES AND THE RECURRENT COST CRISIS 2.6 As shown in table 2.1, it is clear that the Government's expenditure adjustment policy has been at the expense of capital expenditures, which declined from 8.5% of GDP in 1990 to 4% in 1994. As capital expenditures constitute a share of public investment expenditures, the public investment-GDP ratio declined in parallel, from 9.2% of GDP in 1990 to 6.1% in 1994 and is now well below the average level in Sub-Saharan Africa (17.9% in 1993). 2.7 The paradox is that this steep decline in capital expenditures has not prevented the unfolding "recurrent cost crisis" as declining revenue has prevented the Government from meeting the recurrent expenditure requirements of its investments. In 1994, actual recurrent expenditures, (comprising wages and O&M), were estimated at about GNF 142 billion in constant 1989 GNF (see 2 This allocation takes cost recovery into account. 3 A portion of this amount is funded by donors in the FINEX. 10 table 2.1), representing nearly 40% of total requirements.4 Since priority is given to wage and wage- related expenditures (which represented about 70% of actual recurrent expenditures in 1994), the recurrent expenditure gap reflects primarily the gap in O&M expenditures as described in para. 2.5 above. Box 2.1 The Cost of Inadequate Provision for Road Maintenance Guinea has a road network of about 18,000 kilometers By the end of the First Republic, the road network had nearly collapsed. Given the importance of an operational network to the development of private sector activity (notably in agriculture), road rehabihtation became a priority. As a result, PIP allocations to the road sector increased rapidly to an average of more than 25% of the investment budget between 1989 and 1994. In 1994, maintenance requirements of the network (resuting in part from previous investments) have been estimated at about GNF 5.7 billion for routine maintenance and about GNF J3.6 billion for periodic maintenance. Investment projected to year 2000 (including in rural roads) will contribute to routine maintenance requirements of GNF 7 5 billion and GNF 15 billion for periodic maintenance in 2000 Detailed calculations are shown in annex 4. The recurrent cost crisis is best exemplified by road maintenance, as requirements are much higher than actual expenditures: the Government experiences difficulties in providing an adequate budget for road maintenance. In 1994. Government allocations for road maintenance amounted to GNF 2.5 billion (38% of routine maintenance requirements, only GNF 1.2 billion of which was disbursed. The Government has also allocated its resources to uses other than road maintenance. For example, the budget for road maintenance has been used to rehabilitate roads in Conakry. and the Government continues to spend resources, albeit modest ones, on construction. In addition, allocations to rural road maintenance have been insignificant, due primarily to the institutional arrangement governing sharing of funds for road maintenance between the Ministry of Public Works and the Ministry of Agriculture. The result is the continuous deterioration of the Guinean road network, still aggravated by the country's dramatic climatic conditions. The cost of inadequate provision for road maintenance is staggering. For example. between 1988 and 1993. donors spent about US$32.5 million (more than GNF 31 billion) to rehabilitate around 430 kilometers under the Transport Sector Project. This amount is exorbitant when compared with the estimated GNF 19 billion necessary for routine and periodic maintenance of the entire 18,000-kilometer network. In addition, donors have been compelled to implement significant road rehabilitation programs of their own or other donors' previous in%estments (e.g., the road to the airport). What are the options? Today, the donor community has become reluctant to invest in road sector financing without an adequate guarantee that roads will be maintained subsequently. and most donors view maintenance financing as a priority iexcept for rural roads). Based on experience in other African countries, the World Bank has developed a "best practice" model, involving changes in road maintenance and tax collection management with implications for a unified budgeL An evaluation of the desirability and applicability of this model in the case of Guinea could help the Government and donors arrive at a long-term solution to the maintenance problem. 2.8 The recurrent cost crisis has contributed to a considerable slowdown in implementation of the PIP. BND and FINEX disbursement rates deteriorated from 81% and 86%, respectively, in 1990 to 34% and 60% in 1994. More importantly, it has seriously affected the return on investments. The This figure reflects an estimate of recurrent cost requirements based on the R coefficient which provides a rough estimate of recurrent costs generated by capital expenditures. Annex 3 contains detailed calculations for these figures and a discussion on the limitations of the R coefficient and three sensitivity analyses on the R coefficient. One of them indicates that for an R coefficient of 7%, the recurrent cost gap in 1994 would have been at 25% of requirements. 11 cost of inadequate funding of recurrent costs is illustrated in the case of road maintenance in Box 2.1. Government's Allocation Preferences 2.9 The recurrent cost crisis has been fueled by the Government's allocation preference to capital expenditures. In spite of the dramatic decline in domestic resources, the Government has increasingly favored 100% domestically-financed projects (from GNF 12 billion in 1990 to GNF 27 billion in 1995), while its contribution to FINEX declined from 61% of BND in 1992 to 35% in 1994 and even became a prime candidate for budget (and expenditure) cuts (table 2.3). 2.10 Since 1990, BND disbursements have largely been insufficient to cover even the domestically-financed BND budget (except in 1991). In fact, BND implementation (including counterpart funds) was GNF 27 billion in 1990, compared with a budget for domestically-financed expenditures of GNF 33 billion. This imbalance deteriorated after 1990, and in 1994 BND implementation was less than half the budget for domestically-financed expenditures (at 44%). Table 2.3 Affordability of Government-financed investment allocations (GNF billions, current prices) . .1 .... 990 .! i:199 ':.::::* 199 1995 Primary surplus as% of GDP (1) 6% 5% 4% 3% 2% 2% (A) TOTAL BND BUDGET 33.3 44 40.8 59.2 55 52 less budget for counterpart funds (CPF) 2.3 8 0.8 18.2 13 3 = Total domestically financed BND budget o/w: 31 36 40 41 42 49 as a % of BND total 93% 82% 98% 69% 76% 94% (i) Budget for 100% Government-financed projects (2) 12 14 15 17 23 27 as % of BND total 36% 32% 37% 29% 42% 52% (ii) Budget for local contribution to FINEX (LCF) (2) 19 22 25 24 19 22 as %of BND total 57% 50% 61% 41% 35% 42% (B) TOTAL BND DISBURSEMENTS 26.9 39.9 31 31.8 18.5 As % BND budget 81% 91% 76% 54% 34% As % total domestic financing of BND Budget 87% 111% 78% 78% 44% As% of local contribution to FINEX Budget (LCF) 142% 181% 124% 133% 97% (C) FINEX BUDGET 202.3 2273 274.8 275.0 250.0 265.4 budgeted local contribution to FINEX budget (in %) 9% 10% 9% 9% 8% 8% (1) Excluding investment expenditures. (2) 1990-1992: Staff estimates. 2.11 Until 1993, the Government did in fact have sufficient resources to ensure full disbursement of its contribution to FINEX projects; thus, it is apparent that the Government's preference for allocating available resources to its fully domestically-financed projects reflected to some extent its uneven commitments towards donor-assisted projects as specified by credit and grant agreements. 12 In 1994, however, total BND disbursements (at GNF 18.5 billion) would have been insufficient to cover the budget for local contributions to FINEX (GNF 26 billion), even if the Government had allocated all BND expenditures to FINEX. D. DONORS' RESPONSES TO PUBLIC EXPENDITURE ISSUES 2.12 Although justified by the numerous needs of an economy in reconstruction, rapid increases in external assistance until 1990 and sustained commitments between 1990 and 1994 have contributed to the unsustainability of investment expenditures. In the past five years, donors have become increasingly confronted with tradeoffs between continued service to the stakeholders and sustainability concerns; and with the Government's uneven commitment to the programs they support. The Rationale for Domestic Contributions to Donor-Supported Projects 2.13 Donors also adopt a wide array of positions about whether to require local contributions to the programs they support. Donors who favor large amounts of external financing in projects (and/or limited binding conditions) invoke long-term commitment to assistance, while more stringent conditions are reportedly rooted in short- to medium-term sustainability concerns. To some extent, the two positions about local contributions reflect different perceptions about the Government's main reason for failing to finance its contributions to donor-supported projects: insufficient commitment and/or insufficient resources (although, as shown in table 2.3, it has become clear in 1994 that the Government's resources were no longer sufficient to ensure full disbursement of its contribution to FINEX projects). Domestic contributions financed from counterpart funds are particularly noteworthy here. These funds seem to have captured the worst of two worlds -- the fungibility of resources does not favor discipline, while external financing of recurrent costs does not promote sustainability. Table 2.4 Local contribution rates to FINEX and Bank programs, and the R effect Sector % of local contribution to R coefficient as % of capital R codyitent as 56 cuenally financed project expenditures of PIP as 9 of PIP Rural sector 5% 9,0 Mines/industriy/water (c) 16% 29% Infrastructure 6% 8% Social 15% 20% Administration 10% 10% Weighted average PIP (a) 7.0% 14% 11% Weighted average Bank, as % total 11% 12% 10% credit (b) (a) based on the 1995 PIP profile. (b) based on the 1993 portfolio profile. (c) includes telecommunications. 2.14 Determination of the level of domestic contributions to donor-financed projects is not subject to rigorous criteria. Cross-sector variation indicates that donors are more demanding in sectors whose recurrent costs are high (mines, industry, and water) and comparatively lenient in the social 13 sectors, where other considerations dominate how external financing shares are determined. However, overall, concerns about sustainability seem no longer to be reflected in the level of contributions (table 2.4). 2.15 As Government resources became increasingly tight, the response of most donors was to lower the requirements on levels of domestic contribution in the design of their projects (or increase the share of external financing in their projects). As a result, contribution requirements to FINEX have fallen from an estimated 10% of FINEX in 1991 to 8% in 1995. Beyond technical assistance and training, more than 10% of the investment budget is now spent on O&M, adding 36% to allocations in the current budget (figure 2.1).. Similarly, donors have increasingly been compelled to finance Government's contribution to their own or other donors' projects. (Annex 3 provides a disaggregation of recurrent financing by donors, including the Bank.) Figure 2.1 O&M by budget category 20,000 18,000 16,000 - 14,000 1 12,000 10,000 z 8.000 6.000 4.000 2,000 *O OPIP FINEX NPIP BND 5nOCurrent Budget 2.16 The growing willingness of donors to cover a larger share of total project costs serves the immediate purpose of continuing services to project beneficiaries. The presence of donors is critical to sustain the level of investments Guinea needs to ensure priority services to the population. As discussed in chapter 1, fluctuations in mining revenue have the potential for endangering priority expenditures. This uncertainty advocates for continuous, if not strengthened, donor support. 2.17 However, large shares of external financing may constitute a serious sustainability risk in the longer term.5 As donor support to Guinea is also subject to uncertainty for exogenous reasons, sustainability concerns become critical. Increasing the return on public investment is particularly serious in the face of Guinea's debt problems (as discussed in chapter 1). For instance, in the case of an institution-building project, one donor admits that although one of the projects that it has financed in Guinea embodied a clause about the Government's contribution requirement, the project had paid for all operating costs, and its structure collapsed after the project came to an end. This could also reflect lack of Government ownership of the project. 14 2.18 In addition, recurrent cost financing by donors raises allocation issues. First, it can nurture the chronic underfunding of and underspending on O&M by the Government. The clearest example is financing for the current budget of the Ministry of Agriculture. Current budgetary allocations to agriculture O&M were GNF 251 million in 1994 (against a wage bill of GNF 17,305 million). In 1995, donors' allocations to O&M in the agriculture investment budget were GNF 13.98 billion, or 28% of the investment budget. Thus, it is clear that financing of the operating budget has liberated resources and allowed the MAEF to maintain a large wage bill (figure 2.1). Second, it contributes to the growing disparity between BND and FINEX allocations, which, as discussed earlier, is due in part to the Government's uneven commitments to FINEX contributions and preference for its own projects. This is visible through the heavy presence of donors in some key sectors such as agriculture and infrastructure (particularly public works and urban development) which has allowed the Government to reduce its contributions to these sectors towards new priorities such as administration. What Are the Options? 2.19 The donors are increasingly confronted with the tradeoff between the sustainability of projects and uninterrupted service to the stakeholders. To improve donor coordination and the impact of investment, donors collectively would need to strike a difficult balance between these two poles. To improve the sustainability of donor-supported projects, two principles should be followed: (a) arriving at a clear sense of development priorities; and (b) modulating projects and programs to reflect sustainability concerns more realistically. 2.20 As discussed in this chapter, in the absence of a domestic increase in revenue mobilization, sustainability concerns will inevitably lead to the conclusion that the volume of investment is unsustainable. In that case, scaling down interventions would require a -more focused collective view about what the Government's priorities should be and in what sectors the Government should primarily be involved. 2.21 Three steps could be taken to address sustainability concerns in an operational manner. First, criteria for shares of external and domestic financing in donor projects should be defined formally. For example, externalities generated by investments in social sectors could justify a greater share of external financing at the expense of sustainability concerns (which seems already to be the practice). Second, to the extent that the impact and sustainability of investments in some sectors can be measured (e.g., changes in social indicators through quick surveys), an operational link could be made between the share of external financing in the sector and improvements in the sector's indicators. And third, imperatives for high shares of external financing might also be mitigated by some appreciation of the Government's potential for improving resource mobilization and allocating resources more effectively. Donors could then modulate the volume of their investments more formally, based on Government's revenue performance and credit compliance. 6 For a complete discussion on FINEX and BND allocations, see the Country Economic Memorandum (CEM), par. 2.26-2.28, World Bank, 1995. 15 3. PUBLIC EXPENDITURES ON HEALTH, EDUCATION, AND AGRICULTURE AND INTERSECTORAL ALLOCATIONS A. INTRODUCTION 3.1 Chapter I demonstrated that public investment growth will be limited unless the Government improves its domestic resource mobilization. As Guinea's debt-GDP ratio and debt service-export ratio continue to increase, the fragility of Guinea's treasury has become apparent. 3.2 Chapter 2 showed why budget imbalances, particularly between capital and recurrent expenditures, affect the impact of public investment: notwithstanding massive spending on, for example, the road sector, economic conditions are not really improving, while underfunded O&M allocations in key sectors (e.g., for medicines and textbooks) are not improving living conditions. 3.3 In particular, since macroeconomic constraints and budgetary imbalances may limit investment growth in the next 3-5 years, it is clear that the Government must improve its public resource management and increase its return on public investment dramatically if it wishes to sustain the economic and sectoral progress that it has made thus far. B. KEY ISSUES 3.4 There is considerable scope for more efficient allocation and use of public resources within key sectors. For example, allocations to the health sector are below the average for Sub-Saharan Africa and largely benefit the wealthy. Although the Government has substantially increased gross enrollment rates in primary school, budgetary allocations are going disproportionately to a relatively inefficient higher education system. And although the largest share of the budget goes to agriculture and its wide array of programs, little is known about how the budget is allocated to and spent on these programs, and there are questions about the effectiveness of some agriculture programs. Costing of health and primary education strategies and road maintenance by year 2000 casts doubt about the affordability of current sectoral strategies, reflecting to some extent the Government's difficulty in formulating a coherent development strategy based on affordable and consistent priorities. C. FUNCTIONAL COMPOSITION OF EXPENDITURES: DISPARITIES BY SECTOR 3.5 Expenditure allocations are largely inefficient. Too little is going to basic programs in the social sectors and too much to productive sectors, which may crowd out funding for priority expenditures. A likely case of misallocation is the Garafiri hydroelectric project discussed in annex 10. 3.6 Expenditures on social sectors declined by 25% between 1991 and 1994, although their share in total expenditures has remained constant at around 19%. But by all regional standards, expenditures on social sectors are low in Guinea: an average of 26% in Sub-Saharan Africa (1985- 89) and 38% in developing countries (1990). Guinea's spending on education has now reached the 16 continent's average of the previous five years (14% of total expenditures), but health expenditures in 1994 remain below the SSA average in 1985-90 (6%) and is half the developing countries' average in 1990 (8%). In particular, per capita expenditures on health and education (US$4 and US$13 in nominal terms) are about a third of the average of 21 SSA countries in 1993 (US$11 and US$33). It is also noteworthy that Guinea spends more on defense (6.6% of expenditures) than on health (4.4% of expenditures). In addition, Guinea's expenditures on general services and social services are roughly equivalent. Table 3.1 Sectoral composition of total expenditures: Percent of total expenditures (constant 1989 GNF) 1988 199 1991 1991 1993 1994. SSAA(.DDC(7 General Public Services .Yl% 67. 116% 17l./ 20.1% 18.6% L% % Defense 4.6% 5.2% 8.0% 6.7% 6.8% 9.4% 8% 13% Administration 10.1% 11.5% 9.6% 11.1% 13.3% 9.3% 22% 23% Economic Services 40.5% 41 5% JL/2 4LL% 42.3% 371 22% 2,21 Agriculture 13.6% 16.9% 16.2% 18.9% 17.2% 17.7% 6% 6% Energy 2.5% 1.4% 1.2% 1.6% 2.9% 2.7% 2% 2% IndustryfTourism/Mines/Water 6.9% 6.4% 7.1% 8.6% 5.6% 4.7% 2% 2% Transportation + Public Works 14.2% 14.1% 10.2% 13.1% 14.2% 10.6% 10% 12% Construction and Urban Development 3.3% 2.8% 3.8% 2.8% 2.3% 1.4% 2% 1% Social Services 12.7% 11.4% 193 112N 19.8% 19.2% 26% 39% Education 6.9% 7.3% 14.3% 13.0% 14.1% 14.1% 14% 17% Health 5.2% 3.7% 4.4% 3.8% 5.2% 4.3% 6% 8% Social: others 0.5% 0.4% 0.5% 0.3% 0.5% 0.8% 6% 12% Regions UY2R 0.1% 0.5% 0U5 U Q91 Common exp. (net ofdebt relie) 81.2% 43.9% 41.1% 53.0% 38.8% 45.9% less Public Debt (1) 49.5% 13.8% 18.6% 33.7% 20.5% 21.2% =Comm, exp, net above the line (2) 31.6 30 22.5% 1.A% J13% 24.8% olw extemal/dom. interest on debt 12.9% 11.2% 10.9% 8.8% 7.4% 11.0% olw O&M exp. SBK/OBK(3) 4.7% 2.8% 4.7% 4.4% 2.0% 0.0% olw personnel and O&M 12.2% 12.5% 4.1% 4.3% 6.1% 6.8% olw transfers (4) 1.9% 3.5% 2.8% 1.9% 2.8% 6.9% error 0.2% 0.2% 1.6% 0.1% -0.4% 0.3% Total above the line 100% 100% 100% 100% 100% 100% 100% (6) 100% (6) (1) including: arrears, amortization, differed payments, ST credits, all of which appear below the line in the TOFE. (2) including interest on external and domestic debt which appears above the line in the TOFE. (3) SBK/OBK is the Govemment-owned bauxite mine. (4) including dette viagbre (pensions), remboursements droits. (5) Sub-Saharan Africa 1985-1989. (6) Including other functional expenditures. (7) Developing Countries 1990. 3.7 Another noteworthy trend in Guinea is the consistently high expenditures on agriculture (17% of total expenditures in the period 1988-94). And in fact, this average is almost three times the Sub-Saharan average and the developing countries' average (6%). Average allocations to transportation and public works also comprise a large share of the total expenditures, reflecting the needs to rebuild the devastated infrastructure after 1985. At 13% of the total. expenditures during 17 1988-94, the share of expenditures on infrastructure is above the Sub-Saharan Africa average (10%). D. PUBLIC EXPENDITURES ON HEALTH Key Issues 3.8 Although the Government has addressed several alarming health trends -- primarily high infant, child, and maternal mortality-- many existing challenges remain. The Government has not yet found a way to free up expenditures to meet these health challenges, and the outlook in the health sector is not promising: * The Government has an ambitious, commendable, yet unaffordable health strategy, and public expenditure patterns in health are contrary to strategy statements. Per capita spending on public health care is extremely low when compared with Sub-Saharan Africa and with international standards. Moreover, public health expenditures are declining sharply in real terms. * Implementation of the investment budget has declined considerably in the past five years, jeopardizing the efficiency of donor programs. * Current expenditures are inadequate and go predominantly toward salaries; nonwage current expenditures (particularly medicines) have reached a critical level, threatening recent sectoral achievements and severely affecting overall performance in the sector. * Given the high incidence of preventable diseases in Guinea, insufficient resources are allocated to primary health care and prevention programs. In addition, public health expenditures disproportionately benefit the wealthy, who are also the main users of private services. In relative terms, the poor contribute more of their household expenditures to health care than do the wealthy. In fact, public health care is relatively more expensive for the poor than private health care is for the wealthy. * Cost recovery in primary health care exacerbates social inequality because the poor are the main users of those services. In addition, the presumption is that the poor cannot (or will not) pay for primary health care, due notably to the poor quality of services and the costs of travel and wait times in rural areas. * Modern private -health care is embryonic, limiting the redeployment of resources towards basic programs. * The quality and utilization of public services are still at issue. Sectoral resources -- particularly human resources -- are concentrated in the capital city and in a few geographical areas, limiting access to and the quality of services in many other districts of the country. * Analysis is hampered by inconsistency among data sources, notably those published by the MOF, the MOP, and the MOH. 18 Health Status and Health Sector Policy 3.9 Despite the fact that Guinea's GDP per capita is about 30% above the SSA average, its health outcomes compare poorly with most of its neighbors (table 3.2). Table 3.2 GNP and health status indicators in Guinea and Sub-Saharan Africa GDP (15S) Crude Life Infant Under-fhe Total Mortaity Rate Expectanel Mortality Mortality Fertilit) Rate Ideaths per tyears) Rate (per Rate (per (children per 1,000 1.000 lIve 1,000 live womenP inhabitants) Ibrrhs) biribsi Guinea 43"' 2013 44 136 237 6' Sub-Saharan Africa 330 13.6 52 94 175 5.8 Source: MOH data and World Bank database (1990). 3.10 These poor outcomes belie the fact that the Government's sectoral policy since 1984 has sought to make affordable and sound health services accessible to the population. The policy, based on the principles of primary health care (PHC) enunciated at the 1978 WHO/UNICEF conference in Alma-Ata and at Bamako in 1987, seeks to integrate curative, preventive, and promotional health care; improve individual, family, and community health, especially among vulnerable groups; and promote participation by beneficiaries. The policy also seeks to strengthen sectoral planning, management, and financial capacity, decentralize decision-making to the regional and district levels, adopt a more rational use of human resources, promote community involvement, and ensure adequate financing for the sector. 3.11 The PER has estimated the cost of the country's health strategy to determine its feasibility and affordability. It appeared from the onset that the targets set for the 16 programs in the strategy have unrealistic financial implications, since they seek primarily to serve fundraising rather than planning purposes. However, 7 priority programs were retained as they address 68% of disease prevalence. Total current budget for these programs and for improved access to primary health care will amount to GNF 16 billion in year 2000, while the total allocation per capita would be US$11.2, which is consistent with the cost of the "essential package" of US$9 to US$12 identified in the World Bank's analysis presented in the 1993 World Development Report and Better Health in Africa. (Annex 6 provides detailed calculations for costing the health strategy.) Organization and Performance of the Health Sector 3.12 The Government has indeed made some progress in the health sector: it has increased child immunization coverage to more than 60% and PHC facility coverage to 80%; developed an essential and generic drugs policy and introduced drug cost recovery; improved human resources and skill mixes; and enhanced sectoral management and administration. But it has done little to address several other shortcomings, e.g., many districts are understaffed and provided irregularly with medicines, laboratory facilities lack reagents and effective quality control processes, and funds for supervision, maintenance, vehicle repair, and fuel are insufficient. As a result, health facility utilization rates are low and the quality of care is poor. 3.13 In addition, regional disparities in health resources and skill mix persist (table 3.3); for example, health sector staff is concentrated in Conakry (61% of the country's midwives, 100% of its 19 dental technicians, 59% of its laboratory technicians, and 49% of its medical doctors and dentists are in the capital city). The Government should continue to seek more effective redeployment -- a priority expressed, but not yet actualized. Table 3.3 Health providers/population ratios by region Loner Middle Upper Fore-t All Sub- Ratio Conakry Guinea Guinea Guinea Guinea Guinea Saharan Africa I docor for 2,895 4,118 10,666 11 024 10,979 6,570 24.600 1 midwife for: 5,677 8,104 47,404 37,798 24,766 15,024 - 1 para-medical for: 997 1,073 1.675 1,338 1,143 1,233 2,200 Source: (Three-Year Rolling Plan, 1994-1996). 3.14 As with public sector health care personnel, personnel in the formal private sector are concentrated in Conakry (which accounts for more than 50% of pharmacies and about 60% of the existing 126 private health facilities of the country) and in the regional capital cities. The partnership between the Government and the private sector is currently limited to a few local NGOs. The Government recognizes the importance of the private sector, but unfortunately, it has not sought to develop it systematically. (Annex 6 provides a more detailed discussion on the organization and performance of the health sector.) What Does The Government Spend on the Health Sector? Economic Composition of Budget and Expenditures 3.15 During the 1992-94 period, health expenditures declined in real terms. On a per capita basis, the drop was significant: from US$7.1 per capita in 1988 to US$1.8 in 1994 (table 3.4). Similarly expenditures dropped from 1.3% of GDP in 1988 to 1.0% of the GDP in 1994 (see annex 1, tables 1.13, 1.15 and 1.18), a performance that compares poorly with other countries that have a similar income distribution. However, public health expenditures as a share of total expenditures remained fairly constant at 4.5% of total public expenditures (table 3.4). The decline in public financing in real terms and the imbalances due to an insufficient and declining current nonsalary budget limit the effectiveness and efficiency of the sector and are contrary to the Government's expressed policy. 3.16 In 1995, a significant effort was made to endow the historically low health current budget more adequately (from 3.9% of the total budget in 1994 to 5.4% in 1995). Despite this increase, the current budget for the sector remains inadequate, aggravated by the large proportion of funds spent on salaries (which reached 94% of the current budget in 1994), and incomplete budget implementation, especially for current nonsalary expenditures. Health sector budget allocations are shown in annex 6, table 6.1. 3.17 The investment budget showed a upward trend in real terms (+32% between 1988 and 1995). Donors' allocations remained remarkably stable during 1990-95, so that most of the investment budget growth was born by the BND, which increased by 40% in real terms during 1990-95. Between 1988 and 1994, 73% of the sectoral budget was actually spent. But the disbursement rates on the current and the investment budgets differed sharply -- an average of 97% for the first, compared with an average of 65% for the latter. Implementation of both the BND and FINEX budgets declined during the period, and deficiencies in the implementation of the FINEX budget 20 reduced donors' contribution to only 25% of actual investment expenditures (or 13% of total sectoral spending). As a result, investment expenditures declined from GNF 12.3 billion to GNF 4.4 billion in real terms. The decline in the FINEX budget's disbursements, due to insufficient counterpart funding and the critically low budget for current nonsalary expenditures, is also indicative of weaknesses in donor coordination. Table 3.4 Health expenditures: 1988-1994 (in GNF millions) 1988 1999 1991 1992 1993 1994 Aveage Expenditures i.n current GNF Current expend. health 4,960 5,506 10,182 12,641 12,414 12,264 9,661 Investmentexpend. health 10,085 10,623 11,733 8,582 17,075 8,522 11,103 Total expend. health 15,045 16,129 21,915 21,223 29,489 20,786 20,765 Real Expenditures (constant 1989 GNF) current expenditures health 6,366 4,614 7,130 7,593 6,961 6,607 6,545 as % of total current exp. 3% 2% 4% 4% 4% 4% 3.6% investment expend. health 12,329 8,574 8,025 5,063 9,321 4,429 7,957 as % of total investment 8% 5% 6% 3% 6% 5% 5.5% Total expend. health 18,695 13,188 15,155 12,656 16,282 11,036 14,502 as % of total expenditures 5% 4% 5% 4% 5% 4% 4.5% Tot. exp. health as % of GDP 1.3% 0.83% 0.78% 0.51% 0.79% 0.44% 0.78% Tot. exp. health/capita (in constant 1989 USS) 7.1 3.5 3.6 2.4 2.8 1.8 3.1 Source: MOF. Functional Composition of Expenditures 3.18 During the 1991-94 period, expenditures for primary care represented an average of 24% of total expenditures on health, secondary-care facilities captured 22%, tertiary care 22%, and administration 18% (table 3.5). This budgetary allocation, and especially the fact that there has been no significant increase in the proportion of funds allocated to primary health, care, is inconsistent with the country's policy for the sector. However, three factors mitigate this observation: (i) about three-quarters of expenditures incurred at the secondary-care level went to prefect-level hospitals (which also deliver primary health services); (ii) the increase in expenditures for tertiary care in 1994 was only temporary, largely for rehabilitation works performed at the two large hospitals in Conakry (US$10 billion in 1994, or 60% of the investment budget, compared with 17% and 22% in 1991 and 1992, respectively); and (iii) the hospitals included in the tertiary care category performed largely secondary-care tasks. 3.19 Current nonwage expenditures for primary health care facilities have been inadequate. The bulk has been spent on drugs (table 3.5), but even this amount has been inadequate. The share remaining for other current expenditures -- the maintenance of infrastructure and vehicles -- is insufficient, creating deficiencies in supervision, patient referral, and vehicle operation. 21 3.20 Public expenditures for medicines, which were increased in 1994 to about 64% of current nonsalary expenditures, are less than US$0.25 per capita, and compare poorly with the average expenditure of US$2 in African countries. Table 3.5 Health expenditures per levels of care and for administration 3.21 The Government (1991-1994) policy to redeploy health (as percentage of total health expenditures) personnel to the primary care level is not reflected in 1991_199 _M1993___94_ verag the expenditure figures, and Primary care 34% 29% 12% 22% 24% the distribution of wage Current 10% 17% 12% 12% 13% expenditures per level of Non wage 1% 1% 1% 1% 1% Drugs 1% 1% 0% 1% 1% care and administration has Capital 23% 12% n.a. 10% 11% not changed in the past Secondary care 24% 27% 15% 22% 22% several years. In 1991, the Current 13% 21% 14% 17% 16% Non wage 2% 2% 1% 3% 2% salaries of primary health Drugs 1% 1% 1% 2% 2% care personnel represented Capital 11% 6% n.a. 6% 6% 29% of total expenditures Tertiary care 19% 20% 8% 42% 22% Current 8% 11% 8% 8% 9% for wages; secondary-care Non wage 3% 2% 2% 2% 2% salaries captured 36%, Drugs 1% 1% 1% 1% 1% Capital 11% 9% n.a. 34% 13% tertiary-care salaries 16.5%, Administration 24% 25% 7% 14% 18% and administration 18% -- Current 6% 11% 7% 7% 8% figures that stayed at the Non wage 1% 1% 1% 1% 1% . Capital 17% 14% n.a. 7% 10% same levels in the ensuing TOTAL 100% 100% 100% 100% 100% years. The figures cast Current 38% - 59% 42% 44% 46% - doubt - on -whether the Non wage 7% 7% 5% 7% 6% Drugs 3% 3% 2% 4% 3% Government has been able Capital 62% 41% 58% 56% 54% to redeploy staff effectively. Source: MOH data. Who Uses Public Health Services? 3.22 Wealthier groups report more illnesses and, when they are ill, are more likely to seek treatment (annex 6, figure 6.1 and annex 6, tables 6.2 and 6.3). Among individuals who reported an illness in the previous month (based on the 1994 Integrated Household Survey), 60% of the poorest quintile' did not seek any treatment, compared with about 30% of the wealthiest quintile. When the poor did seek treatment, they relied primarily on private traditional medicine; only about 12% sought The expenditure quintiles used in the analysis are for the entire population; we formed them by ranking individuals according to per capita total household expenditures as reported in the 1994 Integrated Household Survey. Each quintile contains exactly 20% of the population, with the first quintile referring to the poorest 20% of the population and so on to the fifth quintile, which refers to the wealthiest 20% of the population. There may, however, be measurement errors in the total expenditure variable used to classify individuals into quintiles. In the aggregation of household expenditures, an average of 42% of expenditures are for food and 58% for nonfood items. This food share is significantly lower than that recorded in many other Sub-Saharan African countries. 22 treatment in the public sector.2 In contrast, more than 50% of ill persons in the wealthiest quintile used public hospitals, which are more expensive than other health care facilities. The Equity of Public Health Expenditures: Do They Benefit The Poor?3 3.23 Wealthier quintiles receive much larger per capita public expenditures on health than do poorer quintiles (figure 3.1). This result is not surprising given the utilization patterns discussed earlier. The per capita expenditures of GNF 4,062 for the wealthiest quintile and only GNF 346 for the poorest quintile represent a more than tenfold difference (see annex 6, table 6.5). The wide gap is due primarily to the unequal distribution of hospital-level expenditures, given that the two wealthiest quintiles absorb almost all of the public hospital subsidy. MOH spends twice as much of its current budget on hospital care than on primary health care, yet only the wealthier groups use these services. Figure 3.1 Per capita public expenditures on health, by quintile: 1994 3500 3000 . 2500 gohospitals 2000 e3 -. hcalth ccntcrs 000 5i 00 0 Q uintiles Source: Table 3.2 and Integrated Household Survey, 1994. 3.24 The differences among quintiles at the primary health care level are not as dramatic; the wealthiest quintile receives GNF 975 per capita, and the poorest receives GNF 265. Per capita public health expenditures are also higher in urban area than in rural ones, reflecting the fact that the wealthier quintiles are concentrated in urban areas. Furthermore, per capita public health expenditures are concentrated in the capital, Conakry, where they are twice as high as in other regions (figure 3.2). These results imply that, if the goal is to distribute public health care equitably, resources are not focused effectively on primary health care facilities, which are used by poorer groups. How Well Targeted Are Public Health Expenditures? 3.25 The poorest quintile receives about 2% of public expenditures on hospital subsidies; the wealthiest quintile receives 55% of those public expenditures. This distribution is alarming -- the most unequal for any African country for which incidence analysis has been undertaken (see box 3.1). 2 Treatment in the private sector consists primarily of visits to traditional practitioners in their homes. Only the wealthiest groups visit modem private practitioners. A methodology of the analysis is shown in annex 6. 23 Figure 3.2 Per capita public expenditures on health subsidy, by region: 1994 3500 M 3000 2500 5 2000 hospitals 1000 ~ 500 hah etr 0 Conakry Lower G. Middle G. Upper G. Forest G. Average Regions Source: Table 3.2 and Integrated Household Survey, 1994. 3.26 A comparison of the distribution of public health expenditures with the distribution of the population indicates that all public health expenditures are distributed unequally (see annex 6). The poor receive a smaller share of public health expenditures than their share of the population, and the wealthy receive a larger share. The poorest quintile also receives a much smaller share of public expenditures on health centers - 10% versus 36% among the wealthiest quintiles. Box 31 T Targötiu f Publc xpenitures o Health: e Iaä maional Comparisons One way to meaure hov effectiVe public expenditures are targeted is to cömpare-the hares of expenditures.that go lternatively to the poorest and wealtist qüintiles. The poorest quintile shoul receive a disproportiònate share of the subsidy ,--ideally irie tian 2O%/ for t o reasons hey are less able o purchase services in the private sector and thus depend more heavily on publicly provided services, and their, health needs are greater, since they typically :have more children. As-showfi in the tble below, p lli: health expenditures in all seven Äfrican:countries for which an incidence analysis f public expenditures onihealtli has been undertaken are not weäargeted at the poorest 20% of the population. -In almost all of those countriesthe poorest 20% receive significartly less than 20/ of the public health expenditures, and the wealthiest.20% receive more than:20%. Expenditures are particulåily poorly targeted in the,thtee West African cöuntries för which we have information (Cöte d'öire, Ghina, 'anid Guiiea.)(Theonly:exception is South Africa, where the middle classrbenefits the mst from public services, t where the wealthiest groups opt for private services In contrast, two non-African countries that target.public expenditureä at health:are Colombia andMalaysia. Gtiinea is ineffective at tàrgeting håltli expenritires to the poornprimarily becatis the poor in Guinea use public services less than;do the poor:in other African countries, and,.when they. do use public services, they rely less on the less expensive services provided.by primary health care facilities. In contrast, the wealthier groups benefit exclusively from the expensive services provided in hospitals. Iniddênce f public expenditres on b-elth in selecteid cointr (percent),4 ...r...t20 iW thiest20% Guinea (1994) 4 48 Afiricat Môte d,jvoire (1993) 10 30 Ghana (1992) 12 33 Kenya'(1992/3) 14 24 Madagascar (1993) 18 24 Maläwi (1990/1) 6 17 South Africa (1993) 16 17 Tanzania (1993) .13 23 Other Countries Colombia (1992) 28 12 Malaysia (1989) 29 Il Sources: Dayton and Demery, forthcoming. *preliminary results. 24 The Burden of Health Care Expenditures: On the Government or on Households? 3.27 The Government contributes an average of 55% of total per capita expenditures on public health care; households contribute the remaining 45% (table 3.6). Public health expenditures are more important for the poorer groups, capturing a larger share of the total spent on public health care. Nevertheless, the total amount spent by both the Government and households is greatest for the wealthiest quintiles. In fact, total spending on health increases dramatically by expenditure quintile. Per capita spending on public health care is 13 times larger on the wealthiest quintile (GNF 4,062) than for the poorest quintile (GNF 346). Table 3.6 Total per capita spending on public health care: Government and household contributions, 1994 Qu___til _ Go_ernment Householdi Total Hosehold spendin as % of total Poorest 346 152 498 30% 2 676 350 1,026 34% 3 1,361 837 2,198 38% 4 1,949 1,370 3,319 41% Wealthiest 4,062 3,648 7,710 47% Average 1,677 1,358 3,036 45% Source: Figure 3.2 and Integrated Household Survey, 1994. Cost Recovery and the Burden of Household Spending for Public Health Services on the Poor 3.28 One of the most controversial debates about the public health care system in Guinea pertains to the effect of the current cost recovery program on the poor. The broad consensus is that about 40% of the population cannot afford a consultation at a public health care facility during some part of the year. Yet this common wisdom has never been confirmed with empirical analysis, given the. absence of data on the ability to pay. The results of the 1994 Integrated Household Survey can now be used for an analysis of the ability to pay for health care,4 but the findings were not available in time for this.report. Nevertheless, it is possible to assess the' burden of health spending on the various expenditure groups. 3.29 Two conclusions emerge about per capita spending on consultations at public and private health care facilities in Guinea (table 3.7). First, relative to household expenditures, the burden of health care spending on the poor is greater than on the nonpoor. In fact, public health care is more expensive for the poor than private health care is for the wealthy. For all health care, the poor spend almost twice as much of their nonfood expenditures as the wealthy do. Second, poorer groups spend more on private health care (predominantly traditional health care) than they do on publicly provided health care. 3.30 Why do people in Guinea, especially poor people, choose not to use public health services? The answer cannot be ascertained on the basis of the cost recovery fees only. As discussed, the poor pay more for consultation fees in the private sector. However, when the consultation fee for a visit to a public sector facility is added to the travel and opportunity costs of their time, public sector services are much more expensive for the poor than private sector services (annex 6, tables 6.7 and 6.8). For instance, a visit to a public primary health care center takes an average of 9 hours, whereas Analysis of willingness to pay was initiated by USAID to follow up on the findings of the present section. The results of this analysis should be available in 1996. 25 a visit to a (private traditional) caregiver takes less than an hour. (It is not known how much of this time is spent in travel or wait time.) Hence, there is a presumption that costs associated with travel and waiting contribute to making the public health care system too expensive for the poor. To give the poor an incentive to use public health care, the Government should seek not only to improve the quality of services, but also to reduce the time spent on consultations. Table 3.7 The burden of household expenditures on health: 1994 Quimifles Per capita Per capita household expenditures on Health spending as a share of per capita nonfood health care consultations * nonfood expeadilures (percent) expenditures Public Pritate All health Public Private Al health sector Sector secior sector Poorest 31,439 152 237 388 048 0 75 1 23 2 56,849 350 478 828 0.62 0.84 1.46 3 102,055 837 423 1,260 0.82 0.41 1.23 4 188,408 1,370 856 2,226 0.73 0.45 1.18 Wealthiest 697,192 3,648 1,877 5,525 0.52 0.27 0.79 Average 215,113 1,358 917 2,276 0.63 0.43 1.06 Rural 102,376 603 577 1,180 0.59 0.56 0.76 Urban 434,704 2,350 1,143 3,493 0.54 0.26 0.31 Source: Integrated Household Survey, 1994. a Per capita household expenditures on health were calculated as the total number of visits multiplied by the average fee paid (for each type of visit) divided by the population for each quintile. Since the survey did not collect information on more than one visit during the four-week recall period, it is likely that this level of spending underestimates total health care expenditures. In the future, this table will be revised based on expenditures on health from another section of the questionnaire. Policy Options * Increased health allocations will be necessary to improve health status in Guinea. However, health care objectives and allocations must be prioritized and costed to increase the effectiveness of health sector expenditures. Additional resources should be directed primarily to basic health care programs, to reach the minimum package of US$9-US$12 recommended in the 1993 WDR. Savings can be obtained from tertiary health care programs. In addition, nonsalary expenditures (particularly medicines) should be increased substantially from these minimum GNF 4.6 billion in 1996. Additional analysis will be required for clarifying the contribution of cost recovery for this budgetary component. The quality of basic programs should also be improved with a more effective redeployment of staffing towards rural areas. * Reasons for the uneven success of cost recovery should be examined further. Gross social inequalities should be corrected, notably with the introduction of cost recovery in secondary and tertiary care, support to private sector development in tertiary care, the initiation of saving mechanisms to improve the payment capacity of the poor, and improvements in public service quality. * Data collection and processing and the analytical capacity of the MOH should be improved. 26 E. PUBLIC EXPENDITURES IN THE EDUCATION SECTOR Key Issues 3.31 The current Government was left with a legacy of poor educational indicators and an inefficient allocation of resources. In response, it implemented a series of reforms that sought to increase enrollment rates in primary education, improve access to education among certain disadvantaged groups in the population, and increase public spending on education, particularly at the primary level. Our assessment indicates some positive results of the Government's efforts: an increase in gross enrollment rates in primary education, notably among girls; improved budgetary management; and decentralization at the Ministry of Pre-University Education and Professional Training (MEPUFP). But it also indicates that the Government still faces several severe problems with its educational system:s * Despite dramatic increases in gross enrollment rates (GER), enrollment in primary and secondary education is low, as are completion rates. * Equity. The poor are less likely to enroll in public (or any) school and thus receive a smaller proportion of public expenditures on education. Poor girls and students from rural areas have the lowest enrollment rates in public education and benefit less from public expenditures on education. As a share of nonfood expenditures, the poor spend less on education than other groups, another factor causing their low enrollment rates. Total per capita expenditures on education (including government and household contributions) are four times smaller for the poorest quintile than for the wealthiest quintile. * Overall public expenditures on education almost doubled between 1988 and 1991 and have now reached the average among countries in Sub-Saharan Africa. However, per capita allocation is still about a third of the SSA average. From 1991 to 1993, public expenditures on primary education have been declining. * The current strategy for 100% enrollment in primary education for 2020 cannot be achieved unless a significant effort is undertaken to reallocate resources towards primary education. * At the secondary education level, the absence of any capital expenditures seriously questions the actual value added by the Government compared with the role that the private sector could play. The same can be said of the vocational and education training (VET) system, given the dilapidated condition of the VET centers and the insufficient links with labor-force needs. * Higher education is inefficient and absorbs more than a third of public education expenditures. Besides, only the wealthiest benefit from higher education, creating inequalities in per capita recurrent costs by educational level. * The role of the private sector is limited; Guinea had only 95 private primary schools in 1993. The private education sector in Guinea is among the least developed in the region, accounting Annex 7 describes the organization of the education system. 6 21 countries for which data were available for 1993. 27 only for 4.3% of primary school students, 2.3% of secondary school students, and 22% of VET students. Education Outcomes: Enrollment and Completion Rates 3.32 Although Guinea ranked last in the UNDP Human Development Index of social indicators from 1992 to 1994, its pre-university educational system has improved considerably since 1990 with the first phase of its reform program (Programme d'ajustement sectoriel de l'ducation, PASE, 1990-1994). The PASE has increased the gross enrollment rate (GER) in primary schooling from one of the world's five lowest (29%) in 1989 to 40% in 1994. Yet, it has not made similar improvements at the vocational training and higher education levels. 3.33 The primary education system is also very inefficient. Although GER has increased dramatically since 1990, only 12.8% of first-year enrollees went on to complete the primary level and pass the end-of-primary exam in 1992. Thus, only 3.7% of the age cohort actually completes primary school; of these, only 0.7% are from rural areas, and only 1.1% are girls. 3.34 Enrollment rates in primary education among the poorest 40% of the population are half the national average and only one-third of the average for the richest 40% (table 3.8). At the secondary level, enrollment rate disparities are even more pronounced. Overall, enrollment rates among girls are twice as low as they are among boys; girls also miss school and repeat grades more often than boys do, and are less successful at the entry exam for lower secondary education. Enrollment rates in urban areas are more than twice as high as in rural areas. The bias in the educational system against rural students is even more striking than the bias against girls: fewer than 20% of all students in primary schools are from rural areas, despite the fact that the rural population represents more than half of all potential students. Middle and Upper Guinea have the lowest enrollment rates, both overall and for girls. (For a more detailed discussion on enrollment rates, see annex 7.) 3.35 Completion rates in secondary education area also low. Only a third of candidates pass the exam at the end of lower secondary education, and then one each at the end of 12th grade and at the end of the 13th grade. Table 3.8 Gross enrollment in primary and secondary education, by expenditure group: 1994 (percent) Printiq _ _ _ _ __ seconfir* ._ _ ... _.. _._....... Poorest Richest Average Poorest Richest A%erage 40% 40% 40% 40% All Groups 21 77 44 4 33 19 Boys 30 91 53 5 43 25 Girls 10 64 34 2 21 11 Rural 20 52 27 2 7 5 Urban 22 86 78 16 40 38 Source: 1994 Integrated Household Survey, preliminary results. 3.36 In the VET system, the bias against girls is flagrant. In the post-primary vocational training centers, female students represented only 12% of all students in 1985, but an even more alarming 1.7% in 1992. Nineteen institutions offer 30 separate programs to boys. Girls have a choice of only 3 programs, all in one institution. 28 Higher Education: Quality and Cost-Effectiveness 3.37 Higher education in Guinea has been subject to large policy shifts in the past two decades, having gone from a period of rapid expansion to one of draconian cutbacks and physical deterioration. The legacy of excessive expansion, weak managerial effectiveness, and an inefficient use of scarce resources has had a negative impact on performance; the effectiveness and efficiency of the system and the quality of education are low. The system is insufficiently related to the economic and social environment and Guinea's developmental needs. And key indicators continue to deteriorate; for example, the number of female students in higher education has declined from 19% in 1984 to 6% today. The sector's effectiveness is also increasingly hampered by student strikes. 3.38 Higher education is overcrowded and underfunded with a build-up of large internal arrears for food purchases, equipment, and supplies. Many responsibilities are ill-defined, and many institutions have overlapping functions and programs. Conflict and organizational inertia are pervasive. Under the Education Sectoral Adjustment (PASE), the Financial Affairs Directorate of the Ministry of Higher Education, Scientific Research and Culture (MESRC) has developed the capacity to prepare an operating budget that reflects program priorities. Expenditures on nonsalary, nonsubsidy expenditures in higher education have increased considerably, and the MESRC has developed the capacity to account for and handle these expenses in an efficient manner. The same success has not been achieved in the development of an investment budget, however, as the Public Investment Program continues to bear little relationship to sectoral priorities. Yet, since 1984, the central government has become virtually the sole funding source of higher education (aided sporadically by the international community). Monthly stipends to students range from GNF 30,000' for a full fellowship to GNF 5,000 for "social assistance". As a result, scholarships represent 48.7% of the institutions' budgets and 21.3% of the MESRC budget in 1995. Although the sub-region average, as a share of the higher education budget, ranges between 50% and 60%, this is still too high. An Assessment of the Government's Education Strategy 3.39 In 1989, the Second Republic Government published its current education strategy, with a commitment toward improving the educational system overall. The heart of the strategy was to improve education at the primary level -- in particular, to increase gross enrollment rates among all groups of students, and to increase the amount of public expenditures devoted to this level of education. But the Government also sought to address a deteriorating VET and higher educational system.8 3.40 In 1990, the Government embarked on its PASE, an extensive administrative and financial reform of its pre-university sector. The program consisted of a series of administrative and policy reforms, to be followed by further expansion and improvements in the primary educational system, Amounting to about one quarter of the average monthly salary of a primary school teacher. 8 The Government's 1994 statement of VET sector reform calls for a significant change in how this education level does business, to move it toward a demand-driven, rather than supply-driven, system. However, the system does not yet possess adequate managerial capacity to implement this far-reaching reform. The Government expressed its commitment to developing managerial capacity and establishing basic quality assurance controls and maintenance and upkeep functions. 29 including private schools. The key priority at the primary school level was to increase GER from 28% in 1989 to 70% in the year 2000; the projection has since been scaled back to a more manageable but still ambitious 53%. The Government has also placed priority on increasing access among girls and rural children and improving the quality of schooling and management, and has embarked on an ambitious effort to reduce the illiteracy rate, from 74% in 1990 to 37.5% in the year 2000. The Government also established targets to increase resources for education. The pedagogical input expenditure has increased to a high US$7 per pupil. All financial targets established for the sector under the PASE were met and surpassed, with the result that expenditures for nonsalary operating costs improved significantly, from 2% of pre-university current expenditures to 15%. On the other hand, the reduction of illiteracy has received little financing, mainly from donors. Cost Implications of the Government's GER Strategy 3.41 The Government's objective, articulated in the IDA-assisted project "Equity and School Improvement", is to reach 100% GER in primary school by year 2020, with an intermediary objective of 53% in year 20009. Three scenarios were envisaged. The results show that the strategy could be attainable only if efforts towards class construction and training of teachers intensify significantly after the project's life. In addition, as more children go on to secondary schooling, sizable donor assistance will also be required to increase capacity at this time; with an estimated annual increase of 5% in the number of children enrolled in secondary school, this level might face a capacity constraint even before year 2000. (The methodology and detailed calculation tables are presented in annex 7). What Does the Government Spend on Education? Table 3.9 Education Expenditures: 1988-1994 (GNF millions) .19 88 1990 1991 .992 1993 1994 Average. Expenditures in current GNF Current expend. education 13,687 23,336 52,553 61,554 62,223 63,378 46,122 Investment expend. education 5,831 7,821 17,632 11,164 17,581 3,633 10,610 Total expend. education 19,518 31,157 70,185 72,717 79,804 67,011 56,732 Real Expenditures (constant 1989 GNF) current expenditures education 17,565 19,556 36,799 36,973 34,889 34,143 29,987 as % of total current exp. 90/ 10% 19% 20% 21% 20% 16% investment expend. education 7,129 6,313 12,060 6,586 9,596 1,888 7,262 as % of total investment 4% 4% 8% 4% 6% 2% 5% Total expend. education 24,694 25,868 48,859 43,559 44.485 36,031 37,249 as % of total expenditures 7% 7% 14% 13% 14% 14% 11% Tot. exp. education as % of GDP 2% 2% 3% 3% 3% 2% 2% Tot. exp. education/capita (constant 1989 USS) 9.4 6.9 11.5 9.6 7.5 5.8 8.5 Source: MOF. 9 As of 1994, the Government was slightly ahead of its pace, at a GER of 40%. 30 3.42 Economic composition of expenditures. A striking feature of investment expenditures in the education sector is the absence of investment in secondary education and VET, and the absence of external aid for higher education (table 3.9). Investment expenditures fell precipitously from a peak of 8% of total investment expenditures in 1991 (as donor aid inbreased in response to the PASE) to 2% in 1994, while current expenditures remained more stable at around 20% of total current expenditures (see annex 1, table 1.13). 3.43 The budget for wages and salaries has declined by one percentage point annually during 1991-93, from 50% to 47% of the total education budget; this figure is low when compared with other budgets in the Sub-Saharan Africa region (see annex 7, tables 7.3 and 7.4). Actual wage and salary expenditures have evolved differently, rising from 50% in 1991 to 60% in 1992 and to 67% in 1993. This change in the economic composition of educational expenditures reflects not only the wage increase of 1992, but also the relatively weak performance of the investment program. Table 3.10 Functional composition of budget and expenditures (as percentage of total education budget and expenditures) 1991 1992 1993 Educational Level % total % total % total % total % total % total education education education education education education budget expenditures budget expendirures budget expenditures A. Primar) 50% 52% 51% 42% 42% n.a. 1. Wages and salaries 20% 21% 20% 25% 20% 29% 2. Nonwage expenditures 5% 3% 5% 5% 6% 5% 3. Subsidies 0% 0% 0% 0% 0% 0% 4. Administration 3% 3% 2% 3% 3% 1% 5. Investment expenditures 21% 28% 25% 12% 15% n.a. B. Secondary general 17% 17% 16% 19% 17% 20% 1. Wages and salaries 15% 15% 14% 18% 14% 19% 2. Nonwage expenditures 2% 1% 1% 1% 1% 1% 3. Subsidies 0% 0% 0% 0% 0% 0% 4. Administration 0% 0% 1% 1% 1% 0% 5. Investment expenditures 0% 0% 0% 0% 1% n.a. C. VET 8% 4% 7% 7% 6% n.a 1. Wages and salaries 4% 3% 4% 5% 4% 6% 2. Nonwage expenditures 1% 0% 1% 1% 0% 0% 3. Subsidies 0% 0% 0% 1% 0% 1% 4. Administration 1% 1% 1% 1% 1% 1% 5. Investment expenditures 2% 0% 2% 0% 0% n.a. D. Higher 25% 23% 24% 28% 33% 37% 1. Wages and salaries 11% 10% 9% 12% 9% 14% 2. Non-wage expenditures 5% 5% 5% 7% 13% 19% 3. Scholarships 5% 3% 3% 2% 3% 5% 4. Investment expenditures 5% 4% 6% 7% 8% n.a. E. TOTAL 100% 100% 100% 100% 100% 100% 1. Wages and salaries 50% 50% 48% 60% 48% 67% 2. Nonwage expenditures 13% 10% 12% 14% 20% 25% 3. Subsidies 6% 3% 4% 3% 4% 6% 4. Administration 4% 4% 3% 4% 4% 2% 5. Investment expenditures 28% 32% 33% 19% 24% n.a. Source: MEPUFP and MESCR. For the MEPUFP, payment amounts are those of the ordonnancement level, due to the unreliability of the data from the Ministry of Finance. Figures for the MESRC, except for salaries (see previous footnote), come from the mandatement level of the Ministry of Finance. 3.44 Functional composition of expenditures. The shares of primary education in the education budget started to decline after 1992, from 51% to 42% in 1993 (table 3.10). In addition, in 1992 and 31 1993, shares of actual expenditures on primary education in total education expenditures were lower than their budgeted shares, indicating that primary education did not receive disbursement priority. (For detailed figures on budget and expenditures by level of education, see annex 7, tables 7.4 and 7.5.) 3.45 This evolution contrasts with that of tertiary education. As a share of the education budget, tertiary education increased significantly, from 25% in 1991 to 33% in 1993. Similarly, in 1992 and 1993, higher education expenditures amounted to 28% and 37% of education expenditures (higher than their budgeted share), indicating slippage in higher education expenditures. In 1993, expenditures on tertiary education actually exceeded those on primary education by GNF 3.6 billion. 3.46 This situation has given rise to significant inequalities in unit recurrent cost (table 3.11). Unit costs in higher education are 54 times those of primary education. This ratio compares with Sub-Saharan Africa countries, although it is significantly higher than in some countries (e.g., in Niger, the ratio is 17). Inequalities in education expenditures are discussed further in the next section. Table 3.11 Budgeted recurrent unit costs per education level u lLevel Etlhnt 199.. 994 UCit A(GNF), 19r U Primary 471,120 47,625 1 Secondary 109,913 116,812 2.5 VET 8,105 603,514 12.7 Teacher Training 1,049 1,339,557 28 Higher 8,103 2,595,705 54 Note: Amounts differ from the budget recorded by the Ministry of Finance and the budget notified to MEPUFP, for reasons that must still be explored. The figures come from the budget of MEPUFP, given the coherence of its budget monitoring and implementation. Equity of Public Expenditures on Education Among the Population 3.47 Per capita public expenditures by expenditure group, location, gender and region. Per capita expenditure inequalities reflect enrollment rate inequalities. The total per capita public expenditure is about nine times greater for the wealthiest quintile (GNF 19,285 per capita) than for the poorest quintile (GNF 2,116 per capita). The bias in the distribution of public expenditures for higher education toward the wealthiest quintiles is largely responsible for this dramatic inequity (figure 3.3). Per capita public expenditures on education are also biased towards urban areas. Because enrollment is higher among boys than among girls, a significant gender bias exists in the distribution of public expenditures for education. Overall, per capita public expenditures on education for primary and secondary school marginally favor those living in Conakry and Forest Guinea. (For a more detailed discussion on inequalities in per capita public expenditures in education, see annex 8.) 32 Figure 3.3 Per capita distribution of the public expenditures on education: 1994 20 .000 octrtia ry 18.000 . 0teeh/pror I6.000 g*secondary 14.000 . prima ry 12.000 10.000 8.000 .0 6,.000 4 .000 poorest 2 3 4 wealthiest Average rural urban male remaie 3.48 Overall education expenditures can be considered relatively pro-poor, because the poor receive a larger share of expenditures than their share of total household expenditures. However, expenditures on technical and professional training and university-level education are not at all pro- poor (see annex 8 for a discussion based on the Lorenz curve). 3.49 In absolute terms, none of the education expenditures is distributed equitably across the population (see the Lorenz curve in annex 8), meaning that the poorest quintiles receive fewer expenditures for public education than their share in the population. Only expenditures for primary- level education are the most equitably distributed of all educational expenditures. In contrast, the university-level expenditures are strongly pro-wealthy; the wealthiest quintile receives more than 60% of public expenditures on university education. An even more revealing perspective comes from a comparison of the targeting effectiveness of public expenditures on education in Guinea with the effectiveness of those in other countries (box 3.2). . ox 3. Targering Public Expenditures on Education: International Comparisons In Africa, public education expenditures are targeted weakly to the poor in each of the eight countries for which an incidence analysis has been done. In all of these countries, the poorest quintile receives significantly less than 20% of the total public expenditures on education, while the wealthiest receives more than 20%. This distribution can be compared with two countries, Colombia and Malaysia, both of which target public education expenditures effectively to the poor. Among these countries, educational expenditures in Guinea are the least well targeted to the poor. The poorest quintile receives 5% of total public expenditures on education, whereas the wealthiest quintile receives 449'o. When compared just with the two other West African countries for which data are available (C6te d'lvoire and Ghana), Guinea compares equally unfavorably Incidence of public expenditures on education in selected countries (percent)* Poorest 200%" Wealthiest 20% Guinea (1994) 5 44 Core d' hoire (1993) 10 37 Gbana (19921 16 21 Kenya (1992/3) 17 21 Madagascar ( 1993)' 9 44 Mal awi (1990'l) 10 38 SouLh Africa (1993) 14 40 Tanzania (19931 13 23 Other Countries Colombia (1992) 28 12 Malaysia (1989) 29 11 Sources: Dayton and Demery, forthcoming. *preliminary results. 33 Contribution of Households to Financing Public Education 3.50 The contribution of households to public education also increases with household expenditure levels. The poorest quintile spends about one-third as much as the national average (per student), and about one-fifth of what the wealthiest quintile spends per student (see annex 8 for a complete discussion and table on household expenditures on education). Poorer quintiles receive fewer public expenditures per capita, while also contributing less (figure 3.4). Figure 3.4 Per capita household and public spending on primary education: 1994 9,000 8,000 -I 7,000 6,000 a 5.000 [wh.e hold S4,000 pbi 3,000 2,000 1,000 01 quintiles The Burden of Education Expenditures on Households 3.51 Results from the household survey indicate that the poor do not spend a higher proportion of their total nonfood expenditures on public primary and secondary education (see annex 8). The poorest quintile spends the smallest share of nonfood expenditures per capita -- only 1% -- because so few children from the poorest quintile are enrolled in school. Education expenditures as a share of nonfood spending rises by quintile; the fourth quintile spends as much as 2.4% of its nonfood expenditures on education. However, the wealthiest quintile spends only about 1.1%, indicating that this group largely opts out of the public education system. Overall, however, these results indicate that household contributions to public education are rather low, and that there may be scope for increasing private contributions to education. Policy Options * Sustainability. If maintained, Guinea's current enrollment trend for primary school would generate universal enrollment by year 2020. Even if donor support to the country were to continue at its present level, the Government's budget constraints would dictate that it seek alternatives -- particularly private sector development and intra-education reallocations. As demonstrated by actual budgetary spending on education, both the PIP and the current budget show weaknesses that the Government must address with a more global perspective toward local expenditures if it is to pursue its priority in primary education. * Given the share of education in Guinea's overall budget (an average of 11% of actual expenditures for 1988-1994), and in light of its projected expansion at the primary level, the Government must set a stronger priority in its budgetary allocations -- neighboring francophone countries spend more than 40% of their current budgets on education. Moreover, the budgetary share for primary education does not reflect the priority accorded to this level. Therefore, the education budget must be reallocated. The reallocations will probably have to be at the expense 34 of higher education, VET, and higher secondary levels. As discussed in the following paragraph, the private sector cannot and should not compensate for public expenditures at the primary levels. Public expenditures would have to increase at least by 50% in the next two decades, and perhaps double that amount, if the actual increase in the number of school-age children is 4.5% rather the Government's projection of 2.8%. * Private sector development. Private schooling in Guinea was banned until 1984 and has room for expansion. The number of private vocational training centers is expanding rapidly in urban areas; 30 currently offer secretarial and managerial training (however, data on the quality of education in the private sector are not available). The VET sector also offers scope for private- sector involvement, as demonstrated by the recent creation of VET private schools. In urban areas, private education could accommodate up to 15% of primary school pupils, thus capturing some of the costs of new infrastructures and additional salaries. Nevertheless, equity considerations (notably between rural and urban areas), economies of scale, and quality considerations (providing textbooks and training teachers) argue for having the Government take the lead role in providing education as a basic service to all. In that respect, the Government could promote more equitable access to education by encouraging parents and associations to promote enrollment among disadvantaged groups. * Higher education. The current state of higher education calls for in-depth reform. Budgeting must be improved at each institutional level, and the Government must decide which institute it will keep and which it will close; for example, the 1995 budget of the Mining Institute in Bok6 is GNF 418 million annually for fewer than 20 graduates but with 29 full-time teachers. The alternative is to send students abroad for training that would no longer be offered in Guinea, provided that such training is still applicable to the country's developmental needs. In 1991- 1992, more students studied abroad than at the University of Kankan, the second largest higher education institution. The same question must be asked about research institutes: how can a laboratory that is supposed to study "diagnosis, prophylactics, and therapy of human transmissible diseases, epidemiological and epizootiological monitoring of the country, and serum production" possibly function normally with nonsalary expenses of GNF 42.8 million? F. PUBLIC EXPENDITURES IN AGRICULTURE Introduction 3.52 Agriculture is vital to the Guinean economy for broad-based, poverty-targeted development. It generates 24% of the country's GDP, employs 80% of the labor force, and provides livelihood to a large segment of the population in rural areas, where per capita income was less than half the national average of US$532 in 1994. The sector is also to play a key role in diversifying the economy away from its enclave mining sector. Throughout Guinea's economic and political history, public expenditures in agriculture have been assigned an important role in fostering agricultural growth and achieving important welfare objectives. Main Issues 3.53 The following are the main issues in the agriculture sector: 35 * Although growth in sectoral output compares favorably with output growth in other SSA countries, public expenditures on agriculture have yielded less agriculture value-added than envisaged by the Government's strategy formulated in the 1991 Letter of Agricultural Development Policy (LADP). * The sectoral strategy expressed in the LADP is comprehensive, yet contains several shortcomings. One is that it lists too many priorities without proper sequencing of activities to be undertaken; another is that it lacks a proactive strategy for private sector development in agriculture, as well as a strategy for addressing the financial sustainability of and intersectoral linkages among programs. The LADP is currently being revised. * A critical issue in agriculture, the role of the public and private sectors has not yet been delineated in practice, despite the stated intentions of the LADP, resulting in abnormally large public expenditures in the sector. Many programs have scope for greater private sector participation, including rural infrastructure and agricultural extension services. * As a result, agriculture's share of public expenditures is high in comparison with those of other developing countries, but the country lacks a framework and data for assessing the effectiveness of expenditures for most sectoral programs. Overall, public expenditures in the sector do not appear to be efficient. * The PIP still contains an overabundance of integrated rural development projects (IDRPs) and filibres -- legacies of the previous regime. IDRPs are largely unsustainable after their completion; as productive activities, there is a question as to whether filibres should be in the hands of the public sector. In addition, they seem to have high unit and per-farmer costs, and their sustainability has not yet been proven. * Ministry of Agriculture (MAEF) personnel expenditures are high; nonpersonnel expenditures are low, aggravating already low morale and productivity. Some MAEF directorates and services appear to be redundant. Overall, MAEF appears to be too large for the quality of services it offers, and the distorted budget financing of the MAEF and the unsustainability of investment indicate that the Government is doing too much, with donors financing about 80% of O&M. * Data for analyzing the distribution of public expenditures are not readily available; quantitative assessments of programs are also scarce. Current Objectives 3.54 The Government's current objectives in the rural sector are embedded in the Letter of Agricultural Development Policy (LADP) of 1991. They are to promote food security and reduce food imports; promote agricultural exports; provide essential support services to rural producers; and ensure rational (sustainable) management of natural resources. The strategy adopted by the Government in order to achieve these objectives is to: (a) pursue the macroeconomic reforms agreed under the SAL and sector adjustment programs to reinforce the market economy; (b) continue to disengage the State from productive activities and services that can be taken over by the private sector, and concomitantly undertake an active promotion of private sector activities; (c) improve business conditions in order to attract private investment, especially in rural areas; (d) provide agricultural services to the farming community, especially applied research, extension, and planting 36 material; (e) improve rural infrastructure in order to open up productive areas; (f) build up local administrative capacity by encouraging the development of cooperative and informal credit institutions, and by decentralizing the services of the agricultural ministry; and (g) undertake research and development activities and improve the legislative framework for natural resources management (land, forests, systems). 3.55 But the LADP has several shortcomings.10 It lists too many priorities and no sequencing of actions to be undertaken; relies on protectionist trade policy to support the domestic production of rice and vegetable oils while there are import duty exemptions for a sizable quantity of rice that eventually leaks into the market; does not address issues of implementation capacity and training needs necessary to carry out the proposed action plan; plays too much to donors' preferences; does not present a proactive strategy for private sector development for larger scale producers/exporters or for agroprocessing activities; and lacks intersectoral and macroeconomic linkages while recognizing their necessity. Thus it does not assess the financial sustainability of the proposed investment programs, or the integrity of the program as a coherent whole; nor does it make a critical assessment of investment strategies underpinning area development and crop-specific projects or evaluate the degree of priority; cost/benefit/appropriateness of large programs such as the Plan d'Action Forestier de Guinde (PAFG). The design of a sustainable strategy, one that takes due account of macroeconomic constraints, and the assessment of investment strategies and programs are hampered by an inappropriate skills mix at MAEF (too heavily dominated by agronomists or foresters) and a lack of analytical tools and data. 3.56 A revision of the LADP is underway. It will especially address issues of the scope and size of the public sector, the corresponding role and structure of MAEF, and the timing of and priorities for investment in agriculture. Sectoral Performance and Contribution to Overall Growth 3.57 In the past decade, production in the agricultural sector has been satisfactory, contributing substantially to overall economic growth. Since 1991, agricultural GDP growth has outpaced overall GDP growth (table 3.12). Table 3.12: Overall and Sectoral GDP Growth (1988-94)" GDPGrowth Rates%} 1988 1939 1990.991 1992 1993 1994 Overall 6.3 4.0 4.3 2.4 3.0 4.7 4.0 Agriculture 0.1 4 6 32 3.1 3.3 5.0 5.0 Industry 4.3 3.7 3.8 2.6 2.6 3.1 3.1 Services 12.1 3.0 5.4 2.2 2.6 5.3 5.1 1/ Figures from MAEF. These figures differ from those in annex 1, table 1.1. 3.58 The performance of Guinea's agricultural sector is also satisfactory by international standards, despite having started from a lower base since 1984. During the 1988-92 period, Guinea's agriculture outperformed all of its neighbors and stood at well above the Sub-Saharan (around 1% annual growth) and South Asian (around 2% annual growth) averages (see annex 9, figure 9.1). 10 For a complete discussion of the LADP, see the Country Economic Memorandum (CEM), World Bank, October 1995. 37 Sectoral Performance and State of Public Expenditures in Agriculture 3.59 This overall performance, however, masks several trends which indicate that public expenditures in agriculture have yielded a less than favorable result. Indeed, while indices of both real total agricultural budget and public investment (PIP) increased steadily from 1988 to 1993, real agricultural output and other impact measures were less spectacular (See annex 9, figure 9.2). Indeed, the estimated public expenditure elasticity of agricultural value added is only 0.32 (meaning that for every 10% increase in public expenditures, agricultural value added increased only by 3.2% during the 1988-1994 period). 3.60 This low efficiency of public expenditures undermines the food security and natural resource conservation goals of the LADP, with three implications for public vs. private sector interventions for economically and environmentally sustainable growth in agriculture. First, although total food production increased, per capita food production declined at an average annual rate of 0.3%, while food aid in cereals increased from 24,000 tons in 1979-80 to 30,000 tons in 1992-93. Second, the increase in rice output was essentially the result of area expansion, since crop yields have, at best, remained constant since 1992. 3.61 Third, the consequence of area expansion, especially for the cultivation of rice under the prevalent slash-and-burn technology, is deforestation. The index of forest cover shows a declining trend since 1988, although the rate of deforestation has not reached the alarming rate of neighboring Cte d'Ivoire (5%). This, coupled with the fact that fertilizer use is relatively low, means that an agricultural growth strategy should urgently address agricultural intensification and generation and diffusion of improved cultural practices if it is to be sustainable. Current State of Public Expenditures in Agriculture 3.62 The roots of the current mandate of MAEF -- and hence the level, structure, and trends in public expenditures in agriculture -- can be traced to three legacies from the past: the colonial, the First Republic, and the first decade of the Second Republic (discussed in annex 9). Overview and Trends of Public Expenditures in Agriculture 3.63 Overall public expenditures for the agricultural sector averaged about GNF 55 billion during the 1990-94 period (table 3.13), or about 17% of total public expenditures. While expenditures on agriculture have decreased in real terms from GNF 63 million in 1992 to GNF 45 million in 1994 (-30%), their share of total public expenditures has nonetheless remained in a fairly close range, between 17% and 19%. 38 Table 3.13: Agricultural Sector Public Expenditures, 1990-94 (in GNF millions, constant 1989 prices) - 1990 ~ 1991 j 19.92 1993 J 1994 Aeae Total Expenditures 59839 55303 63192 54334 45139 55561 Current Expenditures 5566 8997 9374 8574 9745 8451 olw personnel 93% 98% 98% olw O&M 7% 2% 2% Investment Expenditures 54273 46306 53818 45760 35394 47110 olw FINEX 93% 89% 90% o/w BND 7% 11% 10% (1) Including agriculture, livestock, fisheries, and forestry. 3.64 The agricultural share of total public expenditures is high in comparison with other countries: 17% for Guinea compared with about 7% for developing and 2% for industrial countries (table 3.14). Table 3.14: Government Expenditures on Agriculture: Guinea, Developing Countries, and Industrial Countries ndatrClaa Deveoisg Cou3intris Developed Countres 09I994)~ &#72.1983) (195/988%) (19841988 (19.93).( (19/98) (198&41988) 1. Ratio of agriculture share m pubc 0 .7 0.3 0.3 0.3 0.7 0.7 0.6 expenditures to agricultural output share in GDP 2. Agriculture share in public 17.0 7.6 7.5 7.2 3.3 . 3.0 2.3 expenditures (%) 3. Agriculture share in GDP (%) 24.0 26.3 25.5 23.6 4.7 4.5 3.8 4. Agriculture public expenditures as 15.0 8.9 9.7 11.6 22.5 22.5 22.5 a percentage of agricultural GDP (%) Sources: The Reform of Public Expenditures for Agriculture, van Blarcom, Knudsen, and Nash, (World Bank, 1993) and Ministry of Agriculture, Guinea. Economic Composition of Public Expenditures in Agriculture 3.65 Capital or National Development Budget (BND). During the 1990-94 period, the BND financed an average 10% of the PIP in agriculture, with international donors contributing 90%. The structure of the agricultural PIP is largely donor-driven and donor-implemented. 3.66 Although the PIP has emphasized "structuring" projects with a national scope (33% of the 1991-94 PIP) -- which is a recent change in investment strategy in favor of national services -- the distribution of the 1991-94 PIP is still reminiscent of the period when the Government was committed to undertaking mostly filire projects and IDRPs (see annex 9, figure 9.3 for the composition of the PIP in agriculture). 3.67 The filibre projects (23% of the PIP) appear expensive in terms of cost per farmer or unit of output, and highly capital-intensive. For example, during the 1985-87 period, there were about ten expatriates in the cotton filire in Guinea for 750 hectares cultivated and 44 tons produced; while International comparisons should be treated with caution since expenditures in agriculture under the line ministry are not homogeneous across countries. 39 Burkina Faso, for about the same number of experts, had 100,000 hectares and 124,000 tons.12 Although the filibres are meant to be privatized in the long run, complete privatization (when it occurs) usually does not happen until four or five consecutive phases of the same project have been carried out. For example, after the First and Second Coffee Projects (Relance Cafe) RC1 and RC2, Guinea and the CFD were contemplating RC3. The same is true for cotton and palm oil. The long- term nature of these operations would not be a problem (on the contrary, it is a recognition of the time necessary to achieve real development impact) if there were no questions about the cost effectiveness of the approach itself. 3.68 Similarly, the Government and donors still adhere to the IRDP concept. The 1991-94 PIP contained about half a dozen of these projects, accounting for 14% of the total rural development PIP. These projects have their own management structure parallel to that of the public administration, and generally little or no provision is made for sustaining their activities after project completion. The implementation completion report of one such project, PAG,13 now closed, calls into question whether these operations are sustainable.14 3.69. Current Budget. Personnel expenditures constitute 98% of the current budget of the MAEF.'s Relative to the other ministerial departments, MAEF had the third highest wage bill in 1993, with about GNF 16 billion (following National Defense, GNF 33 billion, and Pre-University Education, GNF 36 billion). Two independently conducted recent reports suggest that MAEF may be overstaffed.'6 The high number of MAEF's civil servants is even more conspicuous given that the number of projects in the Public Investment Program (PIP) decreased by more than one half from 133 in 1993 to about 50 currently, with no corresponding decrease in staff. The large wage bill is thus essentially due to the large work force of MAEF since the average salary at MAEF, about GNF 1.8 million (about US$1,822) in 1993, is much lower than the civil service average of GNF 2.8 million. 3.70 As indicated in the organizational chart of the ministry (see annex 9, table 9.6), many directorates or services could be deemed redundant in view of MAEF's own stated mandate and of what constitutes core agricultural public services. The most conspicuous case is the National Directorate of Agriculture. It still has hundreds of (a) rural development specialists (animateurs ruraux) whose functions clearly overlap with those of extension agents of the National Extension 12 See Lele, Van de Walle, and Gbetibouo, Cotton in Africa: An Analysis ofDifferences in Performance, World Bank, November 1989. 13 Projet Agricole de Guckddou, IFAD-, ADF-, and IDA-financed. 14 However, the analysis (and thus clear conclusion) is rendered difficult by the emergence of area-based natural resources management projects that have some of the IRDP features but are not supposed to have the same fate. In fact, the latter are part of the implementation of a national plan divided into different ecosystems with different priorities. The Fouta Djalon, the number one priority in terms of its degree of degradation, is now being specifically targeted by an important program of several river basin management projects, assisted by many donors. 15 The actual share should be lower because part of all non-personnel recurrent expenditures for international donor-financed agricultural projects are recorded in the National Development Budget. 16 See FAO: Deuxi6me Projet de Services Agricoles, Rapport de Pr6liminaire de Preparation, No 63/93, mai 1994. 40 Service; and (b) food quality control specialists (agents du service de conditionnement), effectively, tax collectors under the former regime whose current function is unclear.17 3.71 Non-personnel expenditures are low. Non-salary current expenditures of the economic services show that a staff member of the Ministry of Agriculture (already among the lowest paid of the civil service) has an annual budget of a meager US$50 for operating expenses, compounding his/her low morale and productivity. Functional Composition of Public Expenditures in Agriculture 3.72 Data necessary to conduct an analysis of the distribution of public expenditures (budget and actual allocations) by main functions and programs of the agricultural ministry are not readily available. This is a serious deficiency. Quantitative assessment of programs in order to take corrective measures at an opportune time is also lacking. 3.73 Rural Infrastructure. The National Rural Infrastructure Project (PNIR) helped reinforce the institutional and technical capacity of DNGR, to the extent that the directorate is now the sole coordinator and implementing agency for all activities pertaining to road rehabilitation and low-land development in rural areas over the entire country. A Master Plan of rural roads under the aegis of the DNGR was drafted identifying 8000 km of rural roads which needed to be rehabilitated as a priority. Funding for the rehabilitation of 5740 km has been secured, and works on 3492 km were completed by the end of 1994. Average annual expenditures on the rehabilitation of rural roads and bridge construction amounted to US$16 million, and cost per km varied between US$18,000 to US$25,000 which is high given the large number of long-term expatriate TAs initially involved in PNIR.19 More than 90% of the program in 1993 (and 98% in 1994) was financed by international donors. For a detailed discussion of the rural infrastructure program, see annex 9. 3.74 This program is yielding high and visible positive impact on rural standards of living as testified by an independently funded beneficiary survey. However, relative long-term cost effectiveness of spot improvement (practiced elsewhere in the sub-region) versus rehabilitation (A la 17 Another entity of concern is the DNFF and its associated organ, the OGUIB (Wood Office). The mandate of DNFF is very broad since it covers the managemerit of all renewable natural resources (except marine) and therefore includes not only the administration of hunting and forestry activities and protected areas management, but also a range of other significant tasks such as soil and water management and conservation, fighting against drought and desertification, urban green areas management, and watershed management. Because of this broad range, DNFF has a staff of 1185 persons, out of which only 173 are foresters - clearly not enough to take care of the Guinean forests -- and 773 are agronomists or agricultural technicians. (See MARA, Activit6s dans le Secteur Forestier, 1ire 4dition, d6cembre 1993). On the other hand, OGUIB, which is in charge of assisting wood processing and commercialization, may be redundant today with the presence of private entities. 18 A Bureau of Development Strategy (now Bureau of Agricultural Policy Coordination, BCPA) has been set up since 1987. It coordinated the elaboration of the LADP and is mandated to monitor public investment programs in the sector. The production of agricultural statistics remains of low Government priority despite the creation of a permanent agricultural statistical service (SNPSA) in 1988. But its funding stopped in 1992 and is contingent upon external donor financing. 19 In addition, while in other countries of the sub-region (e.g., Benin and C6te d'Ivoire) the rural roads program is largely spot improvement, in Guinea, it is full-fledged construction, which costs twice as much. 41 DNGR) needs to be assessed. In addition, although private medium- and small-scale civil engineering firms have been fostered, there may be scope for increasing cost-effectiveness by reducing the role of the public sector in contract administration, using AGETIP types of organizations. This option which has helped reduce the cost of civil works in other Western African countries, should be explored. 3.75 Agronomic Research is undertaken by IRAG in seven specialized research stations corresponding to agro-ecological sub-regions. The research activities are broken down into thirteen areas of research orfilibres de recherche.20 3.76 Total average expenditures for the whole agronomic research program during the period 1989-94 hovered around GNF 3.5 billion (or US$ 4.0 million). The total wage bill amounted to about GNF 1.1 billion, of which GNF 0.2 billion was spent on seasonal and temporary workers for the production and upkeep of trial field and nurseries. Non-personnel current expenses amounted to GNF 0.5 billion; while capital expenditures amounted to GNF 2.6 billion, of which GNF 1.5 billion was for technical assistance, GNF 0.7 billion was for equipment and material, and a little less than GNF 0.2 billion was for training. The economic breakdown of the expenditures on agronomic research shows that 25% of the average annual expenditures during the same period was allocated to personnel, with about 60% to investment. In the latter expenditures category, technical assistance took up 62% (or 37% of the total expenditures on agronomic research). 3.77 The impact of agronomic research has been limited essentially by three factors. The first is the lack of adequately trained scientific staff. The second is extremely weak management systems for finance and accounting, infrastructure and equipment, and staff and research programs. In addition, staff are being taken out of or assigned to IRAG with little consideration for scientific requirements, and staff promotions have little to do with scientific results. Third, because there is no overall master plan, agronomic research is supply driven, either by its financiers or by the researchers. Each of the research stations attempts to address all constraints deemed important by the researchers according to their specific location or areas of expertise. Thus, the 13 areas of research emphasis have been broken down further into 33 research programs, some of which have little relevance to farmers' needs, yet compete for scarce public current funds. Only recently has there been an attempt to prioritize the research agenda, using a weighting procedure to rank crops and research themes according to users' priorities, and to tailor it to demand. The impact of agronomic research has also been limited by the absence of a systematic link with agricultural extension, a key weakness being addressed by the on-going research and extension project.21 3.78 Agricultural Extension. This service has a well-organized and well-motivated work force. However, its effectiveness is dependent upon the availability of technology on the shelves of the agronomic research institute. A study to estimate the quantitative impact of agricultural extension in Guinea is underway and should shed light on the pertinence of the on-going agricultural extension system. World-wide, it is shown that returns to agricultural extension are high: in Guinea, a survey conducted by the monitoring and evaluation unit of SNPRV indicates that farmers in direct contact 20 It is envisaged, after the IDA-supported research prioritization conducted in December 1994, to group them under six major departments: cereals, other annual crops, perennial crops, animal production, enVironment and natural resources, and rural sociology and production systems. 21 In addition to the regular training of the subject matter specialist of the extension service, field contacts (Uniti Experimentale Paysanne), gathering researchers, extension agents, and farmers are institutionalized. 42 with the extension have an aggregate average crop yield 30% higher than their counterparts who are not in direct contact with the service.22 A sizable amount of expenditures on agricultural extension undertaken infilibres, and integrated area development projects is not known. Cost effectiven;ss of the various approaches being an issue, it is important to get data for comparison purposes and to make informed recommendations. For a detailed presentation of the agricultural extension program, see annex 9. 3.79 Livestock Services. The department of livestock services administration (Direction Nationale de l'Elevage, DNE) underwent the most drastic restructuring of all MAEF's technical departments. Livestock services and animal health is a lean organization, after having reduced staff by two-thirds under the IDA funded Livestock Services Rehabilitation Project. Privatization and cost recovery of veterinary services are successful. Functional and economic composition of public expenditures on livestock services are shown in annex 9, tables 9.4 and 9.5. 3.80 Natural Resources Management. This is an area of relatively recent interest and existence (since the mid-1980s), but it has become an important component of the agricultural PIP and it is entirely handled by DNFF. It includes several regional development projects as well as community- based natural resources management projects and forest activities. The total cumulative program in this area is now close to US$190 million, most of which is financed by external donors. More analytical work needs to be done on the nature and efficiency of expenditure in this sector. Policy Options 3.81 With 17% of total public expenditure share, expenditures in agriculture may need to be scaled down. In doing so, it is crucial that the role of the public and private sectors be assessed. MAEF's capacity to define and implement its agricultural development strategy needs to be strengthened, and a core investment program should be identified to ensure that the resources going into the program (including about US$1.0 billion in foreign assistance since 1985) are used to maximum effect. Criteria should be established to decide on the acceptability and priority of investment projects. MAEF should ensure that externally-financed projects fit into the core investment program, that the number of projects becomes more manageable, and that their impact on the ground is commensurate with the experiditures made. Expansion of cultivated area, low and stagnant crop yields, and rising per capita food imports are all indications that the efficiency of investment in agriculture needs to be enhanced urgently. As implementation capacity in MAEF expands and in order to accelerate this process, all projects remaining in the public sector domain should become part of national programs under the line directorates; the establishment of parallel implementation units should be avoided. All of this will require a new partnership between MAEF and international donors. In this context, the role of long-term technical assistance should be reviewed with a view to reducing it over time and replacing it with short-term consultancies where outside expertise is needed. A consensus on these issues needs to be built between the Government and the donor community and reflected in the agricultural policy letter. 3.82 The functions, structure, and size of MAEF must be redefined by taking a critical look at certain functions, such as those carried out by the Service de Contrdle de Qualiti, that were carried over from the First Republic. 22 This finding is consistent with the results reported from an econometric study, Evaluation of T& V-Based Extension in Burkina Faso, by Bindlish, Evenson, and Gbetibouo (World Bank, November 1993). 43 3.83 An agricultural growth strategy should address agricultural intensification and generation (to assess the extent of cultivation expansion and deforestation) and the diffusion of improved cultural practices. 3.84 In the area of rural infrastructure, which absorbs about 40% of public expenditures in agriculture, new modalities of work sharing among DNGR, SNAPE and rural communities, and small private enterprises should be considered. These policy orientations should be spelled out in a policy declaration. To address the issue of overstaffing in MAEF that would result from the above measures, a comprehensive action plan would need to be prepared to ensure a smooth transition and sustainable outcome. In the rural infrastructure, the long-term cost effectiveness of spot improvements versus rehabilitation should be reassessed. Cost effectiveness could be improved by reducing the role of the public sector in contracts administration. 3.85 The current pattern whereby personnel expenditures constitute the vast bulk (97%) of the current budget of MAEF and donors finance 85% of investment expenditures, as well as 80% of current expenditures, is untenable. This situation leads to a completely donor-driven and fragmented investment program, leaving MAEF without the means to pursue its own investment priorities and without the possibility of adequately funding current expenditures of services that are not linked to donor investments. Consideration must be given to reducing the wage bill significantly, thereby freeing resources for other current O&M expenditures and for capital investments. 3.86 More attention should be paid to operation and maintenance expenditures across MAEF and all programs, especially for rural infrastructure and including the management of strict natural resources (gazetted forests), and natural resource management. The anticipated overall public expenditures during the 1996-2000 period for the three high-priority national programs, which IDA currently supports, amount to about US$380 million and are distributed as follows: agricultural services, US$120 million; rural infrastructure US$160 million; and rural water supply US$100 million. The investment share of the forecast total expenditures amount to US$335 million equivalent. Donors have already pledged to support 46% of the rural infrastructure program and about 30% of that for rural water supply; DNGR and SNAPE are actively seeking donor support for the balance of the financing requirement of their respective programs. Unless credible strategies for maintenance- and cost recovery are developed, these further investments in infrastructure especially would only exacerbate the present recurrent cost and O&M crisis. Research, extension, rural infrastructure, and the rural water supply and sanitation program for 1996-2000 would demand about US$27 million for O&M (rural road maintenance alone, US$11 million). G. INTERSECTORAL ALLOCATIONS 3.87 The previous discussion indicates that inadequate allocations to key sectors are in stark contrast to the ambitious objectives delineated for these sectors, and that inefficient allocations within each sector hamper their efficiency. How can the Government reach more realistic sectoral priorities and objectives that are consistent with the current macroeconomic framework, and how can it redress allocation imbalances and define performance criteria for sectoral allocations? Given macroeconomic constraints that seriously limit the extent to which the budget envelope can be expanded, long-term decisions about allocations among sectors, programs, and expenditure categories can be daunting. A critical condition for making knowledgeable choices among sectors is the existence of realistic and affordable long-term economic strategies. 44 Costing Sectoral Strategies 3.88 Estimates of the costs of strategic objectives in the health and education sectors were provided earlier in this chapter; recurrent cost requirements for road maintenance were presented in the previous chapter. Here, we discuss whether these strategies are compatible with the prospective macroeconomic framework. 3.89 In 1994, current expenditures as a share of total current expenditures (excluding interest on debt) were 8% for primary education, 5% for health and 2% for public works (and transportation) -- a total of 15%, corresponding to allocations of GNF 20 billion, 12.3 billion and 6.4 billion respectively. How would these shares change, should the sectoral strategies be carried out and entirely financed? The costs estimated here are for current expenditures only, since these are financed largely by the Government and thus provide a more accurate measure of the financial burden of the strategies on the country. Two cases are examined. The first corresponds to the scenario discussed in chapter 1, whereby the Government's revenue mobilization effort remains neutral, and expenditures adjust to the revenue shortfall to preserve the sustainability of the deficit. The second is the scenario described in the 1995 PFP, whereby the Government makes a substantial effort to improve its revenue mobilization; expenditure and deficit projections are based on the PFP until 1998 and on IMF estimates thereafter. 3.90 Under the first scenario, the Government's neutral tax effort would yield a sustainable growth in expenditures of a maximum of 2.3%. By year 2000, current expenditures should not exceed GNF 320 billion (in real terms). Sectoral financing would require the following: 14% (GNF 37 billion) of current expenditures to health, three times their 1994 level; 21% (or GNF 57 billion) to primary education, or almost three times the 1994 expenditures; and 10% (or GNF 27 billion) to roads,23 more than four their times 1994 expenditures. The total cost of the sectoral strategies would amount to 45% of current expenditures, leaving only 55% for all other expenditures. 3.91 Under the second scenario, the Government's tax effort to meet the PFP targets and beyond would yield sustainable growth in expenditures of not more than 6.9 % in real terms; by year 2000, current expenditures would amount to GNF 421.0 billion in real terms. Sectoral financing would be as follows: 10% of the current budget to health, 15% to education, and 7% to roads. The total cost of the sectoral strategies would amount to 32% of current expenditures, leaving 68% for all other expenditures. These scenarios are displayed in figure 3.5. 23 The percentage for roads is an underestimate, because it includes only current expenditures for road maintenance. Administrative costs and the cost incurred by the Ministry of Public Works and Transport are not included. 45 Figure 3.5 3.92 These charts indicate clearly that, Costs of sectoral priorities regardless of its revenue performance, major reallocation of resources to these sectors will be 19 required if the Government is seriously 1994 - Current expenditures committed to pursuing its respective sectoral as% strategies. In addition, the discussion in chapter 2 8% has raised the necessity of reducing capital expenditures. Should these decline by 5% or 10% s% in the next three to five years as suggested, major 2% reallocations will be all the more called for, and it Primary Education would be absolutely imperative that key sectors Ro maintenance (i. rural roads) be given complete protection from budget cuts. U Others This raises two issues. 2000 - Scenario 1: No tax efffort 3.93 The first is whether or not the sectoral 21% objectives are realistic and affordable. To commit to this objectives, the Government would 14% have to clarify and promulgate an understanding 56% of its long-term development strategy and the role of sectors. However, it is unlikely that the ePrimary Education medium-term cost of these strategies has been DRoad maintenanc (incl. rural roads) estimated, let alone discussed and approved by E Others the entire Government. In fact, it is probable that if all sectors were to produce similar strategies, 2000 - Scenario 2: Tax effort their sum would far exceed 100% of the budget. 16% (Chapter 5 provides an institutional perspective ,....= I 10%I o on how the MOP should assess sectoral strategies and translate them into programs and budgets 7% within a consistent macroeconomic framework, as 67% well as how the MOF should respect priorities in aePrimary Education budgetary allocations and implementation.) IoHealth. BRoad maintenance (incl. rural roads) Second, it is unclear how the Government a Others conceives Guinea's long-term economic development and poverty alleviation strategies or defines the role of the sectors most crucial to meeting its strategic objectives. In the absence of standard procedures for designing credible and affordable sector strategies, it is difficult for the Government to allocate expenditures rationally. Chapter 5 discusses policy options pertaining to intersectoral allocations of expenditures. 46 4. THE INSTITUTIONAL ENVIRONMENT AND THE BUDGETARY PROCESS 4.1 After an initial period of adjustment when basic structural and regulatory reforms were adopted, the reform effort slowed down and was insufficient to complete the transition to a market- based economy. Key among yet-to-be achieved reforms is a clear definition of the public sector's role. In the absence of clear delineation of its responsibilities and functions, the public sector remains involved in a broad number of sectors; public institutions have grown in number and in size, while expenditures, and particularly the wage bill, remain difficult to control. 4.2 Effective control of public expenditures in Guinea has been limited by a weak institutional structure, slack managerial direction, and competing sectoral policies and interests. A major impediment to sound budgetary programming is the budget process itself -- in which capital and current budgets are the responsibility of two separate institutions, with weak analytical links between the budgets. The absence of affordable sectoral strategies, a dearth of technical skills and information systems, and limited performance evaluation criteria -- all have adverse consequences on the sustainability of investment. As a result, the budget does not reflect a coherent set of comprehensive sector programs. The budget preparation process is also open-ended. Until 1995, final budget allocations and expenditure priorities were decided by a few individuals without proper regard for O&M expenditure needs or the managerial autonomy of the sectoral ministries. Ministerial tension for control over the investment budget has intensified in the past few years. The result is a poor institutional framework for supporting longer-term economic and social development in Guinea. A reform agenda towards improving the institutional environment and budget process is discussed in chapter 5. Several issues are key: Institutional Environment * Institutional responsibilities are not consolidated. The numerous institutions with overlapping functions compete for their own interests. * The system for controlling and monitoring public expenditures lacks autonomy, and is characterized by insufficient technical expertise. Budget Preparation Process * The Government lacks a clear medium-term macroeconomic framework establishing the costs of present and alternative policies. * The Government's capacity to assess sectoral needs, evaluate projects, and prepare strategic plans is limited; allocations in the current budget are frequently ad-hoc, those of FINEX reflect mostly donor priorities, and those of BND are open to political imperatives. * The macroeconomic framework, defined in the budget letter, lacks credibility. In addition, no ceiling is imposed on sector budgets, making budget preparation an open-ended process. * Until 1995, final budget decisions were concentrated in just a few individuals. 47 * Until 1995, the current budget and PIP were prepared separately, exacerbating the recurrent cost crisis. * Information systems are weak; previous budgetary performance cannot be used as a guideline for future budgetary allocations. Budget Implementation and Monitoring * Too many players are involved in the expenditure approval process, leading to delays and rent- seeking behavior. * Expenditures do not necessarily reflect budget allocations, hampering the credibility of the budget as a management tool. * Accounting procedures are not always respected, leading to questions about transparency. * Due to delays and changes in priorities, procurement procedures are inadequate. * Information on the PIP implementation of the line ministries is not circulated among those ministries properly. Civil Service and Pensions * The. framework for civil service management is not based on affordability, outcome, and clear definition of functions, and information on the civil service is incomplete. * The pension system is overly generous and management is loose; as such, pension transfers have skyrocketed. Management and Information Systems * Timely, reliable information is absent, because information is poorly circulated, data gathering and processing capabilities are weak, and an integrated budget management system has not been implemented. A. THE CURRENT INSTITUTIONAL STRUCTURE: NUMEROUS AND OVERLAPPING FUNCTIONS 4.3 For the past ten years, the Government's response to institutional performance has been to replace ill-functioning institutions with new ones, but without dismantling the existing ones. (Annex 11 describes the shuffling that has taken place since 1988.) As a result, the institutional landscape now consists of a large number of institutions with imprecise, overlapping mandates. Particularly with respect to the PIP budget, this confusion has given rise to arbitrary expenditure decisions, because none of the institutions is held accountable for performance. 4.4 A new Government reorganization in August 1994 was the third one in ten years, changing the major players in the budgetary process. The Ministry of Finance and Plan, which had been consolidated in 1992, was again separated into two structures -- the Ministry of Finance (MOF) and 48 the Ministry of Plan and Cooperation (MOP). The mandate of the new Ministry of Finance was clarified: in addition to its revenue mobilization functions and current budget preparation and implementation responsibilities, it again became the only institution with authority to approve expenditures under the PIP budget. The Ministry of Plan and Cooperation assumed responsibility for external grant management within its National Directorate of Cooperation (DNIC), formerly part of the Ministry of Foreign Affairs. The preparation of economic management strategies, the collection of data, and the production of economic data were devolved to the Ministry of Plan, and specifically to its National Directorate of Public Investments (DNIP), whose mandate was limited to preparing the PIP budget, participating in negotiations of external financing for public investment, and following up on project implementation. In reality, the roles of the Ministry of Plan and the DNIP were greatly reduced by the creation of a new institution -- the Administration and Control of Large Projects (ACGP).' In part to maintain its involvement in the PIP implementation process and in part to challenge the predominance of ACGP, the Ministry of Plan recently created a Project Evaluation and Follow-Up Unit attached to the Office of the Minister, with the rank of a Directorate. In addition to analyzing implemented programs and projects in the field, the Unit is responsible for establishing a comprehensive information system that reports on the status of PIP programs and project implementation. B. CHECKS AND BALANCES: ACCOUNTABILITY, TECHNICAL EXPERTISE, AND MONITORING CAPACITY 4.5 The system for controlling and verifying public expenditures consists of three levels, each with its own shortcomings and weaknesses that, despite attempts to correct them in the past, still persist. One level of control exists within the Ministry of Finance -- with the General Control Inspection. This group is responsible for performing periodic checks on the authenticity of budget expenditures both during and at the end of the fiscal years. Its mandate, however, is limited to overseeing the Ministry of Finance. Until recently, the Executive level performed a similar task with its Secretariat of Financial Control; its mandate, however, extended to the entire public administration. However, both institutions proved to be ineffective. The absence of an objective assessment of the adequacy of the budgetary process has given rise to abuse and a lack of transparency, as no Government institution is held accountable for its actions. 4.6 Another level of oversight exists within the Government at the Ministry of Economic and Financial Control. The Executive created the entity in late 1994 in response to the Government's ineffective management of public resources. The new ministry absorbs the former Secretariat at the Presidency. Its mandate is to perform periodic verifications of expenditures as budgets are implemented. All central, decentralized, and public agencies are subject to scrutiny. But its ability to discharge its responsibilities is constrained by weak technical and financial expertise, and a dearth of operating resources. 4.7 The third level of oversight rests outside the Executive, in the Chambre de Comptes, one of three chambers of the Supreme Court. The Chambre de Comptes is responsible for the ex-post oversight of the accuracy and justification of expenditures in Government accounts. Its authority extends to the entire public administration. But it is impaired by substantial lags in the submission For more information on the ACGP, see annex 12. 49 of public accounts by the Ministry of Finance. Indeed, the Treasury is liable to the Chambre des Comptes for fines that currently amount to an estimated GNF 50 billion. It is also limited by the weak financial expertise of its personnel -- limited funding to carry out independent audits if the review of Government accounts so warrants; and inadequate dissemination of its audit results. Until recently, in the absence of a Legislative Assembly, the Supreme Court was required to submit its findings to the Executive, which did or did not choose to apply sanctions for abuse or excess. Without broader dissemination of reports subject to public opinion, little could be done to establish the accountability of institutions and individuals. This situation may change substantially with the newly elected legislative body, as the regulatory framework of the Supreme Court begin to converge with the mandate of the Legislative Assembly. Early signs of improvements in the situation remain to be assessed. C. THE BUDGET PREPARATION PROCESS 4.8 The budgetary process involves three levels of planning. The first captures the overall macroeconomic policy objectives of the Government and the services that it wishes to provide. The second captures the total amount of Government expenditures and how they should be financed. The third captures the amount of expenditures to be devoted to each sector, ministry, public agency, or project. In Guinea this fragmented budgetary process remains one of the major impediments to efficient public resources management. 4.9 The absence of an appropriate macroeconomic framework, strategic objectives, and affordable, realistic sectoral strategies and the lack of a binding sector budget cap have made budget preparation an open-ended process. In addition, overbudgeting in relationship to historical revenue performance prevents budgetary targets from being met. In the absence of sectoral priorities, the PIP reflects primarily the perceptions of donors about needs and preferences. In addition, weak project and sectoral evaluation capacities by the DNIP and the line ministries undermine the quality and viability of projects, particularly projects financed by the Government. 4.10 The root of the problem is that the institutions responsible for preparing the current and PIP budgets -- key elements of efficient public resource management -- view them as separate documents that bear no relationship to each other. Only late in the process do relevant officials from both ministries meet to finalize the budget. The result is a poor analytical link between capital and recurrent expenditures, with grave implications for the sustainability of investment projects. Most budget proposals must be scaled-down to more realistic levels before they are presented to the Executive for approval. In the absence of strategic priorities, decisions to cut budget items are often 2 capricious, without consultation among the parties involved. 2 An example of this problem was the negotiation with the IMf on a second-year ESAF in November 1994, which resulted in a substantial cut in proposed budget levels and arbitrary decisions as regards the line items that were affected by these reductions. The hastiness of decisions made at that time invalidated various understandings reached with line ministries during the preparation process, provided for cuts in areas that should have been protected, and undermined the Government's credibility. 50 The Budget Process and Cycle 4.11 The first known formal budget was prepared in 1988, or thirty years after independence, and it was not until December 1994 that a budget was approved prior to the beginning of the fiscal year. In January 1991, the Loi Organique Relative aux Lois de Finances defined the responsibilities for budgetary preparation and implementation, and established the modalities for control over debt contracting, procedures for the commitment of expenditures, ceilings on the wage bill, and the size of the civil service, as well as a system of monthly spending plans. 4.12 The Loi des Finances consists of three parts relating to: (a) revenues; (b) current expenditures including public debt, personnel expenditures, operating expenses and transfers; and (c) investment expenditures. The investment budget itself is prepared in two parts: one which contains externally-financed projects (FINEX); and the second, the BND, which includes the Government's participation in the financing of FINEX projects, plus investments which are undertaken and financed solely by the State. Nevertheless, as the external account deficit is financed in large part by grants, projects which are financed by grants and which are managed directly by donors are excluded from the investment budget and are reflected only in the balance of payments. Thus, the public investment budget is only a partial reflection of total public investment in Guinea. The Ministry of Finance is responsible for preparation of revenue provisions and for the current budget. The Ministry of Plan is charged with the preparation of the investment budget. The budget is inclusive in the sense that Government revenue is for the most part accounted for in the budget. While in the past an important part of mining revenue was kept in separate accounts that enabled the Government to undertake expenditures which were not included in the budget, this practice has been abandoned. 4.13 Guinea's fiscal year begins on January 1 and ends on December 31. Nevertheless, the budget year is not closed until end-March. The Ministry of Finance and the Ministry of Plan each send out circulars in late June of each year describing the methodology for budget preparation, and requesting inputs from line ministries and, in the case of the PIP budget, from the donors. The circulars include a schedule of meetings with line ministries. Budget preparation is initiated in mid- October, at which time negotiations between line ministries and the above-mentioned institutions take place. The Ministry of Finance supervises preparation of the current budget and the Ministry of Plan does the same for the PIP budget. Both ministries finalize their budget proposals in November. Joint discussions of the draft budgets then take place in mid-November between those two ministries and the Economic and Financial Coordination Committee (CCEF). Until 1995, in the absence of a Legislative Assembly, the budget was approved by the Executive at a Ministerial Council meeting, normally held in late December. 4.14 Inputs for the preparation of both parts of the budget are provided at the beginning of the budget cycle by the line ministries; and, in the case of the investment budget, by donors, who provide estimates of expected expenditures for externally-financed projects. Since 1990, the Government has initiated a process of decentralization, through the introduction of specific budget line items for regional authorities. Regional institutions are requested to provide expenditure inputs at the beginning of the budget cycle. Inputs, however, are sporadic and often incomplete. Consequently, decisions on the level of resources allocated to regional authorities are made at the center. 51 4.15 In order to improve expenditure control, the Ministry of Finance determined and imposed quarterly ceilings on each budget line in 1995. Although each spending ministry is allowed to choose its spending priorities within this limit, this constraint limits management of sector budgets by line agencies. In addition, the Ministry of Finance has regained total managerial control over the entire budget for material and supplies. At the time of the budget publication, spending ministries were reportedly not informed on the amount of their allocation. Macroeconomic and Sectoral Strategies 4.16 The Government lacks a complete medium-term macroeconomic framework: although the PFP provides for balanced macroeconomic accounts at a three-year horizon, the Government's sectoral policies have not been quantified, nor has their consistency been tested against the expenditure framework. In that context, the Government has no framework for assessing the costs of both its current policies and alternative ones. As discussed in the previous chapter, the ability of line ministries to produce these strategies is uneven, ranging from well-prepared objectives that are too ambitious or bear no relationship to the capacity of the Government to finance them to none. The DNE and DNIP at the Ministry of Plan, who are responsible for the review and subsequent translation into budgetary objectives and priorities consistent with the macroeconomic framework, lack the capacity to analyze the viability of these strategies and their relevance to sectoral objectives and priorities in the budget. 4.17 The macroeconomic framework and sectoral strategies also form the basis for the three-year rolling PIP. The scattered configuration of projects in the Plan prevents the use of it as an effective planning instrument: the Plan includes projects that are managed by the Government through the line ministries. It also includes projects for the benefit of decentralized public or private enterprises under loan retrocession contracts. In addition, the absence of well-established sectoral levels of recurrent and investment spending gives rise to a disjointed budget preparation process. 4.18 Given the fact that the PIP is the only significant instrument through which the Government expresses its broad economic objectives and priorities, the PIP planning process should seek stable sectoral allocations over successive years. However, wide BND shifts in the past three years raise skepticism about the existence of a global vision for Guinea's development, and thus the relevance of the PIP as a planning instrument. 4.19 In Guinea, institutional and financial constraints are the major obstacle to the timely articulation of a sound macroeconomic framework for the annual budget preparation, the first steps in the budget cycle. At the time of budget preparation, several important elements are missing. The first and most important is a viable macroeconomic framework which provides realistic economic growth objectives and defines strategic priorities to be supported by budget allocations. Another is comprehensive information about the previous year's budget performance, particularly as regards the PIP budget, which is key to understanding how much the Government should spend, and how that amount should be apportioned to the various sectors and institutions. 3 For a complete discussion on FINEX and BND allocations, see the Country Economic Memorandum (CEM), para. 2.26-2.28, World Bank, October 1995. 52 4.20 In principle, the Directorate of Plan and Economics (DNPE) of the Ministry of Plan is responsible for producing data to form the basis for the budget's macroeconomic framework and its preparation. Three divisions in the Directorate coordinate efforts to elaborate the macroeconomic framework. One, the Forecasting Division (DP), is supposed to produce short-term macroeconomic projections twice a year as the basis for the macroeconomic framework after approval by the CCEF. But given a dearth of DP staff, this role has been transferred to the Division of Macroeconomic Studies (DEM). In turn, this division lacks the financial resources necessary to carry out surveys and produce data, and it generates crucial economic information with as much as a two-year delay, or not at all. 4.21 The macroeconomic framework produced by the MOP in July lacks credibility in the eyes of the MOF, which awaits the arrival of the IMF and the Bank towards the end of the year to formalize the macroeconomic framework, at a time when budget negotiations with line ministries have long been completed. This results in budget proposals that must be scaled down. In the absence of strategic priorities, decisions on cutting budget items are made in an ad-hoc manner and without consultation with the parties involved. Preparation of the Current Budget 4.22 Preparation of the current budget starts with the transmission of the budget letter from the Ministry of Finance to the line ministries. Until 1995, the budget letter defined the overall spending constraint of the macroeconomic framework but imposed no ceiling on sector budget proposals, making budget preparation an open-ended process. Absence of quantified and affordable sector strategies, clear intersectoral priorities, and the limited capacity of the MOP undermine the translation of sector objectives into programs and programs and into budget proposals. 4.23 In addition, determination of a meaningful budget is hampered by the line ministries' limited capacity to assess their needs because resource management is not based on outcome criteria. The current budget proposal is in most cases a function of the previous year's budget to which arbitrary requirements are added based on inflation rates or other considerations. Disbursements of the previous or current year are not used as yardsticks to measure spending capacity and priorities. This process may change in the future since the budget will have to be voted on by the Legislative Assembly. As a result, budget proposals will be subject to more attentive scrutiny by the Ministry of Finance and the line ministries. A first indication of this effort is the preparation of the 1996 budget on a zero basis whereby every budget line has to be justified. 4.24 Additional obstacles to efficient budget preparation in the line ministries are the absence of key accounting tools, such as stock accounting, and control over personnel management in the hands of the Ministry of Civil Service and the Ministry of Finance. Preparation of the Investment Budget 4.25 The poor coordination between the Ministry of Plan and Cooperation and the Ministry of Finance places constraints on annual investment budgets, which in turn precludes integrating revenue performance and cash management constraints in PIP budgetary provisions. The absence of formal linkages between the PIP and the current budget contribute to rendering the budget inefficient as a management tool. 53 4.26 Another problem is a deficient information system, which precludes a full analysis of previous performance of physical and financial implementation of projects as a basis for budget allocation proposals. In this context, the absence of comprehensive information on external grants and on "grant" projects that are managed by donors is a major impediment to efficient investment planning. These problems have prompted the DNIP to rely on information provided by donors about the expected implementation of individual FINEX projects as the basis for the FINEX budget, and to establish a limit on overall BND expenditures that is consistent with revenue considerations. As a result, the internal composition of the BND has become less relevant. 4.27 Additional criteria for compliance with the macroeconomic and sectoral strategies have been added to form a basis for accepting projects in the PIP.4 However, the DNIP, which is responsible for preparing the investment budget, does not analyze the intra- and intersectoral problems that may arise in reviews of sectoral strategies and related investment. As a result, the DNIP's main criterion for including a project in FINEX is the availability and security of the project's financing. Uneven collaboration between donors and the Government in project preparation forces the DNIP to rely on this criterion. 4.28 As no cap is fixed on sector budget levels, the practice in sector ministries is usually to submit the entire list of existing and new projects to the Ministry of Plan. Ensuing negotiations with the Ministry of Plan are supposed to be carried out collectively for a sector but occasionally some negotiations occur directly between the MOP and the project director. The resulting agreement on composition and level of sector investment can be invalidated by the final arbitration and is always changed during the budget revision. D. BUDGET IMPLEMENTATION AND MONITORING 4.29 Current budgetary levels are insufficient to support the normal operation of Government agencies and to maintain the current level of public investment, with adverse consequences on the efficiency of public administration and the quality of Government services. To arrive at a sustainable budget, the Government must analyze the full financial implications of determining the real operating expenditure needs of public administration under its current institutional structure and the size of its labor force, as well as the operating and maintenance expenditure needs of the public investment program. It must then make difficult decisions about its spending priorities. As troubling, however, is that even the present expenditure targets are too high in comparison with revenue performance. The Time-Consuming Budget Implementation Process 4.30 The process for approval of budgeted current expenditures is relatively simple, with only the Ministry of Finance, line ministries, and other recipient institutions involved in the process. Approval and payment of expenditures under the PIP budget is extremely complex and lengthy, 4 These include: (a) the availability of all feasibility studies; (b) a rate of return of at least 10%; (c) secured external or local financing; and (d) the capacity of the State to finance recurrent costs after the project's completion. 54 involving both the Ministry of Plan (until recently) and the Ministry of Finance at various steps in the process. 4.31 In Guinea, budget implementation extends over a period of sixteen months, because of the difficulty in complying with the requirements of the regulatory framework of public accounts -- that a line of credit for the beneficiary institution be opened for each line expenditure envisaged in the budget; that expenditures be committed, verified, and authorized before payment; that authorization to pay not exceed the total budget credit available for the year; that credits not be paid before services are rendered; that payment be made by cash, check, or direct deposit in a bank; and that any payment be stopped or suspended in part or in total if documents to justify a previous transaction contain irregularities. 4.32 While the legal framework for budgetary accounting is very clear, practices are occasionally disorderly or lack rigor. Two prime examples are compensation operations and the reconciliation of arrears, and justification of expenditures in the provinces. Information on key transactions are missing: Treasury operations and the Central Bank lack corresponding accounts, and the accounting of noncash, or administrative, revenue by the line ministries is sketchy. 4.33 While the public procurement code is adequate in principle, the number of institutions involved in the approval process and the number of steps required to complete the process have led to delays of about one year for the signature of large contracts, and of about three to five months for small contracts. 4.34 The National Directorate of the Budget (DNB) is responsible for the commitment and authorization processes, and the National Directorate of the Treasury (DNT) for payment of expenditures. The first is an administrative function; the second is a public resource management function that includes the payment of public expenditures, which gives it relative autonomy and immense power. Numerous structures within the DNB and the DNT, coupled with innumerable bureaucratic steps, make the budget expenditure approval process extraordinarily lengthy, particularly for the PIP. This process translates into substantial payment delays -- some of which the sectoral ministries believe are intentional, as a way to save resources for other uses -- and lead to disruptions in programs and project implementation. And during liquidity shortages, the lengthy procedures cause a significant accumulation of "technical arrears."5 But most of all, the excessive number of "control" stations promote illicit payments and rent-seeking behavior. 4.35 Due to delays in budget implementation and ad-hoc decisions on disbursements, budget allocations are not reflected in expenditures, impeding the budget's effectiveness as a management tool. Examples shown in chapter 2 illustrate this issue, the chief one being disbursements on road maintenance (GNF 1.2 billion), which amounted to less than 50% of the budgeted amount in 1994. Similarly, one of the main examples of this same issue from chapter 3 -- spending on education from 1991 to 1993 -- shows that whereas the budget for wages and salaries has an implementation rate of 93%, non-wages are realized at 81% and transfers at 72%. Also, disbursement rates by level of Their accumulation is due to the time lag between commitments and end-of-year disbursements. 55 education show that the priority was given to tertiary education in 1992 and 1993. The budget for primary education was disbursed at 48% while 79% of the tertiary education budget was disbursed.6 The Current Budget Implementation 4.36 For a description of current budget implementation procedures, see annex 11. In the past several years, weak revenue performance has required that the budget be revised and reduced at mid- year. Budgetary revisions, in turn, have prompted the Government to establish a hierarchy of priority expenditures that are paid at the expense of others. They include the wage bill; the debt service of external nonrescheduled debt and the interest payments of rescheduled debt; the expenditures of the Ministry of Foreign Affairs (diplomatic missions); and expenditures for the defense of the country's sovereignty, including the military. Nonwage expenditures outside these areas of priority have suffered as a result. In particular, the Government's nonpayment for the use of public utilities (water, electricity, and telephone) has led to an accumulation of arrears to the utility enterprises, with dire consequences for their financial sustainability. 4.37 This type of expenditure prioritization (following the economic composition of expenditures) results from the fact that the budget is not conceived of as a set of integrated and coherent vertical programs (following a functional composition of expenditures), comprised of related salaries, O&M, and investment expenditures. If it were, expenditure management would lead to defining priorities among programs, and expenditure cuts, when necessary, would more evenly affect the different expenditure categories. This type of management would secure full disbursement of public programs, greatly enhancing the efficiency of public expenditures. PIP Budget Implementation: "Les itapes du Tour de Guinie"7 4.38 The Ministry of Finance does not manage the payment of PIP expenditures under FINEX. Instead, PIP expenditures under FINEX are made by a local commercial bank or directly by the donor agency after the line ministry or implementing agency has certified that the goods and works or services have been delivered, or, for direct payments outside Guinea, after the Minister of Finance has submitted a request for direct payment to the contractor by the donor, accompanied by the necessary supporting documentation. The Central Bank records these transactions in the balance of payments. Donors regularly provide information on financial and project activity to the line ministries or implementing agencies and to the DNIP at the Ministry of Plan and the DNB at the Ministry of Finance, but the information is often incomplete. 4.39 The procurement process is one of the major impediments to the pace of FINEX investment implementation. The contract bidding and approval process for the procurement of goods and works or services often requires that creditor-specific procurement rules be applied, and frequently requires international/country-specific competitive bidding. Until the recent past, the line ministry and/or implementing agency was responsible for opening the bid process. The DNIP helped the implementing agency initiate the process at this early stage. Bid analysis was carried out by the line 6 These figures are based on ordonnancement for the MEPUFP and mandatement for the MESRC. For a description of the expenditure approval process, see annex 11. 56 ministry or implementing agency with oversight from the DNIP at Plan, the Directorate of Procurement, and the DNB of the Ministry of Finance. For contracts of more than GNF 1 billion, a procurement committee at the Executive office (whose chairman was appointed by the President) was charged with evaluating bids and selecting contractors. The Minister of Finance signed all contracts. The various players often had established contact with the bidders, which biased them towards one or more of the bidders. The bid evaluation and selection process became protracted, requiring the direct intervention of donors to resolve an impasse or to reinitiate the process. The committee has been replaced more recently by the ACGP, which is expected to coordinate its work with the Directorate of Procurement and Portfolio Management of the Ministry of Finance. But, as is discussed further in annex 12, the ACGP has been given broader powers, allowing it to select the contractor and to sign procurement contracts. ACGP will also be responsible for managing implementation of procurement contracts. 4.40 In contrast to FINEX expenditure payment, the payment of expenditures under BND is cumbersome because it is subject to Government procedures, in which a request for payment must pass through many institutional layers before it is authorized and paid. In light of the mid-year budget revisions already mentioned, the BND has become a major casualty of budgetary underfunding (as discussed in chapter 2). Government efficiency also has an impact on BND implementation, given that part of its financing should come from local currency generated by foreign grants for the Government's contribution to FINEX investments. This is the case with grants from the European Union, the United States, Canada, and France. These local currency funds are deposited in accounts at the Central Bank, but the Government must officially decide how these funds are to be used, and provide for their early release for BND financing. Despite repeated requests from the donors, these funds have not always been made available for projects. 4.41 BND expenditures are paid under various procedures. Normal budgetary procedures apply to most expenditures and are long and complex. The payment of the Government's contribution to the FINEX project financing (which normally covers operating expenses) is often processed on the basis of derogated procedures -- that is, payments are provided up-front, while the services are rendered later. This procedure applies largely to works or services rendered by force account. This last procedure poses some problems with "regularization," since line ministries or implementing agencies are not able to submit the necessary documents of proof of the validity of the expenditure to the Ministry of Finance. Given these problems, the DNT is reluctant to approve expenditures under this procedure. A different problem can occur when expenditures reportedly exceed credit lines. For example, evidence from the 1993 accounts suggest that some sectors and some types of expenditures have had very high payment/order ratios (for example, a 621% agriculture investment expenditure), without accounting reconciliation afterwards. Incentives for the Line Ministries to Improve Their Management Skills 4.42 The line ministries and implementing agencies that are responsible for delivering necessary social services to the population are largely left out of the process. Centralized control has prevented these institutions from delivering services effectively and is one of the major justifications for reaching an understanding of the need for accountability. It has also given sector ministries and other agencies little incentive to improve their budget management skills. Sector ministries' incentives for improving budget preparation is undermined by the fact that final allocations are decided by a limited number of individuals in the central agencies. Similarly, budget revisions and 57 expenditure cuts are decided without consultation with the concerned parties. Line agencies have no control, and usually little information on engagement or disbursement of their expenditures, and their autonomy has been further eroded by the imposition, since 1995, of quarterly ceilings on all budget items. In addition to their inability to ensure financial monitoring of projects, sector ministries are ill-equipped to carry out physical inspection of projects. Inspection missions to project sites, which can be joined by the Ministry of Plan, are infrequent and take place only on request from the donor, the project director, or the Directorate in charge of the project in the line ministry. In 1995, the Ministry of Agriculture has not conducted any mission, while it is reported that only two missions have been carried out by the Ministry of Health since 1990. In the absence of a harmonized set of criteria and procedures for evaluation of project management, inspection missions often prove ineffective. E. CURRENT BUDGET PROBLEMS WITH CIVIL SERVICE MANAGEMENT AND PENSIONS8 4.43 Almost 50% of current expenditures are allocated to the wage bill. Tensions among line ministries for scarce nonwage resources are fierce. In the current fiscal environment, it is almost impossible for any one ministry to obtain adequate nonwage operating expenditure provisions. But the heart of the problem with current expenditure budgeting in Guinea is the predominance and the management of civil service wages and the public pension program. 4.44 The legislative framework for the public service provides for a model of internal organization which requires each ministry to have a predetermined minimum number of directorates, divisions and sections, as well as a staffing plan. The framework (Cadre Organique) which is used as the key personnel management tool, does not provide a clear description of tasks, procedures and relations between units and individuals. Most importantly, it does not define measurable objectives for the civil service units and, as a result, cannot serve as a basis for assessing resource requirements. In addition, the framework adapts to the pressure to hire or regularize personnel and requires regular updating after each institutional change. This lengthy and tedious process often leaves a ministry without an updated structure reference. 4.45 The key impediment to effective civil service management has been the limited capacity of the Ministry of Civil Service Reform to maintain up-to-date information on the status of civil servants. The ministry has in fact recently made efforts to keep all of its personnel records up to date, and has completed the process for personnel in Conakry; the results are now being analyzed. This exercise should be extended to the rest of the country and implemented as a routine practice of the ministry. Poor coordination between the Ministry of Civil Service Reform and the Ministry of Finance is another impediment, as are political pressures to increase new hiring. 4.46 A related issue in civil service management is the public pension program (discussed in chapter 2), which is a large recurrent expenditure cost that will likely be unsustainable in the future. All permanent Government staff are entitled to a pension, regardless of their classification. Civil servants in Guinea do not contribute to their pensions; in other countries in the region, civil service For a complete report on civil service management, see Guinea: Civil Service Reform Management Review, World Bank, September 1994. 58 employees make a pension contribution from their monthly paycheck that is matched by an equal contribution from the Government. Public and private enterprise employees in Guinea are required to contribute 6% of their monthly salaries to the CNSS, matched by an equivalent amount from the enterprises. 4.47 Transfer allocations for the public pension program are also growing rapidly, requiring an urgent analysis of administration and management issues before the burden on fiscal revenue becomes unmanageable. Two problems are key. One is the existence of public enterprise retirees in the public pension system. Questions have been raised about whether worker contributions collected under the CNSS system are invested appropriately to generate real returns, and whether these returns are consistent with the underlying liabilities of the system. Indeed, there is a real danger that these liabilities are being transferred to the public system as the CNSS becomes less sustainable. 4.48 The other problem pertains to the implementation of pension entitlement rules under the public pension system, which is the responsibility of the Ministry of Civil Service Reform. Because the ministry has not updated records of civil service personnel in the last two years, it has not taken action to identify personnel who have reached mandatory retirement age. As the civil service ages, the prevalence of birth record forgeries appears to be growing. More recently, the ministry has been pressured to hire university graduates, and action has been taken to identify pensionable personnel to allow these graduates to be hired in light of the wage bill allocation limits established by the Government's ongoing economic reform program. F. BUDGET MANAGEMENT AND INFORMATION SYSTEMS 4.49 The current state of information gathering and dissemination requires immediate attention. Crucial data for prioritizing public expenditures are not available, impeding the development of strategic approaches for expenditure planning, budgeting, and implementation. Difficulties met in the preparation of the PER for gathering a consistent and coherent database attest to this problem. These issues were noted in methodological remarks at the beginning of each chapter. Of particular importance are: (a) the absence of externally-financed projects from the PIP, making disaggregated data on the PIP budget and implementation meaningless for analyzing the composition and evolution of sector investment in the country; (b) the inadequate information-sharing between the MOP and the MOF on PIP disbursements and the absence of information on BND disbursements by category of expenditure; (c) the absence of coordination between sector ministries (such as MOH, MEPUFP, and MESRSC) and the MOF in preparing budget and expenditure data; (d) the absence of disaggregated information on expenditures by sector and ministry. 4.50 The PIP information system, in principle, contains all the elements necessary for providing a good PIP follow-up that could lay the foundation for preparing the PIP annual budget. However, this system is available only to the DNIP, and information from the Treasury and DNB to the DNIP is stalled. Consequently, the PIP system cannot be utilized to its fullest. Although the Division of Public Debt runs a comprehensive system of computerized debt management, it is constrained by the absence of timely information from other structures of the Ministry of Finance and by limited participation in the decision-making process. The DNB also has a computerized system for budget preparation and implementation, but it, too, is limited by the lack of information on budget transactions, which is the domain of the DNT. Operations on some expenditures are not well documented. Because the activities of the DNT are not computerized and because a satisfactory 59 system for classifying and filing budgetary expenditures has not been established, the DNB spends an unusual amount of time trying to reconcile authorized versus actual payments. A monthly statement of treasury transactions is prepared as a result. Nevertheless, the monthly statement of treasury transactions lacks essential data for linking expenditure transactions to specific budget credit lines, programs, or projects. The payroll is computerized and its system is compatible with that of the Ministry of Civil Service Reform, but the public pension system is not. 60 5. POLICY AND STRATEGIC OPTIONS FOR INSTITUTIONAL REFORMS AND BUDGETARY PROCESS A. ISSUES 5.1 In order to address the constraints on economic and social development and bring the efforts of the last ten years to fruition, Guinea faces an ambitious reform agenda. Key to its success are the following reforms : * Enhanced revenue mobilization to ensure adequate service delivery to the population, fiscal viability, and balanced economic growth; * Support to private sector growth and economic diversification (particularly since an increase in public investment may remain limited in the medium term); * Further concentration of the public sector's role towards policy management and application of a predictable and consistent incentive framework to support the private sector, the provision of social services, and the development and maintenance of economic infrastructure; * More efficient expenditure allocation: (i) within the economic composition of the budget: capital and recurrent expenditures need to be balanced to ensure sustainability of investments, and reallocations between wage-related expenditures and O&M are needed to secure adequate functioning of the administration and maintenance of investments. (ii) Intersectoral allocations need to be redirected towards basic services and other areas justifying public sector intervention; * A transparent and accountable institutional environment reflecting a clear role for the public sector, and budget procedures allowing for meaningful and efficient use of the budget as a management tool. 5.2 This chapter will examine the Government's options towards achieving these objectives. The Government may want to use these proposals as a basis for discussion and, if need be, formulation of a coherent set of alternative strategies. A reform agenda would inevitably place, difficult choices before the Government, even if the reality of political life differs and often waters down the best-formulated plans. The belief, however, is that only a comprehensive set of in-depth reforms will achieve durable improvement in public expenditure management, as past experience suggests that partial reforms cannot sustain the pressure of the overall environment. 61 B. WHY SHOULD THE GOVERNMENT RAISE REVENUE? 5.3 As seen in chapter 1, the Government's efforts at providing a viable fiscal framework and promoting private sector development will be key to Guinea's stabilization, as adequate provision of public services will require higher levels of expenditures. In the context of the PFP, the Government has committed itself to taking steps towards this objective. These are discussed in the following section. The Importance of Raising Revenue 5.4 Because international mining prices are beyond the control of the Government, the sustainability of expenditures will increasingly be determined by the Government's efforts at mobilizing domestic revenue collection. In July/October 1995, the Government, the IVF and the Bank jointly prepared a PFP that called for sustained growth in public expenditures supported by an annual 12% annual increase in domestic revenue between 1995-1998 in real terms (and 11% between 1995-2000 based on IMF projections between 1998 and 2000). The importance of raising revenue to levels projected in the PFP is twofold. First, a strong fiscal effort will be met by donor support, which will be needed if the Government wants to increase expenditures to ensure service to the population and functioning of the economy. Second, sustainability of public debt will depend on accessibility to the Paris Club. These reasons are illustrated in the following section by a "low-case" scenario which focuses on the consequences of not achieving revenue targets. (Annex 2 provides full details on the assumptions and procedures underlying this case.) 5.5 What would happen if revenue targets were not met? The assumption underlying this scenario is that the Government takes only minimal effort to correct the ongoing economic and financial deterioration, so that growth in nonmining revenue would match GDP growth while mining revenue collection would benefit from favorable international bauxite prices and the restructuring of the mining sector. The Government would attempt to maintain its expenditure trends, hence forecasting expenditures at levels projected in the PFP. This scenario would create a critical challenge to the Government: how can it finance a 60% increase in the budgetary deficit? The Government would be facing two options: it could proceed with foreign borrowing to finance its deficit or it could reduce expenditures to a level compatible with its revenue. 5.6. Option one. If the Government chooses to finance the deficit with foreign borrowing, the current account deficit would increase rapidly. Thus, as the country's external and fiscal indicators became worse, the Government's credibility would suffer. Borrowing would become difficult for Guinea under the circumstances. However, to the extent that creditors would still be willing to lend to Guinea, borrowing from abroad would be on increasingly nonconcessional terms, creating an unsustainable debt profile in the medium term as the debt would no longer be serviceable. The by-product would be pervasive economic deterioration and a continued accumulation of domestic and external arrears, as well as a direct cut in the rate of GDP growth, notably by further deterring private investment and worsening the credibility of the Government. 5.7 Option two. Under the second option, the Government would reduce expenditures in order to maintain them at a level consistent with revenue; by year 2000, they would represent GNF 629 billion in constant terms, or only slightly more than in 1995, corresponding to a 2.3% growth in real terms (table 5.1). At this rate, which is below the 2.9% population growth rate, these expenditures would be insufficient to sustain basic social services and other public 62 spending necessary to keep the economy functioning, thereby jeopardizing Guinea's developmental prospects in the long term. This "what-if' analysis concludes that if the Government does not remedy its revenue mobilization efforts along the lines suggested in the PFP, Guinea could be headed for serious economic upheaval, as sustained economic growth will require a more tenable fiscal environment. Table 5.1 Macroeconomic projections: Option 1, Option 2 and the PFP 1994 2000 2100 2000 in GNF tlflions Option I Option i PFP.I%iF ftnancing adjumment total re%enue nominal) 3-3 31) 6C3 0" 61.'3 60 859 4C' total revenue (constant, 1994=100) 343.40 . 472.49 472.49 672.73 in% GDP 10.4% 10.6% 10.6% 15.1% total expenditures (nominal) 580.50 1,089.70 803.10 1,089.70 total expenditures (constant, 1994=100) 580.50 853.00 628.66 853.00 in% GDP 17.5% 19.1% 14.1% 19.1% deficit(l) nominal -237.20 -486.10 -199.50 -230.30 deficit (1) constant (constant, 1994=100) -237.10 -380.51 -156.17 -180.28 in % GDP -7.2% -8.5% -3.5% -4.0% external debt (nominal) 2788.00 4716.00 3651.60 3589.82 external debt (constant, 1994=100) 2,788.00 3,684.38 2,852.81 2,810.04 in % GDP 84.1% 82.7% 64.0% 63.0% 1995-2000: Avg. ann. real growth in revenue (real) 5.0% 5.0% 11% Avg. ann. real growth in expenditures (real) 6.9% 2.3% 6.9% Avg. ann. real growth of external debt (real) 5.4% 0.4% 0.1% (a) PFP projections until 1998; IMF projections between 1998 and 2000. (b) excluding grants. 5.8 Strategy for raising revenue. The problems hampering revenue collection are manifold. They include: expanding informal sector acti-vities; pervasive tax evasion; continued fraud in the customs area, particularly pe-troleum products and cigarettes; poor tax administration; and widespread import duty exemptions. The Government has favored a two-pronged approach to enhance revenue mobilization. On the one hand, it has attempted to strengthen tax admin- istration, and on the other, to implement tax reforms. The Customs Department, which is responsible for collecting the bulk of tax revenue, namely import and petroleum taxes, has been strengthened with the help of a surveillance company to combat underinvoicing. In addition, a new division within the Tax Directorate of the Ministry of Finance has been created to collect taxes from enterprises. 5.9 As spelled out in the PFP, the Government has adopted a strategy to modernize the tax system through further strengthening of the tax administration, a rationalization of taxation of goods and services, and a solution to the problem of pervasive import duty exemptions. In that 63 context, the Government intends to rationalize the tax system by adopting a generalized value- added tax (VAT) in April 1996 to replace the multitude of existing indirect taxes. The number of import duty exemptions has also been addressed in the July 1995 revision of the investment code. Rigorous implementation of the revised investment code should lead to increased collection from the 250 (or more) enterprises which had benefited from such exemptions in the context of the previous investment code.' The Importance of Managing Mining Revenue Uncertainty 5.10 Expanding the nonmining revenue base is also vital, as the two other sources of public revenue in Guinea are subject to uncertainty. First, foreign financing of investments may become more scarce as donors are constrained by budgetary considerations at home and competing demands for assistance, notably from Eastern Europe. Second, international bauxite and alumina prices will continue to fluctuate in the medium term, although secular trends of raw materials are pointing downwards. Between 1995 and 2000, prices of bauxite and alumina are projected to recover strongly and Government's revenue from the mining sector should increase.4 5.11 A favorable outlook in the mining sector for the next five years augurs well for the possibility of improving revenue predictability in order to stabilize expenditure management. Indeed, the past five years' experience has demonstrated that insufficient preparation for the mining revenue downturn has the potential to start a downward spiral in fiscal management with severe setbacks on economic and social progress. 5.12 What are the Government's choices? First, the Government will need to strengthen its medium-term planning capacity and improve the management of the sector's pricing and tax mechanisms. Second, in preparation for the next inevitable downturn in bauxite and alumina prices, the Government has an opportunity to constitute reserves with windfalls from the mining sector. Mechanisms for sterilization of excess mining revenue should be set up. Finally, expenditure management will need to remain prudent. In particular, increases in non- discretionary expenditures (typically rigid in the short term) will need to be controlled. C. PRIVATE SECTOR DEVELOPMENT AND DIVERSIFICATION 5.13 Guinea's economic structure remained fragile after 1985 as the economy was -- and still is -- dominated by a few large foreign-owned companies engaged in bauxite, alumina, and Even though these exemptions were eliminated in principle in 1990, they were extended for many enterprises on an ad-hoc basis, and it was only in the second half of 1994 that the Government started to enforce the provisions of the 1990 budget. 2 This trend is already visible in US commitments to the IDA-Il Replenishment, down to US$750 million, from an original commitment of US$1.2 billion. In real terms, the index for bauxite and alumina prices is projected to increase by 20% between 1995 and 2000. See IMF, StaffReport for the 1995 Article IV Consultation, November 1995. The extent of revenue windfall from the mining sector will also be predicated on policies towards restructuring of the mining sector (notably the mining code), attractiveness of new investments, and improved pricing and taxation mechanisms. 64 diamond mining. In 1990, the sector contributed about 22% of GDP and over 90% of exports FOB revenue. In 1994, these shares were still 20% and 80%, respectively. Yet the mining sector is basically a foreign enclave contributing little to employment, and a good portion of mining proceeds is retained overseas. With little private sector expansion, GDP structure remained unchanged over the last decade, with most of the economic impetus coming from the primary sector. 5.14 The set of constraints hampering private sector development primarily reflect the public sector's difficulty in adjusting to a liberal economic system despite the adoption of important measures to incite private sector initiative. Constraints include an unstable institutional environment lacking predictability and accountability. Uneven application of the regulatory and incentive framework, notably the investment and labor codes, have inflated business costs and promoted rent-seeking behaviors. In addition, infrastructure has remained largely inadequate, due mainly to lack of maintenance. How Could the Government Strengthen an Enabling Environment for Private Sector Investment? 5.15 Developing private sector activity and promoting the diversification of the economy is critical to Guinea's development perspectives. In addition, the development of the private sector will result in expanding the tax base, thereby limiting tax pressure on the formal sector of the economy. Thus, the first step towards building the confidence of domestic and foreign investors is to formulate, adopt, and implement a national private sector development strategy which cuts across the public administration and which has been conceived in consultation with representatives of the sector. 5.16 Increased private sector investment in the Guinean economy will largely depend on the degree of business confidence in the macroeconomic program and stability of economic polices and incentives on the one hand, and on the ability to assess risk in estimating future income streams from investment on the other. Creating the environment which will enhance business confidence should be a key ingredient of Government policy. This demands Government consideration of policy measures which will have a direct impact on private investment; but also of broad efficiency public expenditure management reforms to enhance private sector confidence in the Guinean Administration, and therefore improve private sector responses. The following are broad policies that the Government may consider in the context of its examination of specific measures to enhance private sector investment. 5.17 Institutional environment. To develop a climate of business confidence, incentives should be established for an independent and impartial legal system and for improving the overall consistency and predictability of the institutional framework. Improved accountability would abate rent-seeking behavior. 5.18 Accelerating the pace ofjudicial reform, in particular of measures to enforce business law through a judicial process which favors disclosure of laws, processes and outcomes to the public at large. Ensuring the independence of action of the Supreme Court is key to judicial reform. This may be achieved in the context of a review and revision of the Supreme Court's legal statutes and operational provisions in light of the newly-elected Legislative Assembly, and the need to definite the relationship between the Executive, the Legislative, and the Judiciary. 65 5.19 Creating a formal forum for private sector participation in the discussion of policy reforms. In Asia, the use of formal deliberation councils has been successful in facilitating business-Government consultation. The experience of several countries in the West Africa sub- region (Ghana, Senegal, Mali) with the creation of Private Sector Advisory Groups is consistent with this approach. Nevertheless, considerable preparatory work was undertaken. Initial consultations dealt with recommendations which the Government could easily implement, before expanding to a large business roundtable to tackle the more difficult policy issues. In Guinea, similar groundwork may be necessary prior to the establishment of such a mechanism, as the level of mistrust between the business community and Government officials remains high. 5.20 The regulatory and incentive framework reviewing and amending the regulatory and incentive framework in areas of the investment code and labor legislation. As regards the investment code, a review of legislation and of existing conventions within and outside the investment code was completed in May 1995 and a new investment code was adopted in July. Similarly, the Government could examine how labor legislation could be simplified to allow for greater flexibility on the part of employers to downsize the work force in light of product- demand decreases or shifts in the market. 5.21 Credit to the private sector for business expansion, the start of new ventures and pre- exports is also necessary. The ongoing reform of the financial system has not allowed it to move in that direction. Interest rates remain high, reflecting the cost of doing business in Guinea. There is a need for accelerating the pace of financial sector reform to enhance bank profitability and reduce the risks of the banking system. In the short term, reducing the Government's equity share in bank ownership, and paying arrears owed to private contractors will contribute to these reform objectives. A sound financial system will improve the possibilities for future development of a capital market where private business can gain access to business finance. 5.22 Infrastructure and transport. Private sector development requires an adequate network for input and product movement. Although Guinea has invested massively in public works in order to renovate its deteriorating infrastructure, roads are not being maintained properly and are in serious disrepair, adversely affecting transport costs and related economic activity. (This subject has been discussed in more detail in box 2.1.) D. THE ROLE OF THE PUBLIC SECTOR 5.23 As discussed in chapter 1, the liberal option committed to by the Government in 1985 has entailed dramatic reforms towards liberalization and deregulation of the economy. In essence, the liberal option confines the public sector's role primarily to improving access to basic services, maintaining necessary infrastructure, and ensuring adequate monetary, credit, and trade policies and providing a stable and supportive framework for the development of the private sector (this was discussed in the previous section). Consequently, another critical axis of the liberalization paradigm is the withdrawal of the public sector from all activities in which the private sector has a comparative advantage over the public sector, notably in productive and commercial activities.5 5 See The East Asian Miracle (World Bank Policy Research Report, September 1993), p. 84, on the role of government: "...rapid growth in developing countries has been associated with effective but carefully limited government activism. In the 'market-friendly strategy' it articulates, the appropriate role of the 66 5.24 However, in spite of significant progress towards liberalization of the economy, the public sector in Guinea remains involved in productive activities and still provides a sizable part of employment. As seen in chapter 3, productive and commercial activities still hover around 10% of the public investment budget, particularly in the agriculture sector, explaining the high share of expenditures allocated to that sector. In addition, the budgetary and intersectoral analysis, notably the evidence of a recurrent cost crisis (chapters 2 and 3) provides evidence that the public sector is involved in more activities than it can sustain financially, leading to poor- quality service and a low return on investment. As a result, critical levels of expenditures on primary health programs (and drugs) and primary education strongly advocate for the recentering of activities towards its perennial role as provider of basic services. 5.25 Defining and implementing a liberal vision of the public sector's role. The Government has yet to implement a coherent long-term strategy defining its role in the context of objectives in terms of economic growth, income distribution, and poverty alleviation. (This will be discussed further below in section F on intersectoral allocations.) Clear delineation of the public sector's responsibilities and functions would go a long way towards limiting the growth and number of public institutions while clarifying their mandates and focus. In that context, the functions of the civil service could be defined in terms of outcome or service delivery, greatly enhancing management of the wage bill. (This will be discussed below in section G on institutional and budget process reforms.) 1 5.26 Privatization ofpublic enterprises. In this area, some progress has been made, but more efforts are required. In the context of the PFP of July/October 1995, the Government has agreed to a number of measures for completing the divestiture of five enterprises and preparing an action plan for the divestiture of 46 additional enterprises. For gaining private sector confidence, it is important that the plan be implemented in a credible manner. In that context, there is a need to address the constraints of institutions which are charged with management of the privatization process. Delays have been linked to unclear and duplicative lines of responsibility, and to staff lacking the experience in both the value of assets and in negotiating with potential investors. The recent creation of ACGP was intended to address these problems. Unfortunately, as is discussed in annex 12, ACGP's broad responsibilities, and possible tensions with both the Ministry of Finance (also responsible for privatization) and with the responsibilities of line ministries have the potential for complicating the process further. 5.27 Recognizing these problems, the Government may question how best to secure the privatization process and how to address these institutional issues. The answer may lie in the examination of methods and techniques which are working well elsewhere and already showing promise in countries at similar stages of development. A good way to overcome institutional problems is through use of the private sector to handle a large part of the implementation, leaving privatization policy and overall oversight of the process to Government institutions. An additional benefit to be gained from this approach is the promotion of local analytical capacity, government is to ensure adequate investment in people, provide a competitive climate for private enterprise, keep the economy open to international trade, and maintain a stable macroeconomy. Beyond these roles, governments are likely to do more harm than good, unless interventions are 'market-friendly'." 6 The Government's status report on the public enterprise (PE) privatization program (October 1995) claims the following: completion of the privatization of 14 PEs (another two currently in the process); completion of restructuring of three PEs; and a new plan for the privatization of 21 PEs and six PEs previously in the program but whose privatization has become irrelevant. 67 as foreign advisory services and investment banks would rely heavily on local consultants. This has been done with great success in settings as varied as Mexico, Poland, Hungary, Morocco and Russia, and is now being applied in Ghana. E. EXPENDITURE ALLOCATION: HOW CAN THE GOVERNMENT SOLVE THE RECURRENT COST CRISIS? Capital Expenditures are Unsustainable 5.28 Investment levels projected in the PFP and beyond appear unsustainable unless the Government raises additional revenue. Fully accommodating recurrent cost requirements (generated by the projected 7.8% growth of capital expenditures) would entail public expenditure levels that are incompatible with the PFP revenue projections between 1995-1998 and beyond. The recurrent expenditure gap (see figure 5.1) would increase to about GNF 85 billion in year 2000 and further worsen the return on investment. Figure 5.1 Actual capital and recurrent expenditures (1989-1994); PFP and IMF projections for capital and recurrent expenditures (1995-2000); Recurrent expenditure requirements (in constant GNF; 1989-100) 350 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 - capital expenditures -U- recurrent expenditure requirements -0-recurrent actual 5.29 The Government would have essentially two options to resolve the crisis: mobilize revenue beyond levels projected in the PFP; or cut capital expenditures. Either option would need to be accompanied by a reallocation of recurrent expenditures towards O&M. First Option: Mobilize Revenue 5.30 A significant effort in revenue mobilization would certainly help to avoid potentially dangerous cuts in capital expenditures. Can the Government raise revenue beyond PFP projections? On average, revenue would need to increase by a minimum 25% in real terms annually between 1995-2000 to cover the recurrent cost requirements of PFP-projected capital expenditures, bringing the tax revenue/GDP ratio from 10.4% in 1994 to 18.6% in 2000. The fiscal effort would be significantly higher than that projected in the PEP and IMF projections (15.1% revenue/GDP ratio in year 2000). On the basis of Guinea's past and current revenue performance, mobilizing revenue beyond PEP projections appears optimistic. 68 Second Option: Cut Capital Expenditures 5.31 If the Government does not mobilize additional revenue, cuts in capital expenditures seems unavoidable. The dilemma facing the Government is that capital expenditures are already low when seen against the country's many needs. Indiscriminate cuts in core investment will merely hamper Guinea's developmental prospects even further. 5.32 The two expenditure growth scenarios presented in chapters I and 3 - and in section B above -- a 6.9% expenditure growth under the base case (PFP scenario) and a 2.3% growth in expenditures under the low case scenario -- can be used to show the type of annual reduction in capital expenditures that would be necessary to close the recurrent expenditure gap. Under the PFP, capital expenditures would need to decline by a minimum of 5% and a maximum of 10% annually between 1996 and 1998 to close the gap (figure 5.2). (Detailed calculations for these figures are in annex 3.). Figure 5.2 A 5% capital expenditure cut during 1996-2000 with a 6.9% sustainable growth in expenditures (PFP) gap closed 100- - 250 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 - capital expenditures recurrent expenditure requirements -0-- recurrent actual 5.33 If PFP revenue targets were not met, as depicted in the "no tax effort" case, sustainable growth in expenditures would be limited to 2.3% annually. In that case, the gap would take much longer to retrench. Calculations based on an annual reduction in capital expenditures of 10% show that the gap would not likely close by year 2000 (see annex 3). Reallocation of Capital Expenditures 5.34 The Government would have essentially two ways to reduce the relative share of capital expenditures and increase recurrent cost financing -- and a multiple of combinations in between. The first option would entail transferring domestic resources from BND to the current budget, or simply transforming the BND into a current budget. This option. would in fact also address the problem mentioned earlier: the inadequacy of BND resources to cover Government contributions to both FINEX projects and its own projects. However, this option would involve abandoning all 100% domestically-financed projects (about 10% of the PIP). For various reasons, the Government may find this perspective unacceptable. A joint reduction in donor and Government-financed projects may be more feasible. In that case, a rigorous evaluation of the investment budget -- identifying marginal and non-priority projects -- would be critical to decisions about which FINEX and BND projects should be pursued. This exercise will also 69 demand a clear understanding of and agreement on the development priorities of the Government and donors, and a proven commitment by the Government to honor its financial commitments to FINEX participation. The other option would entail reorienting FINEX projects toward increasing the share of recurrent costs financed by donors even further. This option, which is already -- and increasingly -- being applied in some sectors, would raise two issues: the sustainability of projects and donor financing of investment (as discussed in chapter 1). 5.35 These two options show clearly that revenue enhancement is the only "positive option". In fact, only a strong revenue effort will secure sustainable levels of investment, given the uncertainty of donor financing of investment (as discussed in chapter 1). Reallocation of Recurrent Expenditures Towards O&M 5.36 What would happen if current expenditure trends were to continue while the PFP revenue targets were missed? Increases in wages and pensions and their comparative weight in the budget could lead to a full-blown crisis in the medium term. If expenditure growth were limited to a real 2.3% annually, and if the wage bill and pensions grew along the lines projected by the PFP, O&M expenditures would fall precipitously: from 32% of the current budget (net of debt service due) in 1994 to 9.2% by year 2000 (figure 5.3), clearly an undesirable and unsustainable position. Figure 53 Projected growth of wages and pensions with a 2.3% growth in sustainable expenditures (as a percentage of the current budget) 90% 80% salaries 70% - 60% 50% 30% . . 2% pen:ions, n d - 10%! 10% Sca eah ~-- 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 5.37 The implication is that a reallocation of expenditures from the wage bill to O&M will be critical to resolving the O&M gap, regardless of revenue performance. Not only should revenue shortfalls be offset exclusively by wage bill cuts rather than by cuts in O&M expenditures, but even if the Government meets its revenue objectives, current trends in wages and salaries and O&M should also be reversed. Furthermore, reducing pressure on capital expenditures would require more effective allocations of recurrent expenditures -- transfers from salaries to O&M -- as a priority. Reducing the wage bill presents a difficult political choice, which the Government may or may not be in a position to take in the short term. However, as a medium-term term objective, a wage bill cut could be embodied in a civil service reform in which an improved 70 incentive structure should alleviate employment losses. (This will be discussed below in section G on institutional reforms.) 5.38 However, without going that far, the Government could already generate savings by rationalizing the number of agencies and the number of staff in some ministries -- for example, the Agriculture Ministry, where, on the basis of what is known of Ministry activities, 20% to 25% of staff is reportedly redundant. These savings, which represent a possible gain of GNF 3.0 to 3.5 billion, could be allocated to the Agriculture Ministry's O&M expenditures (about 80% of these are currently financed by donors), thus freeing up donor resources for priority sectors. F. CRITERIA FOR INTERSECTORAL ALLOCATIONS: WHAT ARE THE OPTIONS? 5.39 The sectoral and intersectoral analysis indicates that current levels of resources are insufficient to meet the basic needs of the population, maintain the existing infrastructure, and promote a more efficient civil service. In particular, basic services must receive considerably more resources if Guinea wants to reduce poverty and malnutrition in a significant and sustained manner. Hence, raising additional revenue is a sine qua non for adequate public services provision and development. In addition, even if revenue mobilization is increased to the levels projected in the PFP, inter- and intrasectoral choices will have to be reoriented towards efficiency-based allocations of public expenditures, including adequate endowments to O&M. Government and donors will need to enhance coordination and understanding on and commitment to overall development objectives and key priorities. 5.40 As discussed above, public resources for sectoral allocations would go to where market failure and a strong lag in benefits prevents private sector activity (in, for example, agricultural extension) and to the social sectors, essentially health, education, and other basic needs for the population. However, it should be noted that increasing allocations to social sectors is likely to add to the R coefficient as expenditures in social sectors typically generate more recurrent costs than other sectors (see annex 3). If the Government were to finance its strategies in health and primary education, the R coefficient could grow up to 17% (annex 3). As a result, the Government will need to examine carefully the recurrent cost implications of shifts in investment expenditures towards social sectors. This conclusion also has implications on cost recovery policies. If taken at face value, an increase in recurrent costs induced by higher investments in social sectors would seem to advocate for further cost recovery schemes. However, the existing constraints to cost recovery are not well known and should be carefully examined, notably in the context of the private sector's ability, and particularly the poor's, to finance basic services. Examples in some SSA countries (e.g., Malawi) shed a disturbing light on the benefits of cost recovery financing, notably in education. 5.41 How would the Government finance these allocations? The following suggestions show how the Government could generate specific savings. The Government should consider the applicability of these suggestions. 5.42 Gains on transfers appear limited (except for scholarships to higher education), given that direct subsidies to public enterprises, although unjustified, amount only to 1% of GDP (GNF 2.0 billion to SOGETRAG and Air Guin6e) and are due to disappear in 1996. However, evidence shows that subsidies still exist -- GNF 20 billion were reportedly spent on public enterprise subsidies in 1995. In addition, some existing indirect subsidies represent foregone 71 revenue. For example, the tax on airplanes flying over the Guinean territory is paid to Air Guinee, rather than to the Treasury. Completing the privatization/liquidation of the public enterprises should provide additional revenue or allow resources to be reallocated; for example, the privatization of SOTELGUI in the telecom sector would free up about GNF 5 billion in the investment budget. 5.43 The Government should also reallocate resources within sectors as well. Savings in higher education of GNF 1.5 to 2.0 billion annually should be targeted at primary education. Similarly, in the health sector, administrative expenditures could be reduced progressively through rationalization. Tertiary health care could generate savings progressively in phasing out of the hospital subsidy of GNF 2 billion annually, and cost recovery in regional hospitals and in the two Conakry hospitals should be implemented to cover drug needs and 30% of other recurrent expenditures. In the road sector, the Government should also reallocate expenditures from new investments to O&M activities in the existing network, primarily by improving donor coordination. 5.44 The Government could eliminate nonviable or marginal projects as another source of savings. Allocations should be targeted at programs or projects that offer the highest rates of return or exhibit other measures of efficiency such as cost effectiveness. Given the tight liquidity of the Treasury, a basic rule emerging from the analysis in chapter 2 is that it is less expensive and more productive to maximize the return on investment by securing all the resources necessary to its sustainability, even if, as is inevitably the case, securing one investment involves delays or suspensions in others. As demonstrated amply in the road sector, it is less expensive in the long run to have one operational project sustained by an adequate level of recurrent costs, rather than having ten non-operational projects due to insufficient funding. 5.45 Finally, the Government could generate savings by withdrawing from productive activity. In agriculture, withdrawal from productive activities and non-core expenditure programs, could free up 10% to 12% of investments in rural resources, or approximately GNF 10 billion annually. The public sector should also pull progressively out of the mining sector, which represents about GNF 7 billion in investment annually. And, finally, the Government should review its own projects and determine core priorities. For example, projects in other social sectors (primarily sports) could be delayed or redirected towards higher priority expenditures (about GNF I billion annually). G. POLICY AND STRATEGIC OPTIONS FOR INSTITUTIONAL REFORMS AND THE BUDGETARY PROCESS 5.46 Improvement in the institutional environment and the budgetary process conditions any meaningful progress on the allocation front. A principal problem seems to be that the budget is insufficiently transparent while institutions are not thoroughly accountable, nor are their incentive structures inadequate. Consequently, systems of checks and balances, accountability, and sanctions governing the use of expenditures appear ineffective. In addition, mechanisms to ensure that the Government's policy achieves its intended results on the ground are weak. The following section presents broad policy choices for a reform agenda towards improving expenditure management and institutional coordination. 72 Institutional Reform 5.47 A comprehensive review of institutional problems is essential as the basis for institutional reform. This should entail the systematic comparison of broad institutional mandates, internal structures and attributions, and staffing and skill requirements of all institutions responsible for the management of public resources. This would involve ACGP, the Ministry of Finance, and the Ministry of Plan and Cooperation as central players. (ACGP is discussed further in annex 12.) The ultimate objective of this exercise would be to clarify broad institutional mandates, identify overlapping and duplicative structures, and identify redundant skills or skill gaps as the basis for the design and implementation of organizational structures which respond to redefined institutional mandates. Checks and Balances 5.48 Institutional reform should also cover the emerging system of checks and balances which involves the Ministry of Financial and Economic Control, and the Chambre de Comptes of the Supreme Court. The process should aim at clarifying their roles: the first institution controlling the validity of expenditure in the course of the budget year; the second institution exercising expenditure control ex-post. As part of the Supreme Court, the relationship of the Chambre de Comptes with the Executive and the Legislative Assembly should be clarified to ensure independence of action. Both institutions should improve their organizational structures and technical capacity and mobilize the necessary resources to carry out independent audits of public institutions, as necessary. 5.49 It is crucial to make the budgetary process more transparent and put in place checks and balances to deter possible misappropriation. To this end, it is necessary to develop a well- functioning, transparent financial management system underlying the budget process that records and makes transparent actual disaggregated expenditures (disaggregated by ministry and by sector) within a reasonable time frame. A basis for such publication could be the Tableau de Bord des Finances Publiques prepared every month by DNB and DNT. Such systems have been put in place recently in countries such as Namibia and Uganda, and Ghana and Malawi are in the process of instituting variants of this process. A critical associated measure would be to publish and make public actual expenditures, Government's accounts and audits within a reasonable time frame. Budget Planning 5.50 Improving the budget's efficiency as a management tool would require strengthening planning and budgeting capacity. A first step would be the preparation of a comprehensive medium-term framework involving costing of the Government's current and alternative policies and testing affordability of these policies against a three-year macroeconomic framework. Decisions by the Government on sector priorities and financing of sector policies would require a clear development policy articulating the role of the public sector and outcomes of public policies. The second step would involve planning policies over a three-year period in the PIP. Annual budget preparation would then involve the breakdown of the medium-term plan according to priorities. The budgeting exercise would gain significance with the imposition of a binding budget ceiling by sector, reflecting the priority given to each sector under the macroeconomic constraint. 73 5.51 An option open to Government to achieve the above-mentioned objectives may involve a redefinition of responsibilities as regards budget preparation. The budgets are viewed as separate processes which bear no relationship to one another. Missing is the capacity to estimate the operation and maintenance (O&M) requirements of capital investments. Given budget planning deficiencies, it may be best to give full responsibility for the entire budget preparation process to the Ministry of Finance. In this context, the Ministry of Finance would be given full responsibility for preparation of a three-year budget framework. 5.52 The Ministry of Plan and Cooperation would have an input in the process. It would, in close cooperation with line ministries, ensure the coherence of sectoral strategies with macroeconomic objectives, and would undertake a full economic and financial analysis of projects (project appraisal) as a prerequisite for inclusion in the investment budget. The estimated future operations and maintenance expenditures associated with a capital investment would be part of the project's appraisal. Guidelines for the estimation of operational costs associated with projects would be issued by the Ministry of Plan and Cooperation to line ministries. In this context, the technical capacity of the Ministry of Plan would need to be considerably enhanced. The enhanced analytical skills of Ministry of Plan staff could be used to improve the analytical capacity of line ministries. 5.53 The Ministry of Finance would ensure that the number of projects to be included in the capital budget do not exceed available resources. Project requirements for the budget year would be provided with a budgetary commitment and the two subsequent years would be counted as indicative allocations. 5.54 Improving the efficiency of the budget planning process would also require the Government to give consideration to improved use and provision of information and to coordination among Government agencies, and with donors. The options open to the Government as regards donor coordination are detailed in chapter 2 of this report. Civil Service Reform 5.55 The budget planning process necessarily requires that attention be given to wage bill related expenditures in the current budget, which account for around 50% of total current expenditure -- if unchecked, they risk becoming unsustainable in the near future. 5.56 It appears to be essential to do this, not only because a reduction in the overall proportion of the wage bill in total current expenditure would facilitate a reallocation of resources to fund priority services, but also because retrenchment of redundant employees would enable the Government to improve the overall skills of the civil service by hiring younger, better-trained personnel to fill well-defined skill gaps. The decline in quality of government services has been exacerbated by the declining productivity of an aging civil service. The Government needs to examine the advantages of a reform that simultaneously reduces the number of institutions and individuals in the civil service, based on the role of the public sector, the outcome of public spending, and affordability of expenditures, and defines the responsibilities of both institutions and individuals in order to improve accountability, while institutionalizing incentives to increase productivity. Such a reform would need to include changes in the civil service code to allow for flexible organizational structures that respond to the institutional mandate. 74 5.57 A first step would require a broad assessment of institutions in the context of the redefinition of the role of the State in Guinea's economy, as the basis for decisions on their continued operation. A second step in such an endeavor would be to finalize ongoing efforts to determine the current size, distribution, and status of the civil service. The exercise would need to include a complete survey of civil servants approaching retirement age. This would enable the Government to plan future pension funding requirements. Restructuring Pension Programs 5.58 Regarding the public pension system, the Government needs to examine existing legislation, rules, and implementation of pension entitlement. A major concern is that Government employees do not contribute to their pensions. A second concern is that different rules appear to apply to the entitlement of military and civil service personnel. There do not appear to be general provisions for beneficiaries to take their entire entitlement as a lump sum. Allowing beneficiaries to take a lump sum would save the Government administrative costs, and would, over the longer term, reduce the overall fiscal burden. As regards the CNSS, a financial assessment is an urgent priority and should be carried out to determine whether worker and employers contributions to the CNSS system are invested appropriately to generate a level of returns which is consistent with the liabilities of the system. The financial assessment should offer a set of recommendations which could be considered by Government as the basis for restructuring. Budget Implementation 5.59 A redefinition of responsibilities as regards the budget planning process would significantly reduce the number of steps required for approval of budget expenditure, but the process could be further simplified if overlapping and duplicative structures are removed in the context of institutional reforms. A second major impediment to efficient budget implementation is the current information-gathering and dissemination system. An integrated budget management system would provide each player with the amount of detail required for each particular line of work, while also providing a mechanism for recording revenue and expenditure transactions. The system would also serve as a medium for producing and disseminating information, facilitating expenditure control and enhancing accountability. Conclusion 5.60 A favorable outlook predicted for the bauxite sector offers a renewed opportunity to consolidate previous gains and deepen reforms. As demonstrated by experience in the past five years, insufficient preparation for a downturn in revenue can start a downward spiral in fiscal management, with severe setbacks to economic and social progress. Provided the Government succeeds in enhancing the efficiency of its resource management and developing a predictable and consistent framework for the development of the private sector, Guinea could capitalize on past efforts and reap the benefits of its enormous potential thereby lifting its population from poverty. 75 BIBLIOGRAPHY Bindlish, Vishva; Evenson, Robert; and Gbetibouo, Mathurin. Evaluation of T& V-Based Extension in Burkina Faso, Report No. WTP226, November 1993. Food and Agriculture Organization. Guined : Deuxibme Projet de Services Agricoles, Rapport Preliminaire de Priparation, Rapport No 63/93, May 1994. Government of Guinea. Lettre de Politique de Diveloppement Agricole, December 1991. Government of Guinea, MAEF/BCPA, Budget d'Investissement Public du Secteur Agricole, December 1995. Government of Guinea, MARA, Activitis dans le secteurforestier, 12re idition, December 1973. Government of Guinea. 1994 Integrated Household Survey, April 1995. Government of Guinea. Plan d'Action Forestier de Guinje, February 1990. Government of Guinea, International Monetary Fund, and World Bank. Guinea: Policy Framework Paper 1995-1998, July 1995. Heller, Peter. "Public Investment in LDCs with Recurrent Cost Constraint: the Kenyan Case", The Quarterly Journal ofEconomics, May 1974. . "The Underfinancing of Recurrent Development Costs", Finance and Development, March 1979. International Monetary Fund. Guinea: Staff Report for the 1995 Article IV Consultation, November 1995. Lele, Uma; Van de Walle, Nicholas; and Gbetibouo, Mathruin. Cotton in Africa: An Analysis ofPerformance, World Bank MADIA Discussion Paper No. 8297, November 1989. van Blarcom, Bonni; Knudsen, Odin; and Nash, John. The Reform ofPublic Expenditures for Agriculture, Report No. WDP216, October 1993. World Bank. Better Health in Africa: Experience and Lessons Learned, Report No. 13488, September 1994. World Bank. The East Asian Miracle: Economic Growth and Public Policy, Report No. 12351, September 1993. 76 World Bank. FY93 Annual Review of Portfolio Performance, December 1993. World Bank: Guinea: Civil Service Reform Management Review, September 1994. World Bank. Guinea: Country Economic Memorandum, October 1995. World Bank. Mali: Public Expenditure Review, May 1994. World Bank. World Development Report 1993: Investing in Health, Report No. 12183, June 1993. 77 ANNEXES Annex I Table 1.1 Guinea: GDP in constant 1989 prices, 1988 - 1994 (In billions of Guinean francs) Sources and Uses Gross domestic product 1383.5 1438.9 1501.3 1537.3 1582.8 1657.3 1723.6 Imports of goods and nfs 443.8 428.6 470.6 472.5 460.8 488.2 438.4 Total resources 1827.3 1867.5 1971.9 2009.8 2043.6 2145.5 2162.0 Final consumption 1163.6 1172.8- 1237.0 1278.0 1346.4 1375.7 1458.8 Central government 145.1 168.2 186.5 190.5 196.9 195.1 189.1 Other 1018.5 1044.6 1050.5 1087.5 1149.6 1180.6 1269.7 Gross capital formation 241.5 247.3 267.0 249.1 247.4 267.1 229.6 Central government 115.9 119.9 140.7 118.6 113.7 117.0 104.9 Mining companies 26.0 16.4 25.4 41.5 46.6 54.1 21.2 Other 99.6 111.0 100.9 89.0 87.1 96.0 103.5 Exports of goods and nfs 422.2 447.4 467.9 482.6 449.8 502.8 473.5 Total uses 1827.3 1867.5 1971.9 2009.7 2043.6 2145.6 2161.9 Composition Primary sector 322.0 339.1 350.8 363.3 377.0 395.7 415.3 Agriculture 230.8 241.4 248.7 256.4 264.8 278.1 292.0 Livestock 40.8 47.1 49.0 51.0 53.3 55.7 58.2 Fisheries and forestry 50.4 50.6 53.1 55.9 58.9 61.9 65.1 Secondary sector 462.3 .477.7 491.6 496.5 499.5 515.0 531.4 Mining 308.7 322.0 322.6 320.7 315.9 322.2 328.6 Manufacturing 62.1 64.0 67.2 69.9 71.8 75.4 79.6 Water, electricity 3.1 3.3 3.3 3.4 3.4 3.6 3.7 Construction 88.4 88.4 98.5 102.5 108.4 113.8 119.5 Tertiary sector 572.6 593.5 626.1 642.9 669.5 707.5 736.0 Trade 331.9 347.1 366.9 381.6 399.2 423.1 448.5 Transport 71.6 72.2 75.6 78.4 81.7 85.0 88.4 Administration 91.0 96.1 101.6 96.8 97.8 92.9 91.3 Other 78.1 78.1 82.0 86.1 90.8 106.5 107.8 Indirect taxes net of subsidies 26.8 28.6 32.8 34.7 36.8 39.0 40.9 (As a percentage of GDP) Sources and Uses Gross domestic product 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Imports of goods and nfs 32.1 29.8 31.3 30.7 29.1 29.5 25.4 Total resources 132.1 129.8 131.3 130.7 129.1 129.5 125.4 Final consumption 84.1 81.5 82.4 83.1 85.1 83.0 84.6 Central government 10.5 11.7 12.4 12.4 12.4 11.8 11.0 Other 73.6 72.6 70.0 70.7 72.6 71.2 73.7 Gross capital formation 17.5 17.2 17.8 16.2 15.6 16.1 13.3 Central government 8.4 8.3 9.4 7.7 7.2 7.1 6.1 Mining companies 1.9 1.1 1.7 2.7 2.9 3.3 1.2 Other 7.2 7.7 6.7 5.8 5.5 5.8 6.0 81 Annex 1 Table 1.1 Guinea: GDP in constant 1989 prices, 1988 - 1994 7.19z4~8z8 9zt9ol19 992 1993 -199C1 Exports of goods and nfs 30.5 31.1 31.2 31.4 28.4 30.3 27.5 Total uses 132.1 129.8 131.3 130.7 129.1 129.5 125.4 Composition Primary sector 23.3 23.6 23.4 23.6 23.8 23.9 24.0 Agriculture 16.7 16.8 16.6 16.7 16.7 16.8 16.9 Livestock 2.9 3.3 3.3 3.3 3.4 3.4 3.4 Fisheries and forestry 3.6 3.5 3.5 3.6 3.7 3.7 3.8 Secondary sector 33.4 33.2 32.7 32.3 31.6 31.1 30.8 Mining 22.3 22.4 21.5 20.9 20.0 19.4 19.1 Manufacturing 4.5 4.4 4.5 4.5 4.5 4.5 4.6 Water, electricity 0.2 0.2 0.2 0.2 0.2 0.2 0.2 Construction 6.4 6.1 6.6 6.7 6.8 6.9 6.9 Tertiary sector 41.4 41.2 41.7 41.8 42.3 42.7 42.7 Trade 24.0 24.1 24.4 24.8 25.2 25.5 26.0 Transport 5.2 5.0 5.0 5.1 5.2 5.1 5.1 Administration 6.6 6.7 6.8 6.3 6.2 5.6 5.3 Other 5.6 5.4 5.5 5.6 5.7 6.4 6.3 Indirect taxes net of subsidies 1.9 2.0 2.2 2.3 2.3 2.4 2.4 Sources: Estimates of Guinean authonties, World Bank and IMF staffs. 82 Annex 1 Table 1.2 Guinea: GDP in Current Market Prices, 1988-1994 41988, 1-991gf9_ 192 4j~P~A (in billions of Guinean francs) Sources and Uses Gross domestic product 1131.1 1438.9 1860.5 2248.0 2682.7 3036.3 3315.6 Imports of goods and nfs 351.5 428.6 570.2 659.4 791.9 838.8 808.5 Total resources 1482.6 1867.5 2430.7 2907.4 3474.6 3875.1 4124.1 Final consumption 985.4 1172.8 1530.5 1898.1 2382.8 2703.3 3023.0 Central government 126.5 168.2 222.3 257.9 287.9 292.4 289.2 Other 858.9 1004.6 1308.2 1640.2 2094.9 2410.9 2733.8 Gross capital formation 193.2 247.3 325.9 370.1 442.7 490.6 452.4 Central government 92.6 119.9 171.3 176.2 202.1 214.6 201.6 Mining companies 22.6 23.4 40.3 67.4 85.7 99.9 51.9 Other 78.0 104.0 114.3 126.5 154.9 176.1 198.9 Exports of goods and nfs 304.0 447.4 574.2 639.2 649.1 681.1 648.6 Total uses 1482.6 1867.5 2430.6 2907.4 3474.6 3875.0 4124.0 (As a percentage of GDP) Sources and Uses Gross domestic product 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Imports of goods and nfs 31.1 29.8 30.6 29.3 29.5 27.6 24.4 Total resources 131.1 129.8 130.6 129.3 129.5 127.6 124.4 Final consumption 87.1 81.5 82.3 84.4 88.8 89.0 91.2 Central government 11.2 11.7 11.9 11.5 10.7 9.6 8.7 Other 75.9 69.8 70.3 73.0 78.1 79.4 82.5 Gross capital formation 17.1 17.2 17.5 16.5 16.5 16.2 13.6 Central government 8.2 8.3 9.2 7.8 7.5 7.1 6.1 Mining companies 2.0 1.6 2.2 3.0 3.2 3.3 1.6 Other 6.9 7.2 6.1 5.6 5.8 5.8 6.0 Exports of goods and nfs 26.9 31.1 30.9 28.4 24.2 22.4 19.6 Total uses 131.1 129.8 130.6 1293 129.5 127.6 124.4 Sources: Estimates of Guinean authorities, World Bank and IMF staffs. 83 Annex I Table 1.3 Guinea: Selected Indicators, 1988-1994 - 988 1989' 990u1991 1992 993 1994 (Annual percentage changes) Real GDP growth 6.3 4.0 4.3 2.4 3.0 4.7 4.0 Real per capita GDP Growth 3.3 1.0 1.4 -0.5 0.0 1.7 1.1 Consumer price inflation Average 27.4 28.3 19.4 19.6 16.6 7.1 4.1 End of period 26.3 26.0 27.1 12.2 16.1 5.0 2.2 (Index 1986=100) Terms of trade 79.8 87.8 88:9 83.3 73.7 69.2 66.1 Percentage change -10.8 10.0 1.3 -6.3 -11.5 -6.1 -4.5 Bauxite price 72.1 84.6 100.5 95.1 84.6 73.9 67.6 Alumina price 121.6 162.7 204.7 189.0 147.7 129.2 125.7 Export volume 113.6 121.7 123.0 126.2 116.7 132.2 125.6 Import Volume 121.7 116.9 132.9 133.8 130.9 136.9 121.4 Exchange Rate (GNF=US$1) 1/ 474.4 591.6 660.2 753.9 902.0 955.5 976.6 Sources: Data provided by the Guinean authorities; and staff estimates. 11 Period average. 84 Annex 1 Table 1.4 Guinea: Savings and Investment Balances, 1988-1994 11 1988 198~ i9O~2q99' .9Wg .1993s 2'9941~ (In percent of GDP) Government Gross savings 21 -2.1 -0.7 0.4 -0.4 -0.5 -0.4 -1.1 Investment 8.2 8.3 9.2 7.8 7.5 7.1 6.1 Balance -10.3 -9.0 -8.8 -8.2 -8.0 -7.5 -7.2 Private sector 3/ Gross Savings 2/ 6.5 9.0 8.0 8.6 4.7 5.9 5.4 Investment 8.9 8.9 8.3 8.6 9.0 9.1 7.5 Balance -24 0.1 -0.3 0.0 -4.3 -3.2 -2.1 Mining companies Gross savings 10.6 13.1 12.1 11.3 8.3 7.6 5.5 Investment 1.8 1.1 1.9 2.7 3.2 3.3 1.5 Balance 8.8 12.0 10.2 8.6 5.1 4.3 3.9 Rest of Private sector 31 Gross savings -4.2 -4.1 -3.9 -2.8 -3.6 -1.7 0.0 Investment 7.1 7.8 6.4 5.9 5.8 5.8 6.0 Balance -11.3 -11.9 -10.3 -8.7 -9.4 -7.5 -6.0 Total economy Gross savings 2/ 4.4 8.3 8.4 8.2 4.2 5.5 4.4 Investment 17.1 17.2 17.5 16.5 16.5 16.2 13.6 Balance -12.7 -8.9 -9.1 -8.3 -12.3 -10.7 -9.3 (external current account balance) Sources: Estimates of Guinean authorities, World Bank and IMF staffs. 1/ May not add up because of rounding. 2/ Domestically generated savings, excluding foreign grants. 3/ Including public enterprises. 85 Annex 1 Table 1.5 Guinea: Consumer Price Index, 1988-1994 (Period average data) . 1~<w.f988, fgg 19 ggggg A 9 21993 27@ (1987=100) Aggregate Index 127.4 Lig U g9.. 233,5 272.2 291.6 3. Food 133.6 168.6 198.8 231.1 267.9 292.9 306.7 Nonfood 122.4 159.2 192.1 235.4 275.7 290.6 300.9 Clothing 121.7 129.2 150.8 168.0 179.1 196.9 200.0 Health 120.0 148.4 195.8 235.9 258.6 267.1 304.6 Housing 108.2 145.1 173.5 195.8 235.8 247.6 249.0 Electricity and water 117.7 135.3 160.9 207.6 233.1 233.5 239.2 Transportation 140.2 187.0 222.9 310.8 393.4 424.1 427.2 Leisure and culture 115.1 163.3 190.0 217.8 255.1 272.6 285.7 (Annual percent change) Aggregate Index 27.4 2U3 1.4 .i 16.6 ZJ. 4.21 Food 33.6 26.2 17.9 16.2 15.9 9.3 4.7 Nonfood 22.4 30.1 20.6 22.5 17.1 5.4 3.6 Clothing 21.7 6.1 16.8 11.4 6.6 9.9 1.6 Health 20.0 23.7 32.0 20.5 9.6 3.3 14.1 Housing 8.2 34.1 19.6 12.9 20.4 5.0 0.6 Electricity and water 17.7 14.9 18.9 29.0 12.3 0.2 2.5 Transportation 40.2 33.4 19.2 39.4 26.6 7.8 0.7 Leisure and culture 15.1 41.9 16.4 14.6 17.1 6.9 4.8 Source: Data provided by the Guinean authorities. 86 Annex I Table 1.6 Guinea: Central Government Revenue, 1988-1994 (In billions of Guinean francs) Total revenue 1J2 222. 293. 33D,Q 361. 351.4 3433 Mining sector 111.3 150.7 197.6 190.0 162.7 137.5 98.5 Nonmining sector 51.5 71.7 96.3 140.0 198.4 213.9 244.8 Income and profit taxes 3.0 5.0 6.3 13.9 24.6 24.3 30.1 Personal 1.8 1.3 3.6 9.0 13.1 16.5 Corporate 0.8 1.8 1.8 2.6 2.6 3.8 31.2 Payroll taxes 0.4 0.5 0.9 2.3 2.0 2.4 Other 0.0 1.4 0.0 0.0 6.9 1.6 -1.1 Taxes on domestic trade 23.4 38.0 41.6 68.3 101.3 114.5 136.5 Turnover taxes 11 9.5 14.5 19.5 27.4 33.5 36.6 38.5 Excise surcharge 3.4 3.9 4.3 4.6 5.3 6.9 10.2 Petroleum excise tax 10.0 18.7 17.1 35.3 48.9 52.6 65.1 Other 0.5 0.9 0.7 1.0 13.6 18.4 22.7 Taxes on international trade 11.1 14.2 25.7 38.2 46.3 51.2 53.8 Taxes on imports 9.4 12.9 23.9 36.6 44.0 46.0 50.6 Taxes on exports 0.2 0.2 0.1 0.1 0.1 0.1 0.1 Other 1.5 1.0 1.7 1.5 2.3 5.1 3.1 Nontax revenue and miscellaneous 14.0 14.5 22.6 19.5 26.2 24.0 24.4 (In percent.of GDP) Memorandum items: Total revenue 14.4 15.5 1U3 147 135 11.6 1.QA Mining sector revenue 9.8 10.5 10.6 8.5 6.1 4.5 3.0 Nonmining sector revenue 4.6 5.0 5.2 6.2 7.4 7.1 7.4 Income and profit taxes 0.3 0.4 0.3 0.6 0.9 0.8 0.9 Taxes on domestic trade 2.1 2.6 2.2 3.0 3.8 3.8 4.1 Taxes on international trade 1.0 1.0 1.4 1.7 1.7 1.7 1.6 Nontax revenue and miscellaneous 1.2 1.0 1.2 0.9 1.0 0.8 0.7 Source: Data provided by the Guinean authorities. 1 /Including turnover taxes on imports. 87 Annex I Table 1.7 Guinea: Central Government Expenditure, 1988-1994 f ~a l98 ow 9 1 - (In billions of Guinean francs) Total expenditure 276A 347.4 458.4 515.2 571.2 59. 580.6 Current expenditure 152. 18Z& 230.0 280.3 302.7 293.5 311.8 Wages and salaries 45.5 60.7 78.5 102.9 132.1 140.3 145.5 Base salaries 44.6 57.2 74.7 98.8 127.6 134.2 148.0 Other remuneration 0.9 3.5 3.8 4.1 4.5 6.1 -2.5 Other goods and services 50.1 67.5 86.7 96.2 89.4 80.6 76.5 SBK-financed expenditures 12.8 11.7 20.0 24.4' 12.1 1.0 8.5 Other 37.3 55.8 66.7 71.8 77.3 79.6 68.0 Subsidies and transfers 21.5 24.4 17.2 28.0 32.6 31.1 37.3 Public entities 2.2 2.1 3.0 4.6 0.8 0.9 0.6 Subsidies to enterprise and guarantees 11.3 4.4 3.9 7.2 5.0 3.6 3.0 Social and Cultural funds 0.6 1.7 2.3 2.8 3.2 3.9 3.8 Other transfers 7.4 16.2 8.0 13.4 23.6 22.7 29.9 Interest payments 35.8 35.0 47.6 53.2 48.6 41.5 52.5 Capital expenditures 123.5 159. 228.4 234. 2.5.7 2861 268. Domestically financed 21.1 27.5 41.0 39.9 31.7 45.6 38.8 Foreign financed 1024 132.3 187.4 19.. 234.0 24G 230-0 Loans 72.1 86.8 117.4 122.4 134.0 143.9 128.9 Grants 30.3 45.5 70.0 72.6 100.0 96.6 101.1 Net Lending 0.0 0.0 0.0 0.0 2.8 0.0 0.0 Source: Data provided by the Guinean authorities. 88 Annex I Table 1.8 Guinea: Central Government Financial Operations, 1988-1994 ~ ~993~ t 199 (In billions of Guinean francs) Revenue and grants 199.6 274.9 362.1 412.5 478.4 464.1 460.8 Revenue 162.8 222.4 293.9 330.0 361.1 351.4 343.3 Mining sector revenue 111.3 150.7 197.6 190.0 162.7 137.5 98.5 Nonmining sector revenue 51.5 71.7 96.3 140.0 198.4 213.9 244.8 Income and profit taxes 3.0 5.0 6.3 13.9 24.6 24.3 30.1 Taxes on domestic trade 23.4 38.0 41.6 68.3 101.3 114.5 136.5 olw: petroleum taxes 10.0 18.7 17.1 35.3 48.9 52.6 65.1 International trade 11.1 14.2 25.7 38.2 46.3 51.2 53.8 Nontax revenue 14.0 14.5 22.6 19.5 26.2 24.0 24.4 Grants 36.8 52.5 68.3 82.5 117.3 112.7 117.5 Expenditure and net lending 276.4 347.4 458.4 515.2 571.2 579.6 580.5 Current expenditure 152.9 187.6 230.0 280.3 302.7 293.5 311.7 Wages and salaries 45.5 60.7 78.5 102.9 132.1 140.3 145.5 Other goods and services 50.1 67.5 86.7 96.2 89.4 80.6 76.5 Subsidies and transfers 21.5 24.4 17.2 28.0 32.3 31.1 37.3 External interest payments due 35.8 35.0 47.6 53.2 48.9 40.3 49.7 Domestic interest payments due 0.0 0.0 0.0 0.0 0.0 1.2 2.7 Capital expenditures 123.5 159.8 228.4 234.9 265.7 286.1 268.8 Domestically financed 21.1 27.5 41.0 39.9 31.7 45.6 38.8 Externally financed 102.4 132.3 187.4 195.0 234.0 240.5 230.0 Net tending 0.0 0.0 0.0 0.0 2.8 0.0 0.0 Balance, commitment basis Including grants -76.7 -72.6 -96.3 -102.7 -93.1 -115.3 -119.8 Excluding grants -113.5 -125.1 -164.6 -185.3 -210.4 -228.0 -237.2 Changes in payment arrears 7.3 -1.6 37.7 12.1 -97.4 0.3 19.2 Domestic 1.0 4.9 8.2 -0.4 -20.0 -4.2 7.9 External 6.4 -6.4 29.5 12.6 -77.4 4.5 11.3 Balance, cash basis -69.4 -74.2 -58.6 -90.6 -190.5 -115.0 -100.6 Financing 69.4 74.2 58.6 90.6 190.5 115.0 100.6 Net external financing 58.5 83.7 75.7 74.5 199.4 149.7 86.2 Drawings 94.3 124.0 119.9 147.3 160.8 194.0 136.0 Project financing 72.1 86.8 117.4 122.4 134.0 143.9 128.5 Other 22.2 37.2 2.5 24.9 26.8 50.1 7.5 Amortization due -68.7 -73.7 -82.6 -117.3 -115.7 -93.6 -139.5 Short-term credits, net 0.0 0.0 7.6 -0.9 -1.1 -4.3 -5.7 Arrears (reduction-) 21.9 -11.4 30.8 45.4 -219.4 70.9 95.4 Debt relief 11.0 44.8 0.0 0.0 346.8 56.3 0.0 Deferred payments 0.0 0.0 0.0 0.0 28.0 -73.6 0.0 Net domestic financing 10.9 -9.5 -17.1 16.0 -8.7 -34.7 14.3 Banking system 12.3 -8.0 -16.2 16.8 12.0 -7.6 21.4 Nonbank financing -1.4 -1.5 -0.9 -0.8 -20.2 -15.6 5.8 Amortization of domestic debt 0.0 0.0 0.0 0.0 -0.5 -11.5 -12.9 (Wnafcnt) Memorandum items: BalancelGDP Excluding grants -10.0 -8.7 -8.8 -8.2 -7.8 -7.5 -7.2 Excluding mining revenue -19.9 -19.2 -19.5 -16.7 -13.9 -12.0 -10.1 Sources: Data provided by the Guinean authonties; and staff estimates. 89 Annex 1 Table 1.9 Guinea: Monetary Survery, 1988-1994 f9,88 :'1999 Itp1988V (in billions of Guinean francs: end of Denod) Net foreign assets -1.2 13.4 59.0 81.1 112.5 161.3 121.3 Central bank -4.6 6.4 44.3 57.0 74.2 123.7 88.1 Deposit money banks 3.4 7.0 14.7 24.1 38.3 37.6 33.2 Net domestic assets 105.7 92.6 95.1 122.4 158.0 161.4 183.3 Domestic credit 92.1 92.6 95.1 122.4 158.0 161.4 212.3 Public sector 47.5 39.5 23.5 42.4 53.5 45.9 67.9 Government (net) 47.3 39.3 23.1 39.9 51.1 43.4 64.8 State enterprises (gross) 0.2 0.2 0.4 2.5 2.4 2.5 3.1 Private Sector 44.6 53.1 71.6 80.0 104.5 115.5 144.4 Healthy credit 38.5 47.0 63.0 62.0 83.6 106.3 119.4 Nonperforming 6.1 6.1 8.6 18.0 20.9 9.2 25.0 Other items 13.6 0.0 0.0 0.0 0.0 0.0 -29.0 Broad Money 104.5 121.0 154.1 207.7 256.3 315.1 304.6 Money and quasi-money 96.5 112.9 144.4 194.2 238.7 283.5 276.1 Currency 55.7 69.4 84.7 119.4 133.0 166.6 154.8 Demand deposits 24.4 31.4 42.3 66.1 84.7 93.7 97.6 Term deposits 16.4 12.1 17.4 8.7 21.0 23.2 23.7 Foreign currency deposits 8.0 8.1 9.7 13.5 17.6 31.6 28.5 (As a percentaae) Memorandum items GDPIBroad Money 9.2 8.4 8.3 9.2 9.6 10.4 10.9 Currencyldeposits 136.5 159.5 141.9 159.6 125.8 142.5 127.6 Sources: Data provided by the Guinean authorities, World Bank and IMF staff estimates. 90 Annex I Table 1.10 Guinea: Balance of Payments, 1988-1994 (In millions of U.S. dollars. unless otherwise indicated) Trade balance 5.5 144.2 93.5 55.3 -83.1 -65.8 -59.1 Exports (f.o.b) 609.1 731.1 816.7 790.2 656.8 665.0 625.9 Mining exports 573.7 677.1 756.2 719.3 577.4 574.7 498.5 Of which: mining companies 500.7 615.1 678.6 644.3 498.6 466.4 374.9 Other 35.4 54.1 60.6 70.8 79.4 90.4 127.4 Imports (c.i.f.) -603.6 -586.9 -723.2 -734.9 -739.9 -730.9 -685.0 Public sector -159.8 -162.1 -168.0 -153.8 -152.6 -146.6 -138.2 Mining companies -151.8 -140.5 -230.3 -256.4 -234.2 -217.9 -152.6 Other private -292.0 -284.3 -324.9 -324.8 -353.1 -366.5 -394.1 Services, net -283.2 -317.5 -299.7 -260.6 -220.6 -221.4 -216.1 Of which: interest due -106.8 -99.4 -100.9 -92.1 -69.3 -67.9 -76.9 Private transfers -24.9 -42.9 -50.6 -42.2 -63.5 -52.3 -39.7 Official transfers 83.5 98.0 100.9 109.5 135.6 117.9 120.2 Current account balance Including official transfers -219.0 -118.2 -155.8 -138.1 -231.5 -221.7 -194.6 Excluding official transfers -302.5 -216.2 -256.7 -247.5 -367.1 -339.6 -314.8 Of which: mining companies 209.6 292.0 286.0 256.0 150.7 136.2 132.9 Capital movements 76.4 119.4 113.1 106.0 156.6 189.1 50.8 Public, medium- and long-term, ne 51.9 93.0 61.0 52.0 61.8 105.1 -7.7 Project-related loans .146.4 153.5 181.2 175.1 160.2 150.5 131.6 Program loans 46.9 60.7 3.8 32.4 30.4 52.5 7.7 Amortization due -141.4 -121.1 -124.1 -155.4 -128.8 -97.9 -146.9 Public, short-term 0.0 0.0 11.3 -1.2 -1.2 -4.5 -4.5 Mining companies 18.0 16.4 29.9 43.8 83.4 62.7 24.6 Private direct investment 6.5 10.0 10.9 11.4 12.6 15.8 25.9 Other private capital 0.0 0.0 0.0 0.0 0.0 10.0 12.5 Errors and omissions 15.1 -29.3 5.0 -35.6 -7.5 21.1 0.0 Overall balance -127.6 -28.0 -37.7 -67.6 -82.5 -11.4 -143.7 Of which: mining companies 227.6 308.4 315.9 299.8 234.0 198.9 157.5 Financing items 127.6 28.0 37.7 67.6 82.5 11.4 143.7 Official reserves, net 26.9 -21.7 -50.7 -18.2 -3.3 -49.3 34.5 Changes in arrears, net 77.5 -26.1 27.9 9.3 -329.6 79.3 109.2 Debt relief 23.2 75.8 60.5 76.5 384.4 58.9 0.0 Deferred payments 0.0 0.0 0.0 0.0 31.0 -77.5 0.0 Memorandum items: (In percent) Current Account BalancelGDP Including official transfers -9.2 -4.9 -5.5 -4.6 -7.8 -7.0 -5.7 Excluding official transfers -12.7 -8.9 -9.1 -8.3 -12.3 -10.7 -9.3 Sources: Data provided by the Guinean authorities; and staff estimates. 91 Annex I Table 1.11 Guinea: External Debt Service, 1988-1994 (in millions of U.S. dollars) Debt service due before rescheduling 288.8 246.1 262.7 262.1 209.0 191.0 260.7 Interest 106.8 99.4 100.9 92.1 69.3 67.9 76.9 Principal 182.0 146.7 161.8 170.0 139.7 123.1 183.8 Debt rescheduling 23.2 29.7 0.0 0.0 68.2 0.0 0.0 Interest 3.5 15.9 0.0 0.0 21.1 0.0 0.0 Principal 19.7 13.8 0.0 0.0 47.1 0.0 0.0 Debt service due after rescheduling 265.6 216.4 262.7 262.1 140.7 190.9 260.8 Interest 103.3 83.5 100.9 92.1 48.2 67.9 76.9 Principal 162.3 132.9 161.8 170.0 92.6 123.1 183.9 (In percent of exports of goods) Memorandum items: Debt service due before rescheduling 47.4 33.7 32.2 33.2 31.9 28.7 41.7 Interest 17.5 13.6 12.4 11.7 10.6 10.2 12.3 Principal 29.9 20.1 19.8 21.5 21.3 18.5 29.4 Debt service due after rescheduling 43.6 29.6 32.2 33.2 21.4 28.7 41.7 Interest 16.9 11.4 12.4 11.7 7.3 10.2 12.3 Principal 26.7 18.2 19.8 21.5 14.1 18.5 29.4 (In millions of U.S. dollars) External debt outstanding (end of period) 2074.7 2024.5 2123.5 2400.5 2474.7 2593.5 2788.1 Source: Data provided by the Guinean authorities. 92 Annex 1 Table 1.12 Sectoral Composition of Investment Budget, 1988-1994 ~iY2IZi.988 , 1990~ -199 .j 1992 1993 1994 Average (In millions of Guinean francs, constant 1989 prices) Agriculture 21,511 32,074 28,217 44,713 39,330 37,359 33,867 Livestock 3,751 3,086 2,148 1,867 1,942 2,535 2,555 Fishing 7,113 7,919 6,693 4,289 3,295 2,069 5,230 Forests 9,847 12,353 13,548 10,556 14,988 8,195 11,581 Mines 12,632 9,718 9,044 2,341 1,932 1,266 6,155 Industry[Tourism 3,662 1,194 934 566 1,683 1,319 1,560 Water 3,290 10,809 12,270 13,135 9,773 4,995 9,046 Energy 10,952 8,342 6,699 4,514 9,090 8,382 7,997 Construction and Urbanism 13,962 13,236 20,005 11,495 8,969 10,598 13,044 Transportation + Public Works 49,439 53,023 45,566 51,363 53,399 45,340 49,688 Education 12,409 10,756 15,617 19,438 12,675 9,978 13,479 Health 11,370 13,030 12,847 12,516 11,303 14,135 12,534 Other Social Sectors 1,845 2,742 1,332 372 1,096 814 1,367 Administration 8,524 11,915 10,651 9,038 12,969 11,540 10,773 TOTAL PIP .70.307 190.197 185l1 186.204 182.44 158.524 178.84 Sectoral Composition of Investment Expenditures, 1988-1994 (In millions of Guinean francs, current prices) Agriculture 17,197 38,964 45,585 64,741 59,610 54,748 46,808 Livestock 2,325 2,668 4,950 2,494 2,595 -1,865 2,816 Fishing 6,533 10,368 7,304 5,632 4,739 145 5,787 Forests 8,121 15,243 9,861 18,354 16,891 11,340 13,302 Mines 9,643 14,684 16,115 2,623 528 721 7,386 Industry/Tourism 1,758 2,079 510 385 1,084 364 1,030 Water 5,001 3,814 10,560 36,037 22,973 11,316 14,950 olw Urban Water 5,001 3,814 10,560 35,532 22,883 11,220 14,835 olw Hydraulics 0 0 0 505 90 95 115 Energy 6,740 5,090 4,148 7,215 14,534 13,179 8,485 Construction and Urbanism 9,149 11,495 17,800 15,990 13,142 6,592 12,361 Transportation + Public Works 38,665 58,544 45,310 73,002 80,813 45,357 56,948 Public Works 29,473 44,987 28,417 70,424 77,534 45,318 49,359 Transportation 9,192 13,556 16,893 2,578 3,280 39 7,590 Education 5,831 7,821 17,632 11,164 17,581 3,633 10,610 Health 10,085 10,623 11,733 8,582 17,075 8,522 11,103 Other Social Sectors 963 661 922 238 321 58 527 Administration 4,642 11,151 14,616 8,701 13,176 6,705 9,832 Telecommunications 4,724 7,729 4,663 3,522 10,961 4,081 5,946 PIP TOTAL 131 377 200.934 211,707 258.681 276023 168.,625 207,891 93 Annex I Table 1.13 Sectoral Composition of Investment Expenditures, 1988-1994 1988; 199O 1991 1992 1993 1994 Aveage (In millions of Guinean francs, constant 1989 prices) Agriculture 21,023 31,448 31,180 38,196 32,538 28,455 30,473 Livestock 2,843 2,153 3,385 1,472 1,416 969 2,040 Fishing 7,987 8,368 4,996 3,323 2,587 75 4,556. Forests 9,928 12,303 6,745 10,828 9,220 5,894 9,153 Mines 11,788 11,852 11,022 1,548 288 374 6,145 Industry/Tourism 2,149 1,678 348 227 592 189 864 Water 6,113 3,078 7,223 21,261 12,540 5,881 9,349 olw Urban Water 6,113 3,078 7,223 20,963 12,491 5,832 9,283 o/w Hydraulique 0 0 0 298 49 49 66 Energy 8,239 4,108 2,837 4,257 7,934 6,850 5,704 Construction and Urbanism 11,185 9,277 12,175 9,434 7,174 3,426 8,779 Transportation + Public Works 47,267 47,251 30,992 43,069 44,112 23,574 39,378 Public Works 36,030 36,309 19,437 41,548 42,322 23,554 33,200 Transportation 11,237 10,941 11,555 1,521 1,790 20 6,177 Education 7,129 6,313 12,060 6,586 9,596 1,888 7,262 Health 12,329 8,574 8,025 5,063 9,321 4,429 7,957 Other Social Sectors 1,177 534 631 140 175 30 448 Administration 5,675 9,000 9,997 5,133 7,192 3,485 6,747 Telecommunications 5,775 6,238 3,189 2,078 5,983 2,121 4,231 Total 160.607 162.174 144.807 152.614 150.667 8I643 143.085 1988 1990 1991 1992 1993 1994 Average: (As percent share) Agriculture 13% 19% 22% 25% 22% 32% 22% Livestock 2% 1% 2% 1% 1% 1% 1% Fishing 5% 5% 3% 2% 2% 0% 3% Forests 6% 8% 5% 7% 6% 7% 6% Mines 7% 7% 8% 1% 0% 0% 4% Industryffourism 1% 1% 0% 0% 0% 0% 1% Water 4% 2% 5% 14% 8% 7% 7% olw Urban Water 4% 2% 5% 14% 8% 7% 7% o/w Hydraulique 0% 0% 0% 0% 0% 0% 0% Energy 5% 3% 2% 3% 5% 8% 4% Construction and Urbanism 7% 6% 8% 6% 5% 4% 6% Transportation + Public Works 29% 29% 21% 28% 29% 27% 27% Public Works 22% 22% 13% 27% 28% 27% 23% Transportation 7% 7% 8% 1% 1% 0% 4% Education 4% 4% 8% 4% 6% 2% 5% Health 8% 5% 6% 3% 6% 5% 6% Other Social Sectors 1% 0% 0% 0% 0% 0% 0% Administration 4% 6% 7% 3% 5% 4% 5% Other 4% 4% 2% 1% 4% 2% 3% Intal i9.A 100% 1.0% 100M UDYi 1900% 19"rA 94 Annex I Table 1.14 Sectoral Composition of Current Budget, 1988-1994 1988 1990 1991 1992- 1993 1994 g (In millions of Guinean francs, constant 1989 prices) General Public Services 40,369 43,156 40,716 52,169 52,980 40,928 45,053 Defense 16,182 18,220 23,763 22,327 22,569 23,339 21,066 Administration + Others 24,187 24,936 16,953 29,842 30,411 17,589 23,986 Economic Services 15,878 15,381 17,885 17,043 15,722 18,511 16,737 Agriculture 6,466 5,455 7,846 9,310 9,078 9,486 7,940 Energy 793 698 1,069 1,251 1,231 128 862 Industry/Tourism/Mines/Water 4,372 5,970 4,889 5,646 4,597 5,342 5,136 Transportation + Public Works 3,518 2,667 3,326 836 816 3,377 2,423 Construction and Urban Development 729 591 756 0 0 179 376 Social Services 24,096 24,654 39,230 45,273 45,763 41,648 36,777 Education 17,253 19,166 32,092 36,719 36,940 33,234 29,234 Health 6,253 4,522 6,218 7,541 7,370 6,431 6,389 Social Sectors 591 966 921 1,013 1,453 1,982 1,154 Regions 649 482 1,613 1,546 180 0 745 Common exp. net of debt relief 152,966 147,121 193,433 123,550 148,210 200,088 160,895 less Public Debt (1) 41,224 41,893 121,712 59,010 88,612 137,733 81,697 Common Expenditures net (2) 111,742 105,228 71,721 64,540 59,598 62,356 79,198 olw extemal/dom. interest on debt 45,944 39,890 37,251 29,372 23,213 28,229 33,983 olw O&M expenditures SBK (3) 16,427 9,805 14,004 14,656 6,784 54 10,288 olw personnel and O&M 42,813 43,398 12,189 14,356 20,287 16,823 24,978 olw transfers (4) 6,558 12,135 8,277 6,156 9,313 17,250 9,948 TOAL 192,734 188.901 171,166 180,571 174,243 163A42 178.50 '988 1 99ge (As percent share) General Public Services 21% 23% 24% 29% 30% 25% 25% Defense 8% 10% 14% 12% 13% 14% 12% Administration + Others 13% 13% 10% 17% 17% 11% 13% Economic Services 8% 8% 10% 9% 9% 11% 9% Agriculture 3% 3% 5% 5% 5% 6% 4% Energy 0% 0% 1% 1% 1% 0% 0% Industry[Tourism/Mines/Water 2% 3% 3% 3% 3% 3% 3% Transportation + Public Works 2% 1% 2% 0% 0% 2% 1% Construction and Urban Development 0% 0% 0% 0% 0% 0% 0% Social Services 13% 13% 23% 25% 26% 25% 21% Education 9% 10% 19% 20% 21% 20% 17% Health 3% 2% 4% 4% 4% 4% 4% Social Sectors 0% 1% 1% 1% 1% 1% 1% Regions 0% 0% 1% 1% 0% 0% 0% Common exp. net of debt relief 79% 78% 113% 68% 85% 122% 91% less Public Debt (1) 21% 22% 71% 33% 51% 84% 47% Common Expenditures net (2) 58% 56% 42% 36% 34% 38% 44% o/w extemal/dom. interest on debt 24% 21% 22% 16% 13% 17% 19% olw O&M expenditures SBK (3) 9% 5% 8% 8% 4% 0% 6% o/w personnel and O&M 22% 23% 7% 8% 12% 10% 14% o/w transfers (4) 3% 6% 5% 3% 5% 11% 6% TOTAL 100% 10% 1M9A 1.00% "A% i (1) including: arrears, amortization, deferred payments, and ST credits, all of which appear below the line in the TOFE. (2) including interest on extemal and domestic debt, which appears above the line in the TOFE. (3) SBK/OBK is the GOG-owned bauxite mine. (4) including dette viagdre (pensions), remboursements droits. 95 Annex 1 Table 1.15 Sectoral Composition of Current Expenditures, 1988-1994 1988 14990 1991 1992 1993 1994 Average (In millions of Guinean francs. constant 1989 prices) General Public Services 41,100 44,035 46,688 52,530 50,037 42,047 46,073 Defense 16,475 18,591 27,248 22,481 21,315 23,977 21,681 Administration + Others 24,625 25,444 19,440 30,049 28,722 18,070 24,392 Economic Services 16,165 15,694 20,508 17,160 14,849 19,018 17,232 Agriculture 6,583 5,566 8,997 9,374 8,574 9,745 8,140 Energie 807 713 1,225 1,259 1,163 132 883 Industry/TourismMines/Water 4,451 6,091 5,606 5,685 4,342 5,488 5,277 Transportation + Public Works 3,581 2,721 3,814 842 771 3,469 2,533 Construction and Urban Development 742 603 867 0 0 184 399 Social Services 24,533 25,156 44,984 45,586 43,221 42,786 37,711 Education 17,565 19,556 36,799 36,973 34,889 34,143 29,987 Health 6,366 4,614 7,130 7,593 6,961 6,607 6,545 Social Sectors 602 986 1,056 1,020 1,372 2,037 1,179 Regions 661 492 1,850 1,557 170 6 789 Common exp. net of debt relief 289,655 155,693 140,214 177,325 122,169 117,412 167,078 less Public Debt (1) 176,721 49,134 63,438 112,541 64,592 54,123 86,758 Common Expenditures net (2) 112,934 106,559 76,776 64,784 57,577 63,289 80,320 olw external/dom. interest on debt 45,944 39,890 37,251 29,372 23,213 28,229 33,983 o/w O&M expenditures SBK (3) 16,725 10,005 16,058 14,758 6,408 55 10,668 o/w personnel and O&M 43,589 44,282 13,976 14,455 19,160 17,283 25,458 olw transfers (4) 6,677 12,382 9,490 6,199 8,796 17,722 10,211 + Eror 833 812 5,464 203 -1,290 772 TOTAI 196,226 192,747 196.270 181.820 164.565 167.918 182.125 198 1902 991i 192 1993 1994: Averag (As percent share) General Public Services 21% 23% 24% 29% 30% 25% 25% Defense 8% 10% 14% 12% 13% 14% 12% Administration + Others 13% 13%. 10% 17% 17% 11% 13% Economic Services 8% 8% 10% 9% 9% 11% 9% Agriculture 3% 3% 5% 5% 5% 6% 4% Energy 0% 0% 1% 1% 1% 0% 0% lndustrylTourism/Mines/Water 2% 3% 3% 3% 3% 3% 3% Transportation + Public Works 2% 1% 2% 0% 0% 2% 1% Construction and Urban Development 0% 0% 0% 0% 0% 0% 0% Social Services 13% 13% 23% 25% 26% 25% 21% Education 9% 10% 19% 20% 21% 20% 17% Health 3% 2% 4% 4% 4% 4% 4% Social Sectors 0% 1% 1% 1% 1% 1% 1% enions 0% 0% 1% 1% 0% .0% 0% Common exp. net of debt relief 148% 81% 71% 98% 74% 70% 90% less Public Debt (1) 90% 25% 32% 62% 39% 32% 47% Common Expenditures net (2) 58% 55% 39% 36% 35% 38% 43% olw extemal/dom. interest on debt 23% 21% 19% 16% 14% 17% 18% olw O&M expenditures SBK (3) 9% 5% 8% 8% 4% 0% 6% olw personnel and O&M 22% 23% 7% 8% 12% 10% 14% olw transfers (4) 3% 6% 5% 3% 5% 11% 6% error 0% 0% 3% 0% -1% 0% 1% TOTAL 100% i nY m% in0h 1Y 10% 1o0% (1) including: arrears, amortization, deferred payments, and ST credits, all of which appear below the line in the TOFE. (2) including interest on domestic and intemal debt, which appears above the line in the TOFE. (3) SBKIOBK is the GOG-owned bauxite mine. (4) including dette viagdre (pensions) and remboursements droits. 96 Annex I Table 1.16 Guinea: Sectoral Composition of Total Budget, 1988-1994 (In millions of Guinean francs. current prices) General Public Services 38,428 66,259 73,720 102,172 118,249 98,175 82,834 Defense 12,609 21,741 33,937 37,170 40,252 43,323 31,505 Administration + Others 25,819 44,518 39,783 65,002 77,998 54,852 51,329 Economic Services 123,750 206,376 237,714 273,875 292,581 269,200 233,916 Agriculture 39,575 75,189 85,189 119,613 125,295 114,112 93,162 Energy 9,577 11,169 11,321 9,734 18,848 16,364 12,836 Industry/Tourism/Mines/Water 19,427 34,037 39,509 36,592 32,725 24,500 31,132 Transportation + Public Works 43,182 68,878 71,368 88,453 99,282 93,502 77,444 Construction and Urban Development 11,989 17,104 30,327 19,485 16,431 20,722 19,343 Social Services 39,736 62,287 99,587 130,165 127,554 125,269 97,433 Education 23,594 36,196 68,664 94,078 89,104 80,889 65,421 Health 14,173 21,540 27,662 33,769 33,851 39,134 28,355 Social Sectors 1,970 4,551 3,262 2,318 4,599 5,246 3,658 Regions 506 575 2,304 2,575 322 10 1,048 Common Expenditures 87,070 125,566 102,427 107,449 106,292 115,749 107,426 TOTAL 289,490 461,064 515,752 616,235 644,997 608,403 522,657 .488 1990 1991 j 1992 1993 , 1994 Average General Public Services 13% 14% 14% 17% 18% 16% 15% Defense 4% 5% 7% 6% 6% 7% 6% Administration + Others 9% 10% 8% 11% 12% 9% 10% Economic Services 43% 45% 46% 44% 45% 44% 45% Agriculture 14% 16% 17% 19% 19% 19% 17% Energy 3% 2% 2% 2% 3% 3% 3% lndustry/Tourism/Mines/Water 7% 7% 8% 6% 5% 4% 6% Transportation + Public Works 15% 15% 14% 14% 15% 15% 15% Construction and Urban Development 4% 4% 6% 3% 3% 3% 4% Social Services 14% 14% 19% 21% 20% 21% 18% Education 8% 8% 13% 15% 14% 13% 12% Health 5% 5% 5% 5% 5% 6% 5% Social Sectors 1% 1% 1% 0% 1% 1% 1% Regions 0% 0% 0% 0% 0% 0% 0% Common Expenditures 30% 27% 20% 17% 16% 19% 22% TOTAL 100% 1m in 19i % 1im 1% 97 Annex I Table 1.17 Guinea: Sectoral Composition of Total Budget, 1988-1994 # 98 10 191992 93 194 Average, (in millions of Guinean francs. constant 1989 GNF) General Public Services 48,892 55,071 51,367 61,207 65,949 52,467 55,826 Defense 16,182 18,220 23,763 22,327 22,569 23,339 21,066 Administration + Others 32,711 36,851 27,604 38,880 43,380 29,128 34,759 Economic Services 152,036 167,134 163,010 161,882 160,122 140,569 157,459 Agriculture 48,687 60,886 58,451 70,734 68,633 59,644 61,172 Energy 11,745 9,041 7,768 5,765 10,321 8,510 8,858 Industry/Tourism/Mines/Water 23,956 27,691 27,137 21,688 17,985 12,922 21,897 Transportation + Public Works 52,956 55,690 48,892 52,199 54,215 48,716 52,112 Construction and Urban Development 14,691 13,826 20,761 11,495 8,969 10,777 13,420 Social Services 49,720 51,182 69,026 77,599 70,837 66,575 64,157 Education 29,661 29,921 47,709 56,157 49,616 43,212 42,713 Health 17,623 17,552 19,065 20,057 18,673 20,566 18,923 Social Sectors 2,436 3,708 2,252 1,385 2,549 2,796 2,521 Rggions 649 482 1,613 1,546 180 745 Common Expenditures 111,742 105,228 71,721 64,540 59,598 62,356 '79,198 TOTAL 363a04 379.098 356.736 366.774 356,67 321.966 357,38 98 Annex 1 Table 1.18 Guinea: Sectoral Composition of Total Expenditures, 1988-1994 (In millions of Guinean francs. current prices) General Public Services 41,391 71,425 85,955 99,675 113,378 88,836 83,443 Defense 12,837 22,184 38,914 37,427 38,016 44,508 32,314 Administration + Others 28,554 49,241 47,041 62,248 75,362 44,328 51,129 Economic Services 117,727 181,676 191,430 255,044 243,392 180,927 195,033 Agriculture 39,306 73,886 80,548 106,828 99,124 86,187 80,980 Energy 7,369 5,940 5,898 9,312 16,608 13,424 9,759 Industry/Tourism/Mines/Water 19,870 27,845 35,190 48,510 32,329 22,586 31,055 Transportation + Public Works 41,455 61,790 50,757 74,404 82,188 51,797 60,399 Construction and Urban Development 9,728 12,214 19,038 15,990 13,142 6,933 12,841 Social Services 35,996 49,123 94,530 95,877 112,061 91,637 79,871 Education 19,518 31,157 70,185 72,717 79,804 67,011 56,732 Health 15,045 16,129 21,916 21,223 29,489 20,787 20,765 Social Sectors 1,432 1,838 2,429 1,936 2,768 3,839 2,374 Regions 515 587 2,642 2,592 304 10 1,108 Common Expenditures 87,999 127,154 109,647 107,854 102,688 117,481 108,804 error 650 1,000 7,800 300 -2,300 1,400 TOTAL 283,628 429,965 484,204 561,043 571,822 478,892 468,259 As percentage of GDP General Public Services 4% 4% 4% 4% 4% 3% 3% Defense 1% ,1% 2% 1% 1% 1% 1% Administration + Others 3% 3% 2% 2% 3% 1% 2% Economic Services 10% 10% 9% 10% 8% 6% 8% Agriculture 3% 4% 4% 4% 3% 3% 3% Energy 1% 0% 0% 0% 1% 0% 0% Industry/Tourism/MinesMater 2% 2% 2% 2% 1% 1% 1% Transportation + Public Works 4% 3% 2% 3% 3% 2% 2% Construction and Urban Development 1% 1% 1% 1% 0% 0% 1% Social Services 3% 3% 4% 4% 4% 3% 3% Education 2% 2% 3% 3% 3% 2% 2% Health 1% 1% 1% 1% 1% 1% 1% Social Sectors 0% 0% 0% 0% 0% 0% 0% Regions 0% 0% 0% 0% 0% 0% 0% Common Expenditures å2% L4% älu 42% 38% 4.7% error 0% 0% 2% 0% 0% 0% 0% TOTAL 26% MI 24% 220 20% 1§6%119% 99 Annex I Table 1.19 Tax-GDP ratio in selected SPA countries (excluding grants) Ghana 13.8 21.4 Guinea 13 3 12.8 : Malawi 18.5 18.7 Mauritania 22.4 25.5 Senegal 17.7 14.9 Median SPA 14.6 15.0 100 ANNEX 2 ASSUMPTIONS AND ANALYSIS OF THE UNSUSTAINABLE SCENARIO Current Trends Would Lead to an Unsustainable Budgetary Deficit 1. The methodology used in chapter 5, section B entails developing a basic current-trends scenario through the year 2000 based on the hypotheses presented in chapter 5, table 5.1. This scenario assumes that the Government maintains current trends, with minimal efforts to correct and reverse the economic and financial deterioration; in other words, it is a business-as-usual scenario. But, it does assume that economic activity in the export sector, particularly in mining, will respond to current Government efforts to expand output by rehabilitating key mining enterprises. The first simulation shows that the emerging deficit would necessitate large external borrowing, since domestic financing possibilities are limited. The second simulation calculates implicitly the level of public expenditures consistent with viable deficits; i.e., deficits that can be financed without causing adverse effects on the economy. Simulation 1 indicates that massive foreign borrowing would lead to an unsustainable external debt profile over the medium-term; simulation 2 indicates that reduced levels of public expenditures would disrupt the efficient functioning of the economy and hurt the social sectors. Assumptions Underlying the Current-Trends Scenario 2. For the real sector, the scenario assumes that GDP and its deflator would grow annually by 4.6 to 5% and by 4 to 5%, respectively, through the year 2000. These rates reflect the recent improvements in the economic performance of Guinea's economy with the policies and measures adopted by the Government to move toward a more decentralized system of economic management. In addition, the scenario assumes that the ratio of nonmining to mining GDP would increase slightly given the Government's policy to diversify the economy. Private consumption as a share of GDP would decline as the formal-sector is increasingly taxed. This assumption reflects the Government's frustration with revenue collection efforts: because it is unable to mobilize receipts, it raises taxes on the increasingly narrower tax base of the formal sector, thereby reducing purchasing power and mitigating consumption expenditures. Meanwhile, investment as a share of GDP would rise slightly during the period, but the incremental capital output ratio (ICOR) would increase from 3.1 in 1995 to 3.5 by the year 2000, in part reflecting the poor state of Guinea's capital stock, particularly its infrastructure. Undoubtedly, if the Government were to reallocate expenditures toward improving health and education services, while maintaining transport, power, and communication infrastructures, then it would stem the increase in the ICOR. Balance-of-payments projections indicate that current account deficits would exist throughout the period, with a negative savings/investment gap and insufficient foreign transfers to correct the imbalance. 3. For government budgetary revenue, the scenario assumes that nonmining receipts would grow at the same rate as nominal GDP. This unitary elasticity assumption is based on the Government's difficulties in mobilizing revenue for the past several years. Mining-based revenue projections, however, are consistent with the mining export projections of the balance of payments and reflect the new investments anticipated in the mining sector. These assumptions would lead to an average annual growth of 9.9% in total revenue during the 1995-2000 period, with budgetary 101 revenue reaching GNF 603.6 billion by the end of the century. Already when compared with the current Policy Framework Paper (PFP) projections for 1996-98, which assume that the Government measures are adopted, a shortfall of more than GNF 255 billion would emerge by 1998. This shortfall reflects the Government's unwillingness or inability to adopt the necessary measures to strengthen its financial operations. Consistent with the projections in the PFP, total expenditures would increase at an average annual rate of 11.1% (to GNF 1,091 billion) by the end of the decade, as current expenditures grow by 9.3% annually and capital expenditures by 13% (see annex 2, table 2.4). 4. For the 1996-98 period, the external sector projections are largely those retained in the PFP document; nevertheless, there are important differences. The projections are based on fluctuating terms of trade (annex 2, table 2.1), in which a trade surplus emerges beginning in 1997 as mining exports pick up. The projected increase in mining output is a response to an increase in foreign investments as the Bank-financed mining sector investment promotion project is implemented. Export and import price projections are based on forecasts by the Bank's DEC-IEC and the IMF's Research Department. The growing external current account deficit will necessitate higher levels of external financing in the form of capital inflows during the 1994-2000 period, which will raise the stock of external debt. The implication of this foreign financing for Guinea's external debt sustainability is assessed in the forthcoming sections. Annex 2 Table 2.1 Terms of trade projections: 1994-2000 (1989 = US$100) _____________ 1994 1995 1:996 r1:997 ..1993 . 199. 200. Export prices 84.5 90.2 93.0 90.6 91.3 90.1 90.0 Import prices 112.6 117.0 117.6 119.0 119.6 120.9 122.3 Ratio 75.0 77.1 79.1 76.1 76.3 74.5 73.6 Percentage change. -4.8 2.8 2.6 -3.8 0.3 -2.4 -1.2 Scenario Outcomes and Implications 5. Since the scenario is based on the premise that the authorities are either unwilling or incapable to act on either the revenue or expenditure fronts, the budgetary deficit is forecast to deteriorate during the period. The deficit (including grants) would rise from 3.6% of GDP in 1994 to 6% in the year 2000 (annex 2, table 2.2). Note that these deficits are defined on a commitment basis, since no assumption is made on changes in arrears. The crucial question that arises at this point is how the Government will finance an almost tripling of the budgetary deficit to GNF 338.5 billion in the year 2000. For the 1995-2000 period, the cumulative deficit, including grants, would exceed GNF 1.472 trillion. 102 Annex 2 Table 2.2 Government Financial Operations Projections: 1994-2000 (billions of current GNF) 1994 1995 1996 1997 1998 1999 000 Total revenue 343.3 389.1 419.9 461.2 493.1 560.3 603.6 Foreign grants 117.6 122.4 127.3 132.4 137.7 143.2 148.9 Total expenditures 580.5 637.0 735.4 829.5 917.2 1,001. 1,091.0 Overall balance 4 (- deficit) " -119.8 -125.5 -188.2 -235.9 -286.4 -297.9 -338.5 (-deficit)' -237.4 -247.9 -315.5 -368.3 -424.1 -441.4 -487.4 Memorandum items GDP 3,316 3,642 3,965 4,322 4,715 5,149 5,622 Overall balance d (incl. grants) 3.6 3.5 4.8 5.5 6.1 5.8 6.0 (excl. grants) 7.2 6.8 8.0 8.5 9.0 8.6 8.7 Viable deficit d.c (incl. grants) .... 2.1 1.9 1.6 1.6 1.1 0.9 Value of deficit .... 76.5 75.3 69.2 75.4 56.6 50.6 Expenditure consistent with viable deficit .... 588.0 622.5 662.8 706.2 760.1 803.1 Sources: Scenario assumptions and staff calculations; see chapter 5, table 5.1. a Grants are assumed to grow by 4% annually. b Commitment basis, including foreign grants. c Commitment basis, excluding foreign grants. d As percent of GDP, on a commitment basis. IMF projections. 6. Budgetary deficits can be financed from domestic sources and/or external sources. Net domestic financing consists of bank financing and/or nonbank private financing, including a rise in domestic arrears. Except for financing through the accumulation of arrears, domestic nonbank private financing is almost nonexistent in Guinea, which is unlikely to change in the future. Meanwhile, Central Bank financing is in principle governed by the rule that net credit to Government cannot exceed 20% of the previous year's tax revenue. This requirement has largely been respected, although the limit was exceeded in the first semester of 1995, reflecting the tight liquidity position of the Treasury. Notwithstanding this noncompliance, Treasury policy in the past few years has been to disengage itself from the banking system in order not to crowd out the private sector. This was indeed the case in 1989-90 and again in 1993, when net changes in bank credit to Government were negative. But the projections of this scenario do not imply that the Government can pursue this policy in the future. Assuming that tax revenue continues to represent an average of 93% of total budgetary revenue, the theoretical limits on net credit to Government during 1995-2000 are GNF 46 billion (1995), GNF 50 billion (1996), GNF 54 billion (1997), GNF 59 billion (1998), GNF 65 billion (1999), and GNF 71 billion (2000). Thus, even if Central Bank financing were used to the limits permitted, foreign financing would be substantial for the six-year period. Other forms of domestic financing are still limited at this time, despite an emerging treasury bill market. Although the incidence of massive bank financing of deficits leads to monetary creation and inflation, as has been discussed widely in the economics literature, this scenario assumes that inflation will be kept under control and thus that the Central Bank will uge monetary policy prudently. Although the build-up of domestic arrears can be used to finance Government operations in the short term, domestic suppliers will eventually cease to provide goods and services to the 103 Government. Thus, foreign financing would be necessary, which would contribute to the increase in foreign indebtedness. The implications of these projected developments are clear: Guinea cannot continue to pursue budgetary policies that do not account for its fundamental weaknesses in revenue mobilization. In addition, the Government must enhance its public expenditure procedures so that a more rational management of expenditures will enable the economy to function more efficiently. 7. The financing requirements of the budget and of the external current account will force Guinea's external debt to rise. The following simple model illustrates how Guinea's external debt would become unsustainable in the medium term. We show that if the current economic and financial situation were to persist, then Guinea's external debt as a ratio to GDP would continue to rise and become nonserviceable, leading to a pervasive accumulation of arrears. Basically, external debt rises when capital inflows are used to finance the excess of imports over exports, as well as the interest payments on such debt. The capital inflows increase the country's indebtedness vis b vis the rest of the world. The interaction among several variables and parameters determines the evolution of debt: the growth rate of GDP (g), the ICOR (k), the average propensity to save (s), and the average interest rate paid on external debt (i). These variables can be used to derive the following relationship, illustrating the rate of growth of the ratio of debt to GDP (r): D/Q where D is the stock of debt, and Q is GDP. For a given country, r will remain the same or will grow or diminish depending on whether: D/Q Based on recent economic and financial performance in Guinea, the first term of the expression would become larger than (g), implying that the rate of growth of the ratio of debt over GDP will be positive. 8. Annex 2, table 2.3 summarizes the key statistics of the two financing simulations. In both simulations, the growth rates in budgetary revenue are identical (9.9%) or slightly less than the growth in nominal GDP; this rise in revenue is clearly insufficient to prevent the emergence of unsustainable budgetary deficits In simulation 1, the limited domestic financing possibilities compel the authorities to seek foreign financing of public deficits. Under this simulation, budgetary expenditures rise at an average annual rate of 11.1% in nominal terms and 6.9% in real terms during the 1995-2000 period, reaching GNF 1,091 billion and GNF 866.3 billion, respectively. The budgetary deficit as a percentage of GDP, including grants, would increase from 3.5% in 1995 to 6% in 2000. The simulation implies that external indebtedness would rise by more than 3% of GDP during the same period, as is shown in annex 2, table 2.7. Under simulation 2, nominal expenditures would rise by 6.5% annually and by 2.3% in real terms, reaching GNF 803.1 billion and GNF 658.8 billion, respectively by the year 2000. While the resulting budgetary deficits are certainly financeable and will not endanger Guinea's medium-term debt profile, these so-called viable-based expenditures are insufficient to maintain minimum levels of spending on social and services other high-priority items, thus threatening overall economic functioning. Under this simulation, external 104 debt rises at an annual rate of 4.6% through year 2000, but declines by 19 percentage points of GDP. Annex 2 Table 2.3 Financing simulations (annual percentage change) S19 944.995-200( Actual Projected Simulation 1 (foreign financing of deficit) Growth in revenue -2.3 9.9 Growth in expenditures Nominal 0.2 11.1 Real -3.9 6.9 Growth of external debt 7.5 9.2 Simulation 2 (expenditure cutting)' Growth in revenue -2.3 9.9 Growth in expenditures Nominal ... 6.5 Real ... 2.3 Growth of external debt 7.5 4.6 Source: Staff estimates. Based on viable deficits. 9. The evolution in the stock of external debt under simulation 1 is shown in annex 2, table 2.7. During the 1994-2000 period, the stock of external debt (including the Garafiri/Kaleta debt) would rise at an average annual rate of 9.2%, reaching US$4,716 million (82.7% of GDP), compared with US$2,940 million in 1995 (79.6% of GDP). As shown in the table, the ratio of debt to GDP in 1995 falls by 2.5 percentage points from 82.1% in 1994, due to Paris Club debt cancellations; thereafter, it rises steadily. Continuing and persisting deficits in government financial operations imply that the Government will be compelled to borrow abroad on increasingly nonconcessionary terms as the economic situation deteriorates. Inasmuch as borrowing finances public and private consumption rather than productive capacity, exports would not rise fast enough, and the Government would be unable to reduce the current account deficit. Since the debt service requirements would continue to increase, Guinea would have to borrow further to prevent a reduction of imports. The persistence of this economic condition will introduce a deficit/external borrowing spiral that could can only be brought under control through a strengthening of government financial operations, particularly revenue mobilization and more rational public expenditures procedures. 10. The increase in the ratio of the stock of external debt to GDP would cut directly the rate of growth of potential GDP by diverting a portion of domestic output form growth-producing savings towards higher levels of debt service payments. There would be other indirect effects as well. The existence of a large debt overhang in the form of a high ratio of external debt to GDP can reduce the incentive for private investment because much of the forthcoming returns from investment must be used to repay existing debt and therefore acts as a tax on domestic investment. Furthermore, to the extent that a substantial debt leads to debt servicing difficulties, relations with external creditors may deteriorate, thus reducing the amount of trade financing Guinea can obtain. This in turn may make it more difficult or expensive to finance private investments, because imports play a major role in Guinea's investment projects, and economic performance is investment based. Other indirect negative effects on the rate of investment are low profitability due to depressed economic activity, and debt-induced declines in public investment that complements private investment. 105 Conclusions 11. The current-trends scenario shows that if the existing economic and financial situation is not remedied, it will lead Guinea to an abyss. Even though the ratio of nonmining sector revenue to GDP has increased more or less steadily from almost 1.8% of GDP in 1988 to 7.4% of GDP in 1994, the ratio remains low when compared with other Sub-Saharan African countries, and certainly insufficient to maintain a level of public expenditures compatible with the country's overall economic and social objectives. 12. The implications/outcomes of these scenarios are manifold and varied, and not necessarily mutually exclusive. They are as follows: (1) the authorities must embark on a comprehensive program to make the public finance sector self-sustainable; (2) the scope/role of the Government as it exists today must be trimmed and redefined; and (3) there must be a speedy transition to a market economy, with an enhanced private sector that can produce the goods and services demanded with minimum public sector interference. Annex 2 Table 2.4 Principal Scenario Assumptions, 1995-2000 1995,. 1996 - 1997 1998 1999; 20 Output (Annual percent change) Real GDP 4.6 4.7 4.8 4.9 5.0 5.0 Primary sector 5.0 5.0 5.0 5.0 5.0 5.1 Secondary sector 3.8 4.4 4.7 5.0 5.3 5.3 Tertiary sector 5.6 4.7 4.8 4.8 4.8 4.8 GDP deflator 5.0 4.0 4.0 4.0 4.0 4.0 Nominal GDP 9.8 8.9 9.0 9.1 9.2 9.2 ICOR (ratio) 3.1 3.2 3.3 3.4 3.4 3.5 Aggregate expenditures (ratio) Investment /GDP 14.8 15.2 15.6 15.9 16.0 16.1 Private consumption/GDP 81.1 78.6 77.4 77.1 75.4 75.5 Domestic saving/GDP 6.2 7.5 8.2 8.5 10.2 10.4 Public finance (Annual percent change) Revenue 13.34 7.92 9.84 6.92 13.63 7.73 Nonmining 9.83 8.89 8.99 9.10 9.20 9.20 Mining 22.00 5.74 11.80 2.04 24.30 4.61 Expenditures 9.73 15.45 12.80 10.57 9.18 8.95 Current 1.57 13.11 16.03 7.70 8.92 8.78 Capital 19.20 17.76 10.79 12.37 9.43 9.11 External sector (Annual percent change) Export prices 6.8 3.1 -2.6 0.8 -1.3 -0.1 Import prices 3.9 0.5 1.2 0.5 1.1 1.2 Export volume 8.1 6.1 8.0 3.1 17.1 10.5 Import volume 9.4 1.0 3.5 3.6 5.9 11.0 Exchange rate GNF/US$ 986 986 986 986 986 986 (period average) 106 Annex 2 Table 2.5 Central Government Revenue Projections and Expenditures, 1994-2000 (in billions of Guinean francs) 1994 1995 1996 997 1998. "9999 2000 Total revenue 343.3 389.1 419.9 461.2 493.1 560.3 603.6 Mining sector 98.5 120.2 127.1 142.1 145.0 180.2 188.5 Nonmining 244.8 268.9 292.8 319.1 348.1 380.1 415.1 Total expenditures 580.5 637.0 735.4 829.5 917.2 1001.4 1091.0 Current 311.7 . 316.6 358.1 415.5 447.5 487.4 530.2 Wages and Salaries 145.5 154.9 172.6 198.1 216.2 236.3 258.2 Other goods and services 76.5 68.5 89.5 113.9 125.5 139.0 152.5 Subsidies and transfers 37.3 45.6 39.7 44.5 47.7 51.1 54.9 Interest due 52.4 47.6 56.4 55.1 58.1 60.9 64.6 Capital 268.8 320.4 377.3 418.0 469.7 514.0 560.8 FINEX 230.0 265.4 309.3 337.1 381.9 401.6 421.7 BND . 38.8 55.0 68.0 80.9 87.8 112.4 139.2 Nominal GDP 3,315.6 3,641.5 3,965.2 4,321.7 4,714.8 5,148.6 5,622.3 Memorandum items: (In percent of GDP) Total revenue 10.4 10.7 10.6 10.7 10.5 10.9 10.7 Mining sector revenue 3.0 3.3 3.2 3.3 3.1 3.5 3.4 Nonmining sector revenue 7.4 7.4 7.4 7.4 7.4 7.4 7.4 Total expenditures 17.5 17.5 18.6 19.2 19.5 19.5 19.4 Current expenditures 9.4 8.7 9.0 9.6 9.5 9.5 9.4 of which: wages 4.4 4.3 4.4 4.6 4.6 4.6 4.6 Capital expenditures 8.1 8.8 9.5 9.7 8.9 10.0 10.0 ofwhich: BNE) 1.2 1.5 1.7 1.9 1.7 2.2 2.5 107 Annex 2 Table 2.6 Mining-Based Revenue Projections, 1994-2000 (in billions of Guinean francs) 1994 1995 1996 1997 . 1998 1999 2000 Taxes on mining companies 2.2 2.0 0.0 0.0 0.0 0.0 0.0 Other special taxes 1.8 3.7 2.8 2.7 2.7 2.9 3.2 Diamonds 0.6 1.7 1.3 1.7 2.0 2.2 2.5 Precious metals 0.0 1.1 0.0 0.0 0.0 0.0 0.0 Gold sales by BCRG 1.2 0.9 1.5 1.0 0.7 0.7 0.7 Taxes on mining products 88.0 111.6 123.8 138.9 141.9 152.4 155.1 of which: SBK 0.9 0.0 6.5 9.6 9.8 10.1 10.3 Property revenue 0.7 2.9 0.5 0.5 0.5 0.5 0.5 Revenue from new investments and other 5.7 0.0 0.0 0.0 0.0 24.4 29.8 Total 98.4 120.2 127.1 142.1 .145.0 180.2 188.5 Annex 2 Table 2.7 External debt projections (1994-2000) (in millions of US dollars) Net capital GarafirilKaleta External debt Debtl Inflows borrowing (end of period) GDP GDP' 1994 50.8 0 2,788 3,395 82.1 1995 152.4 0 2,940 3,693 79.6 1996 245.4 39 3,224 4,021 80.2 1997 261.6 57 3,543 4,383 80.8 1998 284.5 80 3,908 4,782 81.7 1999 319.4 91 4,318 5,222 82.3 2000 351.0 47 4,716 5,702 82.7 1 Percentage Source: Scenario assumptions. 108 ANNEX 3A Calculation of Required Recurrent Expenditures and Assessment of the R Coefficient Issue 1. The traditional method for calculating recurrent costs generated by capital expenditures' is the R coefficient. As the coefficient is only an approximation, establishing the existence of a recurrent cost gap in Guinea has necessitated two additional steps: (a) complementary estimates of recurrent costs and (b) sensitivity analysis of the R coefficient. Definition of Recurrent Costs 2. Various demarcations of recurrent costs are possible. Economists are traditionally concerned about distinguishing investment from consumption. Accountants are more greatly concerned about whether assets have a life expectancy of more or less than twelve months. These definitions may affect the definition of an investment and recurrent budget, particularly in countries such as Guinea, where donors fund expenditures of a recurrent nature. Definition of the R Coefficien? 3. The Bank uses the standard R coefficient, calculated by P. Heller (IMF) on the basis of investments in Kenya. This coefficient is generally believed to be applicable to Sub-Saharan Africa and yields an approximate estimate of recurrent cost requirements generated by capital expenditures (or Gross Formation of Fixed Capital) by sector. As can be seen from annex 3a, table 3.1, the coefficient can cover a wide range of values for each sector. It is therefore not a precise measure; it is, however, the most accurate instrument for assessing the existence of a recurrent cost gap. 4. The R coefficient measures incremental recurrent expenditures in year t+1 generated by capital outlays in year t. For example, capital expenditures of 100 in year t would generate recurrent costs of 10 in year t+l if R=10%. Recurrent costs of investment measured by the R coefficient comprise salaries and O&M, including overhead costs linked with the management of investments. The coefficient does not capture: (a) general functioning and maintenance of the administration and, (b) one-off current expenditures (i.e., linked to project implementation). 5. Traditionally the R coefficient does not take depreciation of capital into account. Rather, recurrent costs of past investments continue to accumulate over an indefinite period of time. As new projects close, new recurrent costs are generated and the recurrent cost curve rises faster than the capital expenditure curve. In fact, even when capital expenditures decline, recurrent costs continue to rise. In the medium term (10 to 20 years) such a mechanical approach is unrealistic in the long term as recurrent costs decline with the depreciation of capital and technological changes among other factors, so that a typical recurrent cost curve is likely to be bell-shaped in the medium to long term. 1 Capital expenditures are also referred to as Gross Formation of Fixed Capital or GFFC (FBCF in French). 2 For a complete discussion of the R coefficient, see P. Heller, "The Underfinancing of Recurrent Development Costs", Finance and Development, March 1979; and Ibid, "Public Investment in LDCs with Recurrent Cost Constraint: the Kenyan Case", The Quarterly Journal ofEconomics, May 1974. 109 Three Approaches to Recurrent Cost Annex 3a Table 3.1Caclio The R coefficient by sector 6. As reliability on the sole R coefficient Arcultre approach -- although strengthened by sensitivity Fisheries 8% analysis -- would have proven uncertain, two Forestry 4% additional approaches have been used to General agriculture 10% Livestock 14% ascertain the existence of a "recurrent cost gap" Rural development 8%to43% in the composition Guinea's public expenditures: Veterinary services 7% (a) the project approach which consisted of Education Buildines Agriculture colleges 17% calculating recurrent costs of projects based on a Polytechnic schools 17% traditional investment whereby recurrent costs Primary schools 6% to 70% Secondary schools 8% to 72% are generated after the project's life; and (b) Universities 2% to 22% financing of strategies in education and road thospitals 11%to30% maintenance requirements. These calculations General hospitals 18% are used in chapter 3 and are shown in details in Medical auxiliary training 14% annex 7 (education) and annex 4 (roads). Both Nurses college 20% Nutrition Rehabilitation Unit 34% approaches (a) and (b) are compared with the R Rural health centers 27% to 71% coefficient calculation. Ho2using 3% MfanufacturIng. Commerce. 1%.. Roo eCalculation of the Standard R Coefficient for Feeder roads 6% to 14% Guinea Paved roads 5% to 8% Other transports 10% Minine/EneralWater (1) 30% 7. For standard calculation, the R Administration (7) 10% coefficient was weighted according to the share Pand mtmi () 10% of each sector in the 1995 investment budget. Employment f) 10% The weighted R coefficient in Guinea is 14%. This share is comparable to other R coefficients Uranall 10% ruis 10% in Sub-Saharan African countries for which the Source: P. Heller, IMF. R coefficients has been estimated (e.g., Mali: (1) R coefficient from Public Expenditure Review for Mali, 16%). The share of current expenditures in the World Bank, May 1994. investment budget was also calculated on the basis of the 1995 investment budget. About 15% of the 1995 investment budget is devoted to technical assistance, which was considered as recurrent expenditures. The GFFC (gross formation of fixed capital) in the investment budget was 75% in 1995, a higher percentage than in some Sub- Saharan African countries (for example, 55% in Mali) but comparable with others (e.g., 79% in Guinea Bissau). 8. To make the R coefficient more consistent with other recurrent cost approaches, a base year has been chosen (1990) and it was assumed that recurrent costs were met that year (see section below). Actual annual recurrent expenditures are the sum of: (1) salaries and O&M in the current budget; and (2) current expenditures included in the investment budget (25% of investment expenditures). Required annual recurrent expenditures are the sum of (1) current expenditures on a fixed base (1990); (2) 14% of GFFC of year t-1 and previous years; and (3) current expenditures included in investment expenditures (25%). 110 Limitations of the R Coefficient Approach for Guinea Definition of capital expenditures. For calculating capital expenditures in the Guinea PIP, the accounting definition of capital expenditures has beeri used, as the economic distinction is impossible to make without in-depth knowledge of projects. As a result, capital expenditures in the Guinea PIP do not distinguish between one-off capital expenditures (or consumption in the economic sense) and capital expenditures actually generating maintenance requirements. As a result, recurrent cost requirements tend to be overestimated. However, data used for investment expenditures are those provided by the MOP. These data are significantly lower for some years (e.g., 1994) than investment expenditures found in the TOFE (see methodological remark in chapter 1). Use of MOP data tends to underestimate recurrent cost requirements. * Identification of recurrent expenditures versus other types of current expenditures. For estimating actual recurrent expenditures, actual current expenditures (i.e., all salaries and O&M in the current and investment budgets, as shown in table 2.1 of chapter 2) are used as a proxy. However, in addition to the actual recurrent expenditures, current expenditures encompass two other types of current expenditures: (a) current expenditures for general functioning of the administration and (b) one-off current expenditures (i.e., current expenditures strictly limited to the project's life). In this latter case, the overestimation is limited by the fact that current expenditures in the investment budget appear on both sides of the equation (that is, in actual recurrent costs and in recurrent cost requirements). Therefore, they do not affect the absolute value of the gap. However, they do increase the gap/requirements ratio. Based on existing budget figures, it is impossible to distinguish between these different uses of current expenditures as the budget format does not comply well to such disaggregation. This means that actual recurrent expenditures are likely to be overestimated, hence the gap is likely to be larger than shown in the tables. * OBK/SBK expenditures. Actual O&M expenditures in the current budget include on average 20% of O&M spent on OBK/SBK, the Government-owned mining company. However, mining projects on OBK/SBK do not appear in the PIP, and thus inclusion of OBK/SBK O&M expenditures tends to overestimate actual recurrent expenditures. * Disaggregation between current and capital expenditures is available only for the 1995 budget. Disaggregation of actual expenditures would have been more accurate, largely because implementation rates of the investment budget vary significantly from sector to sector. Similarly, disaggregation over several years would have yielded a more meaningful estimate of the share of GFFC in investment expenditures. This imprecision can affect recurrent cost requirements either way. 111 The PER Approach Versus the Project Approach 9. As it would be impossible to determine the life span and timing for the recurrent cost generation for every project in the PIP since 1987, the PER method assumes recurrent cost generation the year following any capital outlay (see para. I above). For some types of projects, such calculation is fairly accurate as recurrent costs are indeed generated during the project's life (e.g., social sectors -- basic education and equity --, agriculture, institution-building). 10. However, for some other projects (e.g., infrastructure or other economic services), recurrent costs are generated one or several years after the project life. For these types of projects, the PER method of calculation distorts the timing of recurrent cost generation during the project life, even if in the medium term (after five years or more) the two methods yield close results. This is illustrated in annex 3a, table 3.2 with a project of US$400 million with an R coefficient of 10% and recurrent costs starting the second year after the end of the project life and increasing progressively (a typical road project profile). Annex 3a Table 3.2 Recurrent cost estimates: the PER vs. a road project years___ Capitalikdtures * Actual recurrent. cots~ (a) :PER estimate (b} Differenc {bH) progect life year I 100 0 0 0 year 2 100 0 10 10 year 3 100 0 10+ 10=20 20 year 4 100 0 10 + 20 = 30 30 After project year 5 0 0 10 + 30 = 40 40 year 6 0 10 40 0 year 7 0 30 40 0 year 8 0 50 40 0 year 9 0 70 40 0 TOTAL 400 160 260 100 Note: Recurrent costs t = 0.10. Capital expenditures t-1. 11. As can be seen from the table, recurrent cost estimates vary significantly from one method to the next. To bring the results of the two methods closer, several assumptions were made for each method. In the PER method, a base year (1990) was selected during which it was assumed that recurrent costs of all investments were financed. The choice of 1990 as the base year is justified as it corresponds to the peak in Government revenue (in real terms), just before the fall in bauxite prices. This assumption also implies that recurrent costs started to be generated only in 1991. 12. In the second method, it was assumed that all projects have a four-year life time and generate recurrent costs the second year after the end of the project life. In that case, new recurrent costs are generated every four years (instead of every year as in the PER method) and they are evenly spread among the years following the project life. In addition, since recording of public investments in Guinea started in 1987, recurrent costs of all investments since 1987 are taken into account. It was also assumed that capital depreciation would lower recurrent cost requirements after ten years. Therefore, in 1999 recurrent costs for 1987 and 1988 capital expenditures are discounted; and in 2000, those for capital outlays in 1989 are further discounted. 112 13. The comparison between the two methods is illustrated in annex 3a, table 3.3. It shows that overall the two methods yield comparable results (on an annual average, PER estimates are about GNF 105 million lower than those of the second method). However, the methods show significant discrepancies during most years, notably during the reference y6Ars 1994 and 2000. Annex 3a Table 3.3 Two methods of recurrent cost calculation .1987 *988 198 4990 t994 1891 1 99 98 84 1994 1994 9 1998 1909 200 Capital exp. 33,052 131.377 161,790 120,455 121,631 108,605 114,461 113,001 118,020 135,649 143,475 155,020 163,116 171,123 rec. cost actual 0 0 0 0 45,671 45,671 45,671 45,671 110,792 110,792 110.792 110,792 136,542 125,124 rec. cost PER 0 0 0 0 17,028 32,233 48.257 64,078 73,280 89.803 108,794 128,880 150,583 173,419 difference actual - PER 0 0 0 28,642 13,438 -2,587 -18,407 37.512 20.989 1,998 -18,089 -14,042 -48,295 avge PER undereslimation 105 The Third Approach: Recurrent Expenditures in the Education Strategy and Road Maintenance Requirements 14. The purpose of this approach is to test the validity of selected R coefficient values. In chapter 3, estimates have been made of the recurrent expenditures involved in financing the strategies for primary education as well as the cost of maintaining the total road network. This section compares first the investment required for financing the primary education strategy in 1999 with incremental recurrent costs generated in 2000 by this investment (see annex 7). In roads, investments made in the context of the Transport Sector Project are compared with the maintenance requirements calculated in annex 4. Education 15. As discussed in chapter 3, calculations show that if the Government wants to finance its primary education sector strategy, capital expenditures in 1999 in primary education will need to approximate GNF 16 billion in real terms (see annex 7). The same calculation shows that additional recurrent expenditures in 2000 due to the previous year's investment would amount to about GNF 7 billion. Comparing the ratio between capital and recurrent expenditures (7/16=44%) with R coefficients shown in annex 3a, table 3.1 above shows the compatibility of the two calculation methods, as the R coefficient for primary school construction programs above falls between 6% and 70% (with a median of about 35%). The wide range of the R coefficient for primary schools does, however, minimize the reliability of such comparison. Roads 16. The R coefficient for the road sector can also be ascertained on the basis of the Transport Sector Project between 1988 and 1993. This project rehabilitated 430 kilometers of roads, an investment of about GNF 31 billion (of which it is assumed 90% was spent on capital expenditures). This percentage could be even higher based on the capital expenditure share in the road sector in the 1995 PIP). Based on calculations for road maintenance requirements used in chapter 3 (see annex 4), average annual periodical and routine maintenance for paved roads amount to GNF 6 million per kilometer. As a result, maintenance of 430 kilometers of newly paved roads would amount to about 113 GNF 2.6 billion, which is about 9% of capital expenditures. This percentage is consistent with the 8% figure in annex 3a, table 3.1 above. However, it should be noted that rehabilitation investments in the road sector usually include some periodic maintenance (as the technical distinction between these two types of expenditures can be difficult to make). Therefore, the R coefficient in the road sector could well be lowered by a margin to account for the inclusion of recurrent costs in investment expenditures. Sensitivity Analysis 17. To determine the existence of a recurrent cost gap in spite of the R coefficient's imprecision, three different sensitivity analyses have been conducted based on the following variations: * variation of the R coefficient the due to cost recovery programs in health, agriculture (rural roads), water, and primary education. This analysis has consisted of taking the lowest value of the R coefficient for these sectors, as shown in annex 3a, table 3.1 above. In addition, capital expenditures in agriculture were lowered to take into account projects which are not supposed to generate recurrent costs for the Government.3 In the energy sector, it was assumed that recurrent costs would be entirely borne by the electrical company. * variation in the nature of technical assistance, resulting in increasing the share of capital expenditures in investment from 75% to 82%; * variation of the sector weight in the PIP, resulting in changing the overall R coefficient. This analysis has been done for 1990 and 2000. Recurrent Cost Gap Based on the Lowest Values in the R Coefficient 18. In this analysis, the following R coefficients have been applied to these sectors: agriculture: 8%; urban water: 0%; energy: 0%; education: 6% (this coefficient assumes almost complete cost recovery in primary education, which is beyond .Guinea's ambitions); health: 17%; and public works (roads): 5%. In addition, total capital expenditures have been lowered by 5%, corresponding to the 40% of the Agriculture PIP made up of projects assumed not to generate any recurrent costs. The 1995 sector weighted R coefficient would be 7% (down from 14% in the standard PER case). In agriculture, the nature of projects differs in that three categories of projects are designed not to generate recurrent expenditures in the Government budget: (1) productive projects whose aim is ultimately to generate profits; (2) lending to public enterprises; and (3) projects to create and/or strengthen cooperatives, decentralized entities, or professional organizations. In 1995, these three types of projects captured almost 40% of the agriculture investment budget. But it is unclear whether these types of projects do not or will not generate recurrent costs. In the four projects in the productive sector (category 1), it is too soon to have evidence of their self-sufficiency since they are still in their early implementation. The one exception is project 1148, which seems to have reached maturity and to be generating profits. In category 2, the budget may not be financing recurrent costs directly, but there is evidence that the Government provides indirect subsidies to public enterprises (for example, Air Guinde) and pays for employees' pensions; hence, these costs should be captured. Category 3 encompasses some projects that contain an element that generates recurrent costs (in particular, EDF projects are integrated and encompass infrastructure components). It is also premature to be confident that these projects are truly autonomous, since they usually have a very long time horizon (for example, Projet de D4veloppement Rural in Guinde Maritime), because one (an EDF project) has already proved to be a failure. 114 19. As can be seen from annex 3a, table 3.4, these minimal assumptions on the R coefficient still show a recurrent cost gap of 25% and 27% of total requirements in 1994 and 1995 (against 37% and 39% in the standard PER calculation tables). Annex 3a Table 3.4 R coefficient: 7% Millions of Constant 1989 GNF 1990 1991 1992 1993 1994 1995 PIP 162 2 144.8 1526 i0' 8' 6 i5'4 less 5% for ag. investments 154.1 137.6 145.0 143.1 83.3 149.5 75.00% capital expenditures 115.5 103.2 108.7 107.4 62.4 112.1 7% + 7% of capital exp t-1 8.1 15.3 22.9 30.4 34.8 Current expenditure base (1990) 138.4 138.4 138.4 138.4 138.4 138.4 current expenditure actual 138.4 139.4 133.0 123.9 119.6 113.5 recurrent (now) from PIP 46.6 41.6 43.9 43.3 25.2 45.2 recurrent expenditure requirements 185.1 188.2 197.6 204.7 194.1 218.5 recurrent actual 185.1 181.0 176.9 167.2 144.8 158.8 RECURRENT EXPENDITURE GAP 0.0 7.1 20.7 37.5 49.3 59.7 GAP as % of requirements 0% 4% 10% 18% 25% 27% RATIO CAPITAL/RECURRENT 62% 57% 61% 64% 43% 71% Variation in the Nature of Technical Assistance 20. In the standard PER calculation of the R coefficient, it has been assumed that all technical assistance (15% of PIP in 1995) is of a recurrent nature. The sensitivity analysis will assume than that only half the technical assistance is of a recurrent nature. In that context, the share of capital expenditures in the PIP would increase from 75% to 82%. Annex 3a, table 3.5 shows that under this analysis the recurrent cost gap increases significantly, from 37% of requirements in 1994 in the standard PER case to 40% in 1994. Annex 3a Table 3.5 Increase in the share of capital expenditures in the PIP (R coefficient: 14%) 1990 199. 1992 1993 1994 1995 1996 PIP 162.2 144.8 152.6 150.7 87.6 157.4 180.9 0.82 capital expenditures (=82% PIP) 133.0 118.7 125.1 123.5 71.9 129.0 148.3 0.14 + 14% of capital exp t-1 0.0 18.6 35.2 52.8 70.1 80.1 98.2 Current expenditure base (1990) 138.4 138.4 138.4 138.4 138.4 138.4 138.4 current expenditure actual 138.4 139.4 133.0 123.9 119.6 113.5 178.2 recurrent (now) from PIP 29.2 26.1 27.5 27.1 15.8 28.3 32.6 recurrent expenditure requirements 167.6 183.1 201.2 218.3 224.3 246.9 269.2 recurrent actual 167.6 165.5 160.5 151.0 135.4 141.8 210.8 RECURRENT EXPENDITURE GAP . 0.0 -17.6 40.6 67.3 88.9 105.0 58.4 GAP as % of requirements 0% 10% 20% 31% 40% 43% 22% RATIO CAPITAL/RECURRENT 79% 72% 78% 82% 53% 91% 70% 115 Variation in the Sector Weight of the R Coefficient 21. The purpose of this sensitivity analysis is to test the validity of applying the same R coefficient throughout a period of time while sector weights in the PIP change. In the case of Guinea, sector shifts in investment expenditures between 1988 and 1994 have been marked by significant increases in expenditures in agriculture, education, water and energy; and similar decreases in mining and transport. Annex 1, table 1.13 shows shifts in investment expenditures between 1988 and 1994. 22. This sensitivity analysis shows that the R coefficient in 1990 is 13%, not a significant difference from 1995 (14%). The composition of the investment budget in 1990 and 1995 is shown below. Projections for year 2000 have been prepared on the assumption that the Government will pursue a policy of rationalization of public expenditures. In that case, the R coefficient would increase and projections for the recurrent cost gap used in annex 3 (and chapter 2) based on a 14% R coefficient would be underestimating the recurrent cost gap in year 2000. 23. Specifically, projections for sector weight and R coefficients in the PIP in year 2000 were based on the following assumptions: * Total expenditures and investment expenditures would increase as projected in the PFP (see annex 3, table 3.1 below for PFP projections). Capital expenditures would amount to 75% of investment expenditures. * Health and education strategies would be financed. Investment in health would amount to GNF 21 billion, or 10% of the investment budget (see annex 5), with heavy focus on primary health care. As a result, the R coefficient in health would increase and it has been assumed to increase to 40% (see ranges in annex 3a, table 3.1 above). This figure constitutes a compromise between cost recovery in primary health care (tending to lower the R coefficient) and the Government's efforts in improving provision of drugs and other O&M in the sector (see chapter 3, policy options for the health sector). In education, the primary education strategy would be financed (GNF 16 billion) while investments in VET and higher education would remain at their 1992 levels. Total investment in education would amount to GNF 18 billion or 8% of total investment. As in health, higher emphasis on primary school building would increase the R coefficient to an assumed 50%. * Cost recovery in the water sector would remain uneven, forcing the Government to assume part of the recurrent charges. The R coefficient for the sector has been lowered by 1/3. * Financing of the Garafiri/Kaleta project would amount bring energy investment to 12% of total investment. It is also assumed that ENELGUI will assume all recurrent charges of the projects. However, as ENELGUI is likely to remain in a precarious financial situation, it would need to receive transfers for the Government. These are counted as recurrent costs for the Government (the R coefficient for energy has been lowered by 2/3). * Agriculture investments would be reduced from 29% to 21% of the PIP as the Government pulls out of programs more adapted to private sector involvement. The rural road strategy would be financed, and a share of the maintenance costs would be picked up by rural 116 communities. As a result, the R coefficient in the agriculture sector would diminish to 6% (from 9%). * Investments in the road sector would be maintained at their previous level (about 20% of the PIP). To account for the presence of periodic maintenance in capital expenditures, the R coefficient has been lowered to 5%. * The mining and telecoms sectors would be privatized. * Other sectors' shares would remain roughly unchanged. 24. These projections show that the R coefficient would increase to 17%, mainly due to the increase in social sector investment. As a result, the Government will need to examine carefully the recurrent cost implications of shifts in investment expenditures towards social sectors. This conclusion also has implications on cost recovery policies. If taken at face value, an increase in recurrent costs induced by higher investments in social sectors would seem to advocate further cost recovery schemes. However, the existing constraints on cost recovery are not well known and should be carefully examined, notably in the context of the private sector's ability, and particularly that of the poor, to finance basic services. Examples in some Sub-Saharan African countries (e.g., Malawi) shed a disturbing light on the benefits of cost recovery financing, notably in education. Annex 3a Table 3.6 Variations in the sector weight of the R coefficient Sector weight in the PIP Sector 19C90 1995 2011:0 Agriculture 29% 32% 21% Mines/Ind/Energy 16% 18% 21% Urban/Transport/Public 35% 28% 33% works/telecom. Social 14% 15% 19% Administration 6% 6% 6% R coefficient 13% 14% 17% 117 Annex 3 Table 3.1 Recurrent costs generated by capital expenditures PFP CASE: 6.9% growth in total expenditures and Investment expenditures projected In PFP (1996-1998) and by IMF (1999-2000) GDP deflator 192.4 202.0 210.1 218.5 227.2 236.3 245.8 255.6 265.8 Millions of Constant 1989 GNF 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 PIP 162.2 144.8 152.6 150.7 87.6 157.4 180.9 191.3 206.7 217.5 228.2 245.9 245.9 75% capital expenditures 121.6 108.6 114.5 113.0 65.7 118.0 135.6 143.5 155.0 163.1 171.1 184.4 184.4 coef K 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 growth of line below 89.3% 49.7% 32.8% 14.4% 22.5% 21.1% 18.5% 16.8% 15.2% 13.8% 13.1% '+ 14% of capital exp t-1 0.0 17.0 32.2 48.3 64.1 73.3 89.8 108.8 128.9 150.6 173.4 197.4 223.2 Current expenditure base (1990) 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 current expenditure actual 138.4 139.4 133.0 123.9 119.6 113.5 178.2 198.2 206.9 216.8 226.9 240.5 274.1 recurrent (now) from PIP 40.5 36.2 38.2 37.7 21.9 39.3 45.2 47.8 51.7 54.4 57.0 61.5 61.5 recurrent expenditure requirements 179.0 191.7 208.8 224.4 224.4 251.1 273.5 295.1 319.0 343.4 368.9 397.3 423.1 recurrent actual 179.0 175.6 171.2 161.5 141.5 152.9 223.4 246.0 258.6 271.2 283.9 302.0 335.6 RECURRENT EXPENDITURE GAP 0.0 16.1 37.6 62.8 82.9 98.2 50.0 49.0 60.4 72.2 85.0 95.3 87.5 GAP as % of requirements 0% 8% 18% 28% 37% 39% 18% 17% 19% 21% 23% 24% 21% RATIO CAPITALIRECURRENT 68% 62% 67% 70% 46% 77% 61% 58% 60% 60% 60% 61% 55% 118 Annex 3 Table 3.2 Recurrent costs generated by capital expenditures Page 1 of 2 CASE IA: 6.9% growth in total expenditures and 2% annual decline in Investment expenditures in real terms until 1998 Millions of Constant 1989 GNF 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 PIP 162.2 144.8 152.6 150.7 87.6 157.4 154.2 151.1 148.1 148.1 158.3 169.3 180.9 75% capital expenditures 121.6 108.6 114.5 113.0 65.7 118.0 115.7 113.3 111.1 111.1 118.7 126.9 135.7 coef K 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 + 14% of capital exp t-1 0.0 17.0 32.2 48.3 64.1 73.3 89.8 106.0 121.9 137.4 153.0 169.6 187.4 Current expenditure base (1990) 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 current expenditure actual 138.4 139.4 133.0 123.9 119.6 113.5 204.9 238.4 265.5 286.2 296.7 317.2 339.1 recurrent (now) from PIP 40.5 36.2 38.2 37.7 21.9 39.3 38.6 37.8 37.0 37.0 39.6 42.3 45.2 recurrent expenditure requirements 179.0 191.7 208.8. 224.4 224.4 251.1 266.8 282.2 297.3 312.9 331.0 350.3 371.0 recurrent actual 179.0 175.6 171.2 161.5 141.5 152.9 243.4 276.2 302.5 323.2 336.3 359.5 384.3 RECURRENT EXPENDITURE GAP 0.0 16.1 37.6 62.8 82.9 98.2 23.4 6.1 -5.2 -10.3 -5.3 -9.2 -13.3 GAP as % of requirements 0% 8% 18% 28% 37% 39% 9% 2% -2% -3% -2% -3% -4% RATIO CAPITAURECURRENT 68% 62% 67% 70% 46% 77% 48% 41% 37% 34% 35% 35% 35% 119 Annex 3 Table 3.2 Recurrent costs generated by capital expenditures Page 2 of 2 CASE IB: 6.9% growth in total expenditures and 5% annual decline in investment expenditures in real terms until 1997 Millions of Constant 1989 GNF 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 PIP 162.2 144.8 152.6 150.7 87.6 157.4 149.5 142.0 151.8 151.8 162.3 173.5 185.5 75% capital expenditures 121.6 108.6 114.5 113.0 65.7 118.0 112.1 106.5 113.9 113.9 121.7 130.1 139.1 coef K 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 growth of line below 89.3% 49.7% 32.8% 14.4% 22.5% 17.5% 14.1% 13.2% 11.7% 11.2% 10.8% '+ 14% of capital exp t-1 0.0 17.0 32.2 48.3 64.1 73.3 89.8 105.5 120.4 136.4 152.3 169.3 187.6 Current expenditure base (1990) 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 current expenditure actual 138.4 139.4 133.0 123.9 119.6 113.5 209.6 247.5 261.8 282.5 292.8 313.0 334.6 recurrent (now) from PIP 40.5 36.2 38.2 37.7 21.9 39.3 37.4 35.5 38.0 38.0 40.6 43.4 46.4 recurrent expenditure requirements 179.0 191.7 208.8 224.4 224.4 251.1 265.6 279.4 296.8 312.7 331.3 351.1 372.4 recurrent actual 179.0 175.6 171.2 161.5 141.5 152.9 246.9 283.0 299.7 320.4 333.3 356.3 380.9 RECURRENT EXPENDITURE GAP 0.0 16.1 37.6 62.8 82.9 98.2 18.7 -3.5 -2.9 -7.7 -2.0 -5.2 -8.6 GAP as % of requirements 0% 8% 18% 28% 37% 39% 7% -1% -1% -2% -1% -1% -2% RATIO CAPITALIRECURRENT 68% 62% 67% 70% 46% 77% 45% 38% 38% 36% 37% 37% 37% 120 Annex 3 Table 3.3 Recurrent costs generated by capital expenditures CASE 2A: 2.3% growth in total expenditures and 10% annual decline in investment expenditures in real terms until 2000 Millions of Constant 1989 GNF 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 PIP 162.2 144.8 152.6 150.7 87.6 157.4 141.6 127.5 114.7 103.2 92.9 83.6 75.3 75% capital expenditures 121.6 108.6 114.5 113.0 65.7 118.0 106.2 95.6 86.0 77.4 69.7 62.7 56.4 coef K 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 growth of line below 89.3% 49.7% 32.8% 14.4% 22.5% 16.6% 12.8% 10.2% 8.3% 6.9% 5.8% 1+ 14% of capital exp t-1 0.0 17.0 32.2 48.3 64.1 73.3 89.8 104.7 118.1 130.1 140.9 150.7 159.5 Current expenditure base (1990) 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 138.4 current expenditure actual 138.4 139.4 133.0 123.9 119.6 113.5 135.5 156.0 175.3 193.4 210.6 226.8 242.4 recurrent (now) from PIP 40.5 36.2 38.2 37.7 21.9 39.3 35.4 31.9 28.7 25.8 23.2 20.9 18.8 recurrent expenditure requirements 179.0 191.7 208.8 224.4 224.4 251.1 263.7 275.0 286.2 294.4 302.6 310.0 316.7 recurrent actual 179.0 175.6 171.2 161.5 141.5 152.9 170.9 187.9 204.0 219.2 233.8 247.8 261.2 RECURRENT EXPENDITURE GAP 0.0 16.1 37.6 62.8 82.9 98.2 92.8 87.1 81.2 75.1 68.8 62.3 55.6 GAP as % of requirements 0% 8% 18% 28% 37% 39% 35% 32% 28% 26% 23% 20% 18% RATIO CAPITAL/RECURRENT 68% 62% 67% 70% 46% 77% 62% 51% 42% 35% 30% 25% 22% 121 Annex 4 Table 4.1 Financing requirements for periodic and routine road maintenance 1995-2000 (constant GNF millions) PERIODIC MAINTENANCE ROUTINE MAINTENANCE RURAL ROADS TOTAL Dirt Roads Paved Roads Regional Roads TOTAL Dirt Paved Regional Cky TOTAL kms periodic maintenance kms entretien periodique kms periodic maint. Roads Roads Roads Sani- ROUTINE TOTAL CAM total CAM total CAM total Peri- tation Routine Periodic MAINTE- GENERAL par km costs per km costs per km costs odic Mainte- Mainte- NANCE (a) (b) (c) (d) (e) (f) Maint. (i) (j) (k) (1) nance kms nance 1 1994 5,470 0.75 4,103 1,380 5.0 6,900 7,000 0.38 2,625 13,628 1,915 1,104 2,450 0.2 5,469 493 194 5,663 19,290 1995 5,364 0.75 4,023 1,486 5.0 6,900 7,000 0.38 2,625 13,548 1,877 1,189 2,450 0.2 5,516 1,281 619 6,135 19,683 1996 5,258 0.75 3,944 1,592 5.0 7,430 7,000 0.38 2,625 13,999 1,840 1,274 2,450 0.2 5,564 1,769 804 6,368 20,367 1997 5,258 0.75 3,944 1,592 5.0 7,960 7,000 0.38 2,625 14,529 1,840 1,274 2,450 0.2 5,564 2,232 1,028 6,592 21,121 1998 4,978 0.75 3,734 1,872 5.0 7,960 6,007 0.38 2,253 13,946 1,742 1,498 2,102 0.2 5,343 3,225 1,367 6,710 20,656 1999 4,753 0.75 3,565 2,097 5.0 9,360 4,535 0.38 1,701 14,625 1,664 1,678 1,587 0.2 4,929 4,697 1,858 6,787 21,412 2000 4,528 0.75 3,396 2,322 5.0 10,485 3,145 0.38 1,179 15,060 2,038 1,858 1,101 0.2 4,996 6,087 2,490 7,486 22,547 ROUTINE ROAD MAINTENANCE INVESTMENT BUDGET (n) Financ- Rural Road Fund Budget Disburse. Periodic Rehabi- olw Total disbursed % ing Budget Disburse- deficit deficit maint. litation Sanitation Invest. disbursed need ment (0) Cky(p) 1994 5,469 2,500 1,200 -2,969 -4,269 13,821 71,474 7,359 85,295 47,291 55% 1995 5,516 4,500 -1,016 122 A. PERIODIC MAINTENANCE Dirt Roads (a) Numbered changes according to dirt roads which become paved roads. Calculation and hypothesis on paved kms of roads. 1) 1990-1996: Mamou-Kankan (400kms), linear progression. 2) 1990-1995: Bubreka-Boffa-Kolaboui (200 kms), linear progression. 3) 1994-1995: Serdou-N'Zerekore (100 kms), included in '95 estimates. 4) 1998: Coyah-Pamalop: (9 kms); 1998: Kissidougou-Tokonou (280 kms). 5) 1999:. 1/2 troncon Tokonou-Kankan (450 km/2). 6) 2000: 1/2 troncon Tokonou-Kankan (450 km/2). (b) cost of periodic maint.: GNF 4.5 million/km every 6 yrs. 21 Paved Roads: (c) the network increases as the dirt roads are rehabilitated (see 1/ above). (d) cost of periodic maint.: GNF 75 million every 15 yrs (resurfacing). 31 Regional Roads: (e) the network for which maintenance responsibility is unclear (MAEF or MTP). The size of the network remains at 700 km for the whole period. However, a portion will be rehabilitated by the rehab. program of the DNGR (MAEF). The rehabilitated kms appear as of 1998 in the column "Rural Roads". (f) cost of periodic maintenance: approx. GNF 2.5 million every 6 yrs. 41 Rural Roads (g) Number of kms developed through the rural roads rehab. program B. ROUTINE MAINTENANCE 11 Dirt Roads (i) GNF 0.35 million/km periodic maintenance. 21 Paved Roads (j) FNG 0.35 million/km cantonnage et 0.45 million/km patching 31 Regional Roads (k) GNF 0.15 million/km routine maintenance and GNF 0.2 million/km re-routing. 41 Sanitation - CKY (1) Rehabilitation of primary roads in Conakry: division of responsibility between MTP and MAT is unclear. 51 Rural Roads (m) GNF 0.35 million/km periodic maintenance. INVESTMENT BUDGET (n) the budget investment, net of studies. (o) number of the projects which are being reconstructed/rehab of roads and bridges: 3218;3224;32253228; 3249;3251;3253;3257;3261;3263; 3262;3265;3266 (p) Rehabilitation of primary roads in Conakry: the division of responsibility between MTP and MAT is unclear. 123 Annex 5 Table 5.1: Costing of 7 priority programs of health strategy by year 2000 Budget of the MOH In 1994 Coverage and targeted population for programs Frleed costs of programs In 2000 Program recurrent Total Invest as%of as%of Targeted Indicators Population Coverage Invest recurrent total budget Recurr death Population 1992 en 2000 1992 2000 cost per Invest recurrent budget (estimation) (estimation) capita (1992) cost 27,396 (estimation) in 2000 Maternal Health 119,083 1,786,091 6.1% 1.0% 8.1% Mothers 3 visits/ 2 doses anti-tetanus 556,500 699,501 39% 90% 7,681 549 4,835,469 345,423 Family Planning and new-boms delivery with mid-wifeldoctor 32% 50% 0 0 0 0 Women, age 15-49 use of contraceptives 1,525,000 1,916,871 3% 5% 0 0 0 0 PEV+ 0 0 9.7% Children, age 0-4 5 visits and doses 1,169,000 1,469,391 55% 80% 4,289 10,008 5,042,193 11,765,118 Malaria 55,521 101,021 0.2% 0.5% 11.4% Children. age 04 anti-fever treatment 1,169,000 1,469,391 15% 50% 259 317 190,640 232,626 Diarrhea 1,009,480 1,076,795 0.2% 8.5% 11.3% Children, age 0-5 Use of ORS by age 5 1,169,000 1,469,391 35% 80% 165 2,467 193,400 2,900,298 STDs/AIDS 68,771 199,911 0.5% 0.6% 0.3% Age> 14 et <60 Use of condoms 2,986,000 3,753,296 3% 5% 1,464 768 274,731 144,071 Tuberculosis 110,770 285,370 0.6% 0.9% 6.5% Infected population Short-lerm treatment 9,200 11,564 10% 25% 189,783 120,402 548,665 348,085 Respiratory Diseases 26,000 101,877 0.3% 0.2% 12.5% Children,age 0-5 Mortality rate (survival) 1,169,000 1,469,391 82% 88% 79 27 102,088 34,981 Nutrition/food 23,725 246,373 0.8% 0.2% 8.5% Age 5-15 Non-deficiency of iodine 1,684,000 2,116,728 80% 95% 165 18 332,334 35,413 Total 1,413,350 3797438 8.7% 11.8% 68.3% 196,204 134,007 11519520 15806015 124 Annex 5 Table 5.2: Expenditures by level of care (in GNF 000s) Constant expenditure per capita Increase in health center Increase in budget Total budget activity to 100% disbursment Expenditure Expenditure Expenditure % of active Expenditures Disbursments Budget Budget with in 1994 per capita in 2000+ health centers 100% active actual expected in 2000 programs++ pop=6.4 M pop=8.OM (carte sanitaire) health centers in 1994 In 2000 Primary (PSICS) Recurrent 3,532,000 0.54 0 383 out of 723 0 100.0% 100.0% 0 15,806,015 Investment 2,867,000 0.44 0 53.0% 0 36.0% 60.0% 0 11,519,520 Secondary (HP) Recurrent 4,996,000 0.77 0 25 out of 33 0 100.0% 100.0% 0 0 Investment 1,676,000 0.26 0 75.8% 0 36.0% 60.0% 0 0 Tertiary (CHU) Recurrent 2,323,000 0.36 0 4 out of 4 0 100.0% 100.0% 0 0 Investment 10,055,000 1.55 0 100.0% 0 36.0% 60.0% 0 0 Administration Recurrent 2,183,000 0.34 0 n.a. 0 100.0% 100.0% 0 0 Investment 2,089,000 0.32 0 100.0% 0 36.0% 60.0% 0 0 Total Recurrent 13,034,000 2.01 0 n.a. 0 100.0% 100.0% Q 15,806015 Investment 16,687,000 2.57 0 0 36.0% 60.0% 0 11,519,520 Total 29,721,000 4.57 0 0 0 27,325,535 Total expenditures per capita in 1994 in GNF 4,644 in USD 4.89 Recurrent expenditures per capita in 1994 in GNF 2,037 in USD 2.14 Exch. rate GNF/USD 950 Population (2.9%/yr) 6.4 Note 1: No increase is assumed for expenditure at each level Note 2: The costs for primary level health care will only include the above activities and costs per level. 125 ANNEX 6 THE HEALTH SECTOR A. POLICY AND ORGANIZATION OF THE HEALTH SECTOR Government's Policy 1. Although sectoral policy is based on sound principles and addresses relevant priorities, it sets arbitrary, vague operational objectives for resource allocation and planning purposes. This is the same for the Three-Year Rolling Plan and its 16 programs. Nonetheless, the Government has recognized that it must improve its budgetary and planning processes; in fact, it has decided to make annual, detailed budget analyses and projections, to change the format of the. Three-Year Rolling Plan, and to issue annual, detailed operational plans. 2. In the past few years, the Government has undertaken a systematic analysis of the sector's existing infrastructure (Carte Sanitaire), obtained donor support, and begun addressing needs for new construction, infrastructure rehabilitation, and equipment, with emphasis on health centers and first-referral hospitals. It has also addressed human resource needs by adopting a staffing and redeployment plan and a manpower development plan. In these policy documents, the Government reiterates its commitment to improving staffing at PHC delivery facilities (by offering incentives to personnel to redeploy to the country's underserved districts), and strengthening staff skills through a comprehensive training program. The Public Sector 3. The public health delivery system comprises a network of 700 health posts (in villages), 370 health centers (in rural and urban communes), 29 district (prifectures) hospitals, four regional hospitals, and two tertiary/teaching hospitals, located in Conakry. Three other hospitals are operating under the auspices of mining companies. The Ministry of Health is in charge of formulating, planning, and managing sectoral policy; it discharges these functions through its technical directorates for health programs, and its administrative, finance, and planning directorates, but it is also increasingly decentralizing these functions to regional health inspectorates and district health directorates. 4. The sector currently employs 6,334 persons, or 12.2% of the public sector's human resources. Health centers and hospitals are also using an unknown number of contractual employees who are paid from cost-recovery funds, as well as more than 600 nonsalary traditional birth attendants (TBA). The sector consists of 919 medical doctors and dentists, 1,676 nurses and midwifes, and 197 pharmacists providing acceptable health staff to population ratios. The Private Sector 5. The private sector in Guinea comprises an informal sector in villages and small cities that consists of an unknown, but probably important, number of traditional medicine practitioners (TBAs, herborists, and so forth), as well as a relatively small formal sector that consists of 57 medical doctors and dentists, 51 nurses, and 128 pharmacists. 126 6. A study conducted in 1994 by SEATS/JSI estimated that as many as 1.5 to 2 million rural residents use traditional services; the same study quotes the Labor Department's figure of 1,170,000 people who seek services in formal private sector clinics, medical practices, and laboratories. The partnership between the Government and the private sector is currently limited to a few local NGOs (Association Guinjenne de Bien-Etre Familial, the Association for Voluntary Surgical Contraception, and CERAC-SANTE), several international NGOs (the French and Belgium Mddecins du Monde, Mdecins sans fronti6res, and Medicus Mondi), and the Catholic Church. The Government recognizes the importance of the private sector, but, unfortunately, it has not sought to develop it systematically. It should seek to strengthen this partnership by supporting the formal and informal private sector and developing the public sector in a complementary manner to it. Health Budget Allocations Annex 6 Table 6.1 Health sector budget allocations: 1988-1995 (constant GNF 1989 millions) Bod1990 1991 1992 1993 1.994 995 Current budget health 6,253 4,522 6,218 7,541 7,370 6,432 6,296 as % of total current 3% 2% 4% 4% 4% 4% 5% Investment budget health 11,370 13,030 12,847 12,516 11,303 14,138 14,949 as % of total investment 7% 7% 7% 7% 6% 9% 10% BND health 2,354 1,397 1,530 1,213 2,660 3,252 3,258 as%oftotalBND 7% 5% 5% 5% 8% 11% 12% FINEX health 9,016 11,633 11,317 11,303 8,643 10,886 11,691 as % of total FINEX 7% 7% 7% 7% 6% 8% 9% Total Budget health 17,623 17,552 19,065 20,057 18,673 20,570 21,245 as % of total budget 5% 5% 5% 5% 5% 6% 8% as% of GDP 1% 1% 1% 1% 1% 1% 1% Per capita real (US$) 6.6 4.6 4.5 3.8 3.1 3.3 3.2 Per capital nominal (US$) 5.4 5.7 6.5 16.4 5.7 6.3 6.5 Note: Budgetary data provided by the Ministry of Finance and the Ministry of Health are not consistent and at times do not support meaningful analysis. Furthermore, the data cannot be used readily to compare expenditures per level of health care delivery, budgetary allocations among regions, or major health programs. In this presentation, the data provided by the Ministry of Health-which are more detailed-were used for the analysis of the budget implementation and for the breakdown per level of care, while the more aggregated data from the Ministry of Finance were used to illustrate the budget and the budget allocation levels, and for part of the health expenditures analysis. B. EXPENDITURE INCIDENCE ANALYSIS ON THE POOR IN THE HEALTH SECTOR 7. To measure the distribution of public expenditures on health and education among different groups in Guinea, we use an expenditure incidence analysis. The analysis maps the monetary value of the in-kind transfer implied in the public education to individuals, making it possible to trace how Government spending on education or health is distributed among the population and across regions. 8. Expenditure incidence analysis (sometimes referred to as benefit incidence) has become an established methodology since the pathbreaking work on Malaysia by Meerman (1979) and on Colombia by Selowsky (1979). The methodology has enjoyed a resurgence recently, which is reviewed in Van de Walle and Nead (1995). Few applications have been attempted in Africa. The distribution of these subsidies is determined by two broad factors. First, it depends on government spending itself, and how it is allocated within the sector. The lower the spending and the greater the 127 effective cost recovery, the lower the subsidy embodied in the service provided. Second, the distribution will depend on household behavior--on who uses the service that the Government provides. It is only by using the service (by sending a child to a primary school, or visiting the outpatient department at a hospital) that individuals and households can lay a claim to the in-kind transfer that is implicit in the subsidy. 9. Expenditure incidence analysis thus combines two sources of information: data on the government subsidy (estimated as the unit cost of providing the service less any cost recovery back to the Government, as shown in annex 6, table 6.4) that is allocated to the different categories of service (primary schooling, in-patient hospital care, and so on); and information on how these services are used by individuals and households, based usually on household surveys. Households that reported a current enrollment in the 1994 Integrated Household Survey were considered to have benefited from the relevant annual subsidy: in the absence of the subsidy, they would have to pay this amount to finance the schooling of their children. Similar procedures were used for estimating the incidence of health spending. The unit costs of health care (given as the government subsidy per health facility visit) were allocated to households who reported visits to the health facilities distinguished in the analysis (health centers/clinics, hospital in-patients and hospital out-patient). 10. In general, government expenditures will be distributed more equally when the spending is concentrated on services used widely by the population, and especially by poorer groups. If public expenditures are concentrated in primary education, or in primary health facilities (such as health centers-which are widely-used services benefiting poor and non-poor), public expenditures will tend to be distributed more equally. However, if governments spend more on high-cost services that are not generally used by poorer groups (such as university education or in-patient hospital care), the incidence of spending is likely to be more unequal. In sum, the expenditure incidence of public spending depends both on the allocation of public expenditures within the sector, and on the behavior of households. Who Uses Public Health Services? Annex 6 Figure 6.1: Health facility use for illness by quintile: 1994 100% 80% public health centet 6 0% .pu blic hospital C3n o p r a ate nt 20% poorest 2 3 4 wealthiest Source: Integrated Household Survey, 1994. 128 Annex 6 Table 6.2 Rate of Illness and Rate of Seeking Treatment Quintile Share of Population Who Reported Share of Those Who Reported an IllnessfInjury During the lllnessllnjury and Sought Previous 4 Weeks (percent) Treatment (percent) Poorest 24 40 2 26 43 3 28 52 4 27 58 Wealthiest 32 69 Average 27 54 Rural 28 48 Urban 25 66 Male 26 54 Female 28 53 z Annex 6 Table 6.3 Utilization of outpatient public health care facilities for preventive and curative services, by quintile11 Preventive Health Care Visits Ellig Public hospital 17% 0% 1% 42% 31% 23% Health center 43% 68% 68% 54% 34% 48% Total public 59% 68% 69% 96% 66% 71% Private Homeb 41% 32% 31% 1% 8% 17% Private hospital 0% 0% 0% 0% 4% 2% Private clinic 0% 0% 0% 3% 11% 5% Other private 0% 0% 0% 0% 12% 5% Total Private 41% 32% 31% 4% 34% 29% Curative Health Care Visits Public hospital 3% 10% 20% 29% 47% 26% Health center' 34% 42% 41% 33% 28% 34% Total public 31o 52% 61. 62% 71% 6./16 Private Homeb 64% 47% 37% 34% 15% 35% Private hospital 0% 1% 0% 2% 5% 2% Private clinic 0% 0% 1% 1% 4% 2% Total Private 48/ lffi 25% 39% Source: Preliminary results, 1994 Integrated Household Survey (April 1995). 1/ As reported by individuals surveyed in the 1994 Integrated Household Survey, based on a four-week recall period. Only 4% of all health care visits reported were for preventive services. a Health center category includes self-reported visits to health centers. maternities, dispensaries, and "other" public facilities. b The home of the health care provider or patient. In most cases, this is the home of a traditional medicine practitioner/healer. 129 The Equity of Public Health Expenditures: Do They Benefit The Poor? 11. Expenditure incidence analysis identifies which groups in the population receive the public transfer implied in the provision of a public service -- in this case, health. (Annex 8 provides full details on the procedures and data underlying the analysis as it applies specifically to education). The analysis was applied to two broad categories of public services: health centers and hospitals.4 The first category refers to all primary health care facilities, including health centers, maternities, dispensaries, and health posts. The second includes all types of secondary and tertiary-level hospitals. The analysis examined per visit costs for each category in each region (annex 6, table 6.3). Costs for hospitals and inpatient and outpatient care were calculated separately, based on a cost ratio of 6:1, which was estimated in a study by MOH (1995). Outpatient hospital visits cost about one-third more than do visits to a primary health care facility. The implied per capita subsidy from the public health service per visit costs GNF 1321 per person who used the services. These per-visit costs are broken down by expenditure group, region, and gender, thus identifying which groups in Guinea are receiving the implied public transfer. Annex 6 Table 6.4 Recurrent cost per visit for public health care by region: 1994 Canakry Lower Middle Upper Forest All Guinea _____________________Gu__ne__Gut.s G(uinea Guinea Health centers (Primary-level health care) Nonwage (GNF millions) 12 51 86 67 79 295 Total (GNF millions) 1,300 581 493 515 642 3,532 Total number of visits 481,620 967,452 653,592 649,260 1,163,136 3,915,060 Cost per visit (GNF) 2,700 601 755 792 552 902 All hospitals (Secondary/tertiary health care) Wage (GNF millions) 1,849 1,151 869 849 1,128 5,846 Nonwage (GNF millions) 458 276 267 298 174 1,473 Total (GNF millions) 2,307 1,428 1,135 1,146 1,303 7,319 Number of outpatient visits 1,020,024 613,704 375,840 419,352 486,288 2,915,208 Number of inpatient visits 98,880 121,944 44,220 79,908 92,628 437,580 Cost per outpatient visit (GNF)a 1,430 1,061 1,770 1,276 1,250 1,321 Cost per inpatient care episode (GNF)' 8,579 6,368 10,623 7,653 7,500 7,926 Administration Wage (GNF millions) 562 380 413 382 263 2,000 Nonwage (GNF millions) 173 4 3 3 0 183 Total (GNF millions) 736 383 417 384 263 2,183 All recurrent (GNF millions) 4,343 2,393 2,045 2,045 2,208 13,034 Source: MOH (1995), Hameidat (1995), and Integrated Household Survey results, 1994. Each inpatient care episode is assumed to cost six times that of an outpatient visit. This assumption is based on a recent Ministry of Health (1995) report that compares the cost ratios at Monou Prefectural Hospital, 1994. 4 Data limitations prohibited a more disaggregated analysis. 130 Annex 6 Table 6.5 Per capita health public subsidy: 1994 Urban Rural Total Quinnies Male Female Total Male Female Total Male Female Total Share Health Centers Poorest 68 321 205 303 235 267 295 238 265 10% 2 106 140 123 375 368 372 353 352 353 13% 3 383 504 445 473 563 522 452 550 505 19% 4 745 635 690 573 560 566 675 600 636 23% Wealthiest 913 1,013 963 708 1,285 1,019 875 1,070 974 36% Total 736 767 751 421 457 441 534 557 546 Hospitals Poorest 831 299 543 129 10 66 152 19 81 1% 2 887 645 767 365 218 287 407 248 323 6% 3 1,095 1,728 1,417 576 790 692 700 991 856 15% 4 1,406 1,803 1,603 1,518 498 946 1,452 1,188 1,313 23% Wealthiest 2,913 3,285 3,098 4,019 2,220 3,048 3,116 3,061 3,088 55% Total 2,048 2,408 2,226 707 451 569 1,186 1,082 1,131 All Health Poorest 899 620 748 432 244 332 447 257 346 4% 2 992 785 890 740 586 659 761 600 676 8% 3 1,478 2,232 1,862 1,049 1,353 1,214 1,152 1,542 1,361 16% 4 2,151 2,438 2,294 2,091 1,059 1,512 2,126 1,788 1,949 23% Wealthiest 3,825 4,298 4,060 4,727 3,505 4,067 3,991 4,138 4,062 48% Total 2,784 3,175 2,978 1,128 908 1,010 1,720 1,639 1,6771 Annex 6 Table 6.6 Cost ratios per inpatient/outpatient Visit, results from Monou Prefectural Hospital: 1994 _________________________AttCurret 5,260 General medicine (outpatient) 30,820 General medicine (inpatient) 6 Pediatrics (outpatient) 5,233 Pediatrics (inpatient) 26,456 Inpatient cost/outpatient cost 5 Maternity (outpatient) 8,572 Maternity (inpatient) 53,019 Inpatient cost/outpatient cost 6 Surgery (outpatient) 6,560 Surgery (inpatient) 54,308 Inpatient cost/outpatient cost 8 Emergency 11,770 Dental visit 10,531 Average inpatient cost/outpatient cost 6 Source: Ministry of Health, 1995. 131 Methodology for Measuring Targeting Effectiveness of Public Health Expenditures 12. The targeting effectiveness of public health expenditures can be assessed in two ways: in relative terms, by comparing the distribution of public health expenditures with the distribution of total per capita household expenditures; and in absolute terms, by comparing the distribution of the public health expenditures to the distribution of the population (annex 6, figure 6.2). The cumulative share of individuals in the population, ranked by per capita expenditures, is measured on the horizontal axis. The vertical axis measures the cumulative share of public health expenditures and household expenditures. Annex 6 Figure 6.2 Distribution of public health expenditures: 1994 100 . 80al health u alth centers cumulative share of 60 subsidy/ expenditure 40 20 0 20 40 60 80 100 cumulative share of population 13. In relative terms, expenditures on primary health care and all health are pro-poor, in that the poorest quintiles receive a larger share of public health expenditures than their share of total household expenditures. (Annex 6 discusses the methodology used for measuring effectiveness of expenditure targeting.) At the lower end of the distribution, the curve for public expenditures on hospitals is below the curve for total household expenditures, indicating that poorer groups receive fewer public health expenditures than their share of total household expenditures. 132 Cost Recovery and Costs Associated with Travel and Wait Times Annex 6 Table 6.7 Self-reported costs of public and private health care services (per consultation): 1994 Consultation Fee Public Public hospital 2,186 4,062 2,014 729 2,554 Health center 596 447 760 447 819 Private Home visita 1,195 829 1,860 333 2.517 Private hospital 2,926 10,783 2,530 2,946 Transport cost Public Public hospital 712 2,387 560 429 746 Health center 232 310 145 514 180 Private Home visit a 252 467 110 0 709 Private hospital 415 3,143 255 396 Time Spent (in minutes) Public Public hospital 114 108 178 390 116 Health center 150 213 82 550 94 Private Home visit a 81 75 90 55 131 Private hospital 143 806 106 113 Note: Refers to the time spent on travel, waiting, and the consultation itself. Source: Integrated Household Survey, 1994. a Usually refers to a consultation with a traditional health practitioner. Annex 6 Table 6.8 Self-Reported costs of public and private health care services per consultation: 1994 Queiwk Consultation Fee poozrei 2 3 4 rtchest Average Rural Urban Male Female Public Public hospital 971 1,327 1,873 2,554 2,186 4,062 2,014 2,334 2,049 Health Center 364 532 554 819 596 447 760 604 589 Private Home visita 785 650 1,174 2,517 1,195 829 1,860 1,071 1,292 Private hospital 1,500 1,818 2,946 2,926 10,783 2,530 3,008 2,838 Transport cost Public * Public hospital 429 447 798 641 746 712 2,387 560 794 637 Health center 514 164 209 255 180 232 310 145 189 269 Private Home visit a 0 2 64 302 709 252 467 110 230 270 Private hospital 1 1000 0 458 396 415 3,143 255 420 410 Time Spent (in minutes)" Public Public hospital 390 164 114 92 116 114 108 178 116 112 Health center 550 157 113 97 94 150 213 82 148 152 Private Home visit 55 46 76 77 131 81 75 90 74 86 Private hospital 155 135 229 113 143 806 106 167 117 ' Usually refers to a consultation with a traditional health practitioner. b Refers to the time spent on travel, waiting, and the consultation itself Source: Preliminary results of the 1994 Integrated Household Survey (April 1995). 133 ANNEX 7 EXPENDITURES IN EDUCATION Organization of the Sector 1. The educational system of Guinea consists of two broad structures -- pre-university and higher education. The pre-university level, in turn, consists of primary and secondary schooling, as well as vocational education and training (VET). All pre-university structures are directed by the Ministry of Pre-University Education (Ministre de l'enseignement pre-universitaire et de la formation professionnelle, MEPUFP) through eight regional academic inspectorates; in turn, the inspectorates administer 38 prefect-level directorates of education and 210 subprefect-level pedagogic delegates. Higher education in Guinea is managed jointly by the Ministry of Higher Education, Scientific Research, and Culture (Ministre de l'enseignement supirieur, de la recherche scientifique et de la culture, MESRSC), three sectoral ministries, and several institutions -- three institutions of higher education, ten post-secondary technical training schools, and six scientific research centers -- direct all pre-university structures. Enrollment Rates 2. Enrollment rates among girls. Although the primary school enrollment rate among girls increased from 19% to 23% between 1989 and 1993, their attendance has remained at 32% of all enrollments for the past ten years. And in six of Guinea's 38 prefectures, the GER among girls was 10% or less during 1992-93. As private schools at the primary level are still in their infancy, girls comprise 37.8% of their student population, compared with 32% in public schools. The percentage of female teachers (just under 23%) is even lower than the percentage of female students. 3. Urban-rural enrollment rates. The GER among rural students was 27% during 1992-93; the GER among urban students was 76%. During 1993-94, the respective figures were 27% and 78%. In rural areas, only 314 of 5,002 primary school teachers were women in 1992-93. 4. Enrollment rates by region. Conakry and Forest Guinea have the highest rates -- Conakry because it is the main urban center, and Forest Guinea because its long tradition of Christianity emphasizes educational participation (see annex 7, table 7.1). Annex 7 Table 7.1 Gross primary enrolment rates, by region: 1992-93 and 1993-94 (percent) Region G .Total Girs Bys Total Conakry 59 86 73 79 110 94 Lower Guinea 19 47 33 25 44 35 Middle Guinea 15 39 27 26 39 33 Upper Guinea 15 41 28 20 33 27 Forest Region 22 -61 41 32 60 46 Total GER 23 51 37 34 53 44 Note: The sources for the two years are different. Data for 1992-3 come from the MEPUFP; data for 1993-4 come from the preliminary results of the 1994 Integrated Household Survey. 134 Costing of the Government's Strategy for Primary Education by Year 2000: Methodology 5. - To assess the budgetary implications of the project, operators used the following parameters: a 2.8% annual increase in the number of primary school students; a 7% annual increase in primary school GER; the hiring of 600 new teachers between 1995-1999 (except in 1996) and 900 new teachers in 2000 and thereafter; and the construction of 780 classrooms per year between 1995-1999 and 900 in 20C00 and thereafter (an average of 550 new teachers and 800 new classrooms annually between 1995-2000). In the preparation of this PER, we also used these parameters to assess the budgetary implications of the project as well as their effect on an important indicator of educational quality: the ratio of children to teachers and classrooms. We developed three scenarios: the first is based on the project projections; the second is based on a more realistic projection of the growth rate in primary school students; the third is based on the MEPUFP statistics of the effect of the Government's GER projections on indicators of quality.5 As these projections resulted in an agreement between the Bank and the Government of Guinea, they will be used as references in subsequent chapters. Detailed projections appear in the tables below. Results of the Three Scenarios 6. Project projections. A 2.8% increase in the number of school-age children. The first set of projections come from the Equity and School Improvement Project. In year 2000, the Government would essentially meet its 53% GER objective (at 52%). Under this scenario, the pupil-teacher and pupil-classroom ratios would rise, respectively, to 57:1 and 47:1 in year 2000; current budget amounts for the project in 2000 would be GNF 57 billion. The scenario has two important implications. Meeting the projected 100% GER by year 2020 would necessitate increasing both the pupil-teacher and children-classroom ratio to at least 70:1, an unacceptable figure; thus, the number of new teachers and new classrooms would have to be increased by more than 900 annually after year 2000. It is very clear that, failing increased tax revenue and allocation to primary education, significant donor assistance would need to continue to achieve the 2020 objective. Second, as more children go on to secondary schooling, sizable donor assistance will also be required to increase capacity at this time; with an estimated annual increase of 5% in the number of children enrolled in secondary school, this level might face a capacity constraint even before year 2000. 7. A 4.5% increase in the school age cohort. Given the official population growth rate of 2.9% in Guinea, Bank staff have estimated that the school-age cohort will actually grow at an annual rate of 4.5% to 5.0%.6 Assuming the conservative 4.5% rate, the burden on the current budget would amount to GNF 57.27 billion in 2000, close to the budgetary estimate of the project. Yet, without changing the projections for school construction and the number of teachers hired, the pupil-teacher and pupil-classroom ratios would increase to 60 and 52, respectively, in the absence of additional budgetary support. Note that the PER projections do not account for multigrade classes and double shifts, which affect the ratio of children per class and per teacher. In addition, the projections include an increase of 600 in the number of private teachers in 1995, but none thereafter; yet the development of private primary schools would affect the ratios of children per class and per teacher. 6 For details, see Country Economic Memorandum, (CEM), World Bank, October 1995. 135 8. Projections from the Statistical Directorate of the MEPUFP. In the course of discussions on project parameters, the Statistical Directorate of the MEPUFP indicated that the Government could meet its 7% annual increase in primary school GER only if the pupil-teacher and pupil-classroom ratios remained at around 45; ratios above this level would have an adverse impact on GER. The directorate thus estimated that the Government would have to increase the construction of new classrooms to 800 annually, and it would require an average of 978 new teachers annually between 1995 and 2000. Thus, with a 2.8% growth rate in the number of school-age children, current expenditures would increase to GNF 60.4 billion by year 2000; with a growth rate of 4.5%, current expenditures would be GNF 62.5 billion. 136 Project scenario Annex 7 Table 7.2 Costing of Government's Strategy for Primary Education by Year 2000 Primary and lower secondary education: budgetary cost of the objective of reaching 100% of gross enrollment rate by 2022 Page 1 of 4 Estimates based on budgetary projections during preparation of Basic Education and Equity Project financed by IDA (1994) $1= GNF 1,000 1994(14) 1995 1996 1997 1998 1999 2000 2001 2002 Scenaro 1: increaseIan numberof school-age chidren = 2.8%. Program indicators From 1995to 1999: 780 classes/year and 600 teachers/yea From 2000 to 2022* 900 classes/year ad900 teachers/year (a) Primary school Primary school Number of school-age children 1,175,336 1,208,245 1,242,076 1,276,854 1,312,606 1,349,359 1,387,141 1,425,981 1,465,909 % increase in the nuber of school-age children 2.80% 2.80% 2.80% 2.80% 2.80% 2.80% 2.80% 2.80% Gross enrollment rate (1) 40% 42% 44% 46% 48% 50% 52% 54% 55% Number of enrolled children 471,792 506,342 543,423 583,219 625,929 671,767 720,962 763,383 808,301 Increase in number of enrolled children 34,550 37,080 39,796 42,710 45,838 49,195 42,421 44,917 % increase 7% 7% 7% 7% 7% 7% 6% 6% Cumulative increase 34,550 71,631 111,427 154,137 199,975 249,170 291,591 336,509 Total number of public school teachers 9,372 9,972 9,972 10,572 11,172 11,772 12,672 13,572 14,472 Net increase in number of primary school teachers 0 600 0 600 600 600 900 900 900 Total number of public and private school teachers 9,972 10,572 10,572 11,172 11,772 12,372 13,272 14,172 15,072 Number of students per teacher (b) 47 48 51 52 53 54 54 54 54 Infrastructure: number of classes 10600 11380 12160 12940 13720 14500 15400 16300 17200 Number of new classes required 780 780 780 780 780 900 900 900 Number of children per class 45 44 45 45 46 46 47 47 47 (b) Lower secondary cycle Total number of children 84544 90787 97343 104229 111461 119057 127035 135414 144215 Net increase in number of children enrolled 6243 6556 6886 7232 7596 7978 8,379 8,801 % increase (2) 5% 5% 5% 5% 5% 5% 5% Net increase in number of secondary school teachers 0 0 0 0 0 0 0 Number of students per teacher (3) 36 39 42 45 48 51 54 58 137 Project scenario I_ _ I____ _ I__I __IPage 2 of 4 1994 1995 - 1996 1997 1998 1999 2000 2001 2002 In blions of GNF Recurrent costs (calculated by the project and adjusted for the increase in the number of secon dary school teachers) Budgetary base 52.51 52.51 52.51 52.51 52.51 52.51 52.51 52.51 Additional primary level salaries (4 trimesters) (4) (5) 0.24 0.95 1.19 2.15 3.10 4.17 5.59 7.02 Additional secondary level salaries (4 trimesters) (7) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Additional primary level teaching material (8) 0.24 0.26 0.28 0.30 0.32 0.34 0.30 0.31 Additional secondary level teaching material (9) 0.07 0.08 0.08 0.09 0.09 0.10 0.10 0.11 Primary school manuals (10) 0.22 0.22 School maintenance costs (12) 0.23 0.24 Total (13) Current budget 53.07 53.80 54.06 55.04 56.02 57.12 58.95 60.41 Investment for Primary Education and Equity Project 13.62 13.62 13.62 13.62 13.62 15.3 15.08 15.08 Classroom construction 10.92 10.92 10.92 10.92 10.92 12.6 12.6 12.6 Primary level teacher training program (11) 1.6 1.6 1.6 1.6 1.6 1.6 1.6 1.6 Primary school manuals 1.1 1.1 1.1 1.1 1.1 1.1 0.88 0.88 TOTAL 66.69 67.42 67.68 68.66 69.64 72.42 74.03 75.49 Total per enrolled student (in GNF) 131.70 124.07 116.05 109.70 103.67 100.44 96.97 93.39 Recurrent cost per enrolled student 104.80 99.01 92.70 87.94 83.39 79.22 77.22 74.73 138 Project scenario Page 3 of 4 1994 1995 1996 1997 1998 1999 2000 2001 2002 ______________________________Scenario 2: (ncrease In n umber of~I sc 0i-age chlid an=45% _______________________________ I______Fr~in 2000 to 2022: 900 casse/yearen d 900 teachis/year~, Ilss I -- Program indicators (a) Primary school Number of school-age children 1,175,336 1,228,226 1,283,496 1,341,254 1,401,610 1,464,682 1,530,593 1,599,470 1,671,446 % Increase in the nuber of school-age children 4.50% 4.50% 4.50% 4.50% 4.50% 4.50% 4.50% 4.50% 4.50% Gross enrollment rate (1) 40% 42% 44% 46% 48% 50% 52% 54% 55% Number of enrolled children 471,792 514,716 561,544 612,634 668,371 729,180 795,520 856,258 921,634 Increase in number of enrolled children 42,924 46,829 51,089 55,737 60,808 66,341 60,738 65,375 % increase (2) 9% 9% 9% 9% 9% 9% 8% 8% Cumulative increase 42,924 89,752 140,842 196,579 257,388 323,728 384,466 449,842 Total number of public school teachers 9,372 9,972 9,972 10,572 11,172 11,772 12,672 13,572 14,472 Net increase in number of primary school teachers 0 600 0 600 600 600 900 900 900 Total number of public and private school teachers 9,972 10,572 10,572 11,172 11,772 12,372 13,272 14,172 15,072 Number of students per teacher (b) 47.31 48.69 53.12 54.84 56.78 58.94 59.94 60.42 61.15 Infrastructure: number of classes (4) 10600 11380 12160 12940 13720 14500 15400 16300 17200 Number of new classes required 780 780 780 780 780 900 900 900 Number of children per class 45 45 46 47 49 50 52 53 54 (b) Lower secondary cycle Total number of children 84,544 91,308 98,612 106,501 115,021 124,223 134,161 144,894 156,485 Net increase in number of children enrolled (3) 6,764 7,305 7,889 8,520 9,202 9,938 10,733 11,591 % increase 8% 8% 8% 8% 8% 8% 8% 8% Net increase in number of secondary school teachers (6) 0 0 0 0 0 0 0 Number of students per teacher 37 40 43 46 50 54 58 63 139 Project scenario Page 4 of 4 1994 1995 1996 1997 1998 1999 2000 2001 2002 In billions of GNF Recurrent costs Budgetary base 52.51 52.51 52.51 52.51 52.51 52.51 52.51 52.51 Additional primary level salaries (4 trimesters) (5) (6) 0.24 0.95 1.19 2.15 3.10 4.17 4.53 7.04 Additional secondary level salaries (4 trimesters) (7) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Additional primary level teaching material (8) , 0.30 0.33 0.36 0.39 0.43 0.46 0.43 0.46 Additional secondary level teaching material (9) 0.08 0.09 0.09 0.10 0.11 0.12 0.13 0.14 Primary school manuals (10) 0.22 0.22 School maintenance costs (12) 0.23 0.24 Total recurrent costs 53.13 53.88 54.15 55.15 56.15 57.27 58.04 60.60 Investment for Primary Education and Equity Project 13.78 13.78 13.78 13.78 13.78 15.46 15.24 15.24 Classroom construction 10.92 10.92 10.92 10.92 10.92 12.6 12.6 12.6 Primary school manuals 1.1 1.1 1.1 1.1 1.1 1.1 0.88 0.88 Primary level teacher training program (11) 1.76 1.76 1.76 1.76 1.76 1.76 1.76 1.76 TOTAL 66.91 67.66 67.93 68.93 69.93 72.73 73.28 75.84 Total recurrent cost enrolled student (in GNF) 103,222 95,949 88,395 82,512 77,000 71,987 67,788 65,756 (in US$) 108.88 101.21 93.24 87.04 81.22 75.94 71.51 69.36 Recurrent cost per enrolled student (in GNF) 129,994 120,488 110,888 103,130 95,898 91,421 85,586 82,292 (1) 1994 figures given by the MEPUFP. (.b.) in view of the 3-1/2 year training cycle for student-teachers, the objective of 45 students per teacher connot be attained before 1999 if training starts in 1996. (2) it is assumed that the secondary cohort will grow at the same rate as that of the primary (5%). 1 1 1 (3) If there is no new recruitment of secondary-level teachers, the number of students per class will rapidly become unsustainable. (4) 780 new classrooms until 1999. After that and until 2020, hypothesis of 900 new constructions per year (5) Primary-level teaching salary: GNF 132,500/month. (6) Salaries are calculated on a quarterly basis. (7) Secondary-level teaching salary: GNF 302,000/month. (8) $7 per additional student (newly enrolled). (9) $12 per additional student (newly enrolled). (10) $1.1 million/year financed by donors. Progressive decrease (over 20 years) with 20% cost recovery. (11) $1,600 for teacher training (1,000 per year). I I 1 1 (12) Emergency fund for school maintenance: GNF 14,000 per classroom. I (13) Only applies to the number of teachers in the public system (=9972-600). Source: R. Prouty, World Bank. (14) The school year starts in September of the previous year. 140 Government scenario Annex 7 Table 7.3 Government Hypothesis: 53% GER in 2000 and 100% in 2022 can be attained with 45 students per class and teacher Page 1 of 4 $1= GNF 1,000 Program indicators Fonm 1996 to 2022frclassroomnconsti n and tehenrul af toachlev _____________________ ____-____-aaiiiveu a 8(e or,§;pe s and tadh6t: bk,2000.': **', ___ (a) Primary school 1994(14) 1995 1996 1997 1998 1999 2000 2001 2002 Primary school Number of school-age children 1,175,336 1,208,245 1,242,076 1,276,854 1,312,606 1,349,359 1,387,141 1,425,981 1,465,909 % increase In the number of school-age children 2.80% 2.80% 2.80% 2.80% 2.80% 2.80% 2.80% 2.80% Gross enrollment rate (1) 40% 42% 44% 46% 48% 50% 52% 54% 55% Number of enrolled children 471,792 506,342 543,423 583,219 625,929 671,767 720,962 763,383 808,301 Increase in number of enrolled children 34,550 37,080 39,796 42,710 45,838 49,195 42,421 44,917 % increase 7% 7% 7% 7% 7% 7% 6% 6% Cumulative increase 34,550 71,631 111,427 154,137 199,975 249,170 291,591 336,509 Total number of public school teachers 9,372 9,972 10,962 11,809 13,007 14,004 15,421 16,364 17,362 Net increase in number of primary school teachers 0 600 990 847 1,198 996 1,418 943 998 Total number of public and private school teachers 9,972 10,572 11,562 12,409 13,607 14,604 16,021 16,964 17,962 Number of students per teacher (b) 47 48 47 47 46 46 45 45 45 1008.22888 Infrastructure: number of classes 10600 11380 12160 12940 13720 14500 15400 16300 17200 Number of new classes required (4) 780 780 780 780 780 900 900 900 Number of children per class 45 44 45 45 46 46 47 47 47 (b) Lower secondary cycle Total number of children 84544 90787 97343 104229 111461 119057 127035 135414 144215 Net increase in number of children enrolled 6243 6556 6886 7232 7596 7978 8,379 8,801 % increase (2) 5% 5% 5% 5% 5% 5% 5% Net increase in number of secondary school teachers 0 0 0 0 0 0 0 Number of students per teacher (3) 36 39 42 45 48 51 54 58 141 Government scenario Page 2 of 4 In bilions f GNF1994 1995 1996 1997 1998 1999 2000 2001 2002 In billions of GNF cssI____ Recurrent costs (calculated by the project and adjusted for the Increase In the number of secondary school teachers) Budgetary base 52.51 52.51 52.51 52.51 52.51 52.51 52.51 52.51 Additional primary level salaries (4 trimesters) (4) (5) 0.24 1.35 2.86 4.35 6.18 7.93 8.30 11.51 Additional secondary level salaries (4 trimesters) (7) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Additional primary level teaching material (8) 0.24 0.26 0.28 0.30 0.32 0.34 0.30 0.31 Additional secondary level teaching material (9) 0.07 0.08 0.08 0.09 0.09 0.10 0.10 0.11 Primary school manuals (10) 0.22 0.22 School maintenance costs (12) 0.23 0.24 Total (13) current budget 53.07 54.19 55.73 57.25 59.10 60.88 61.66 64.90 Investment for Primary Education and Equity Project 13.62 13.62 13.62 13.62 13.62 15.3 15.08 15.08 Classroom construction [ 10.92 10.92 10.92 10.92 10.92 12.6 12.6 12.6 Primary level teacher training program (11) 1.6 1.6 1.6 1.6 1.6 1.6 1.6 1.6 Primary school manuals 1.1 1.1 1.1 1.1 1.1 1.1 0.88 0.88 TOTAL 66.69 67.81 69.35 70.87 72.72 76.18 76.74 79.98 Total recurrent cost per enrolled student (in GNF) 131.70 124.79 118.91 113.22 108.25 105.66 100.52 98.95 Recurrent cost per enrolled student (in GNF) 104.80 99.73 95.55 91.46 87.97 84.44 80.77 80.30 142 Government scenario Page 3 of 4 1994 1995 1996 1997 1998 1999 2000 2001 2002 Scenarto:2: Increas In!number,ofDschool-ge.children =L 4.6%. Fiom996 t2022: -cIdetoi consructinUn indteacker redulinent7 Program Indicators (a) Primary school Number of school-age children 1,175,336 1,228,226 1,283,496 1,341,254 1,401,610 1,464,682 1,530,593 1,599,470 1,671,446 % increase in the number of school-age children 4.50% 4.50% 4.50% 4.50% 4.50% 4.50% 4.50% 4.50% 4.50% Gross enrollment rate (1) 40% 42% 44% 46% 48% 50% 52% 54% 55% Number of enrolled children 471,792 514,716 561,544 612,634 668,371 729,180 795,520 856,258 921,634 Increase in number of enrolled children 42,924 46,829 51,089 55,737 60,808 66,341 60,738 65,375 % increase (2) 9% 9% 9% 9% 9% 9% 8% 8% Cumulative increase 42,924 89,752 140,842 196,579 257,388 323,728 384,466 449,842 Total number of public school teachers 9,372 9,972 11,348 12,435 13,930 15,252 17,078 18,428 19,881 Net incrase in number of primary school teachers 0 600 1,376 1,087 1,495 1,322 1,826 1,350 1,453 Total number of public and private school teachers 9,972 10,572 11,948 13,035 14,530 15,852 17,678 19,028 20,481 Number of students per teacher 47 49 47 47 46 46 45 45 45 1284.37 Infrastructure: number of classes (4) 10600 11380 12479 13614 14853 16204 17678 19028 20481 Number of new classes requIred 780 1099 1135 1239 1351 1474 1350 900 Number of children per class 45 45 45 45 45 45 45 45 45 1179.71 (b) Lower secondary cycle Total number of children 84,544 89,617 94,994 100,693 106,735 113,139 119,927 127,123 134,750 Net increase in number of children enrolled (3) 5,073 5,377 5,700 6,042 6,404 6,788 7,196 7,627 % increase 6% 6% 6% 6% 6% 6% 6% 6% Net increase in number of secondary school teachers (6) 0 0 0 0 0 0 0 Number of students per teacher 36 38 40 43 45 48 51 54 143 Government scenario Page 4 of 4 1994 1995 1996 1997 1998 1999 2000 2001 2002 In billions of GNF Recurrent costs Budgetary base 52.51 52.51 52.51 52.51 52.51 52.51 52.51 52.51 Additional primary level salaries (4 trimesters) (4) (5) 0.24 1.50 3.56 5.46 7.76 10.06 12.76 14.94 Additional secondary level salaries (4 trimesters) (7) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Additional primary level teaching material (8) 0.30 0.33 0.36 0.39 0.43 0.46 0.43 0.46 Additional secondary level teaching material (9) 0.06 0.06 0.07 0.07 0.08 0.08 0.09 0.09 Primary school manuals (10) 1 0.22 0.22 Primary-level teacher training program (11) 1.6 1.6 School maintenance costs (12) 0.27 0.29 Total recurrent costs 53.11 54.40 56.50 58.44 60.78 63.11 67.87 70.11 Investment for Primary Education and Equity Project 2.7 2.7 2.7 2.7 2.7 2.7 0.88 0.88 Classroom construction 10.92 15.3827224 15.8944537 17.3405311 18.9181726 20.6393479 18.8962606 12.6 Primary level teacher training program (11) . 1.6 1.6 1.6 1.6 1.6 1.6 Primary school manuals 1.1 1.1 1.1 1.1 1.1 1.1 0.88 0.88 TOTAL 55.81 57.10 59.20 61.14 63.48 65.81 68.75 70.99 Total recurrent cost per enrolled student (in GNF) 108.43 101.69 96.63 91.47 87.05 82.73 80.29 77.02 Recurrent cost per enrolled student (in GNF) 103.18 96.88 92.22 87.43 83.35 79.34 79.26 76.07 (1) 1994 figures given by the MEPUFP. (2) it is assumed that the secondary cohort will grow at the same rate as that of the primary (5%). .1 1 (3) If there is no new recruitment of secondary-level teachers, the number of students per class will rapidly become unsustainable. (4) 780 new classrooms until 1999. After that and until 2020, hypothesis of 900 new constructions per year (5) Primary-level teaching salary: GNF 132,500/month. (6) Salaries are calculated on a quarterly basis. (7) Secondary-level teaching salary: GNF 302,000/month. (8) $7 per additional student (newly enrolled). (9) $12 per additional student (newly enrolled). (10) $1.1 million/year financed by donors. Progressive decrease (over 20 years) with 20% cost recovery. . (11) $1,600 for teacher training (1,000 per year). I 1 1 (12) Emergency fund for school maintenance: GNF 14,000 per classroom. I (13) Only applies to the number of teachers in the public system (=9972-600). Source: R. Prouty, World Bank. (14) The school year starts in September of the previous year. 144 Budget and Expenditures in Education (MEPUFP and MESRSC data) Annex 7 Table 7.4 Education budget: economic composition (constant 1989 GNF millions) 1991 1992 1993 Amounts % of rotal ed. Amouots % of lotal ed. AmouUls % of lotal Ie% el level ed. leiel A. Primar) 19,72.1 il0 23,823.9 1')0' 19,292. 1iv 1. Wages and salaries 8,113.2 41 9,484.6 38 9,271.0 45 2. Nonwage expenditures 1,962.8 10 2,325.2 9 2,738.8 13 3. Subsidies 128.8 1 108.1 0.4 114.4 1 4. Administration 1,178.1 6 926.5 4 1,316.9 6 5. Capital expenditures 8,389.2 42 11,905.5 48 7,167.5 35 B. Secondary general 6,762.0 100 7,333.6 100 7,882.02 100 1. Wages and salaries 5,797.7 86 6,804.2 89 6,620.7 84 2. Nonwage expenditures 806.9 12 443.5 6 415.3 5 3. Subsidies 0 0 0 0 0.0 0 4. Administration 157.4 2 330.6 4 284.4 4 5. Capital expenditures 0 0 85.4 1 561.6 7 C. VET 3,267.8 100 3,251.1 100 2,954.1 100 1. Wages and salaries 1,573.1 48 1,855.7 52 1,858.1 63 2. Nonwage expenditures 233.9 7 391.2 11 230.4 8 3. Subsidies 171.8 5 180.2 5 188.4 6 4. Administration 304.0 9 348.6 10 485.8 16 5. Capital expenditures 985 30 824 23 191.3 6 D. Higher 9,886.0 100 11,187.8 100 15,545.6 100 1. Wages and salaries 4,221.4 43 4,415.2 39 4,382.5 28 2. Nonwage expenditures 1,962.8 20 2,393.4 21 5,878.2 38 3. Scholarships 1,914.4 19 1,435.6 13 1,532.9 10 4. Capital expenditures 1,787.3 18 2,943.6 26 3,752.0 24 E. Total 39,687.9 100 47,115.6 100 46,428.7 100 1. Wages and salaries 19,705.5 50 22,559.6 48 22,132.4 48 2. Nonwage expenditures 4,966.5 13 5,553.3 12 9,262.7 20 3. Subsidies 2,215.0 6 1,723.9 4 1,835.7 4 4. Administration 1,639.5 4 1,605.7 3 2,087.1 4 5. Capital expenditures 11,161.5 28 15,673.1 33 11,110.8 24 Note: Although these figures are not consistent with those presented in Annex 1, their trends are similar. 145 Annex 7 Table 7.5 Expenditures for education: economic composition (constant 1989 GNF millions) 1"9. 1992 1993 Educational Level Amounts % of budget Amounts % of budget Amounts % of budget A. Primar) 19,253.0 100 14,526.5 100 10.327.1 100 1. Wages and salaries 7,874.2 38 8,606.1 56 8,840.1 82 2. Nonwage expenditures 1,143.1 6 1,644.9 11 1,418.2 13 3. Subsidies 21.3 0.1 79.6 0.5 69.5 1 4. Administration 1,212.1 6 926.5 6 394.9 4 5. Capital expenditures 10,213.4 50 4,195.4 27 n.a. n.a. B. Secondary general 6,372.3 100 6,421.1 100 6,295.9 100 1. Wages and salaries 5,654.8 89 6,174.0 94 5,952.5 93 2. Nonwage expenditures 539.7 8 246.6 4 342.8 5 3. Subsidies 0.0 0 0.0 0 0.0 0 4. Administration 177.9 3 182.5 3 87.6 1 5. Capital expenditures 0.0 0 0.0 0 0.0 0 C. VET 1,358.8 100 2,341.1 100 2,068.4 100 1. Wages and salaries 1,076.7 69 1,683.8 65 1,756.7 77 2. Nonwage expenditures 172.9 11 416.9 16 139.2 6 3. Subsidies 40.2 3 181.3 7 173.0 8 4. Administration 197.5 13 255.9 10 201.5 9 5. Capital expenditures 68.5 4 59.2 2 na. n.a. D. Higher 8,587.9 100 9,636.9 100 11,429.01 100 1. Wages and salaries 3,881.6 45 4,194.5 44 4,163.413 36 2. Non-wage expenditures . 1,941.3 23 2,366.5 25 5,721.222 50 3. Scholarships 1,136.1 13 817.0 8 1,544.379 14 4. Capital expenditures 1,628.9 19 2,258.8 23 n.a. n.a. E. Total 36,980.0 100 34,289.6 100 30,805.1 100 1. Wages and salaries 18,487.2 50 20,658.4 60 20,712.7 67 2. Nonwage expenditures 3,797.0 10 4,674.9 14 7,621.4 25 3. Subsidies 1,197.6 3 1,078.0 3 1,786.9 6 4. Administration 1,587.5 4 1,364.9 4 684.0 2 5. Capital expenditures 11,910.8 32 6,513.4 19 n.a. n.a. Note: For the MEPUFP, payment amounts are those of the ordonnancement level, due to the unreliability of the data from the Ministry of Finance. Figures for the MESRC, except for salaries (see previous footnote), come from the mandatement level of the Ministry of Finance. Although the above figures diverge from those presented in Annex 1, their trends are similar. 146 ANNEX 8 POVERTY INCIDENCE ANALYSIS IN EDUCATION 1. As discussed in Annex 6, in order to measure the distribution of public expenditures on health and education among different groups in Guinea, expenditure incidence analysis has been used. 2. One caveat concerning the data should be noted. The total public education enrollments generated from the 1994 Integrated Household Survey and those from the MEPUFP differed significantly at the primary and secondary levels. The survey results yield a total enrollment that is 26% higher at the primary level than that reported by the MEPUFP and 78% higher at the secondary level (see annex 8, table 8.1). This issue is being investigated; for the present analysis, we assume that the MEPUFP enrollment figures are correct, since the MEPUFP education statistics are generally considered to be very reliable. Thus, the survey enrollments reported in the Integrated Household Survey were deflated separately for each region (to equal the MEPUFP enrollments), applying the deflator equally to all quintiles. This procedure has slightly biased the results, making the distribution of public expenditures appear more equitable than originally reported in the household survey. To test the sensitivity of the survey results to this assumption, we undertook the analysis in two ways: first, mapping the MEPUFP per student subsidy onto the original survey enrollments, and, second, mapping the MEPUFP per student subsidy to the deflated enrollment figures. The results, which were quite similar, are shown in annex 8, table 8.2. Since there was not a large difference between the two and for the sake of strict comparability with other analyses based on MEPUFP enrollments, we present only the results of the second method in the text. 3. This was not as serious an issue at the technical and professional and university levels, since the survey was not designed to provide reliable estimates of these enrollments at the regional level. For the sake of consistency, we used the same deflating procedure for the analysis at this level as well. 147 Annex 8 Table 8.1 Per student public education expenditures ......Cwnary :5 eUpe G-ine L ine : 'uea:FOrest. eisjdde Guinea All G umes Primary Salary 3,589,773,032 3,434,867,254 3,984,895,016 5,628,243,271 4,054,490,366 20,692,268,939 Non-salary 373,240,236 281,440,976 336,539,268 367,143,156 414,161,788 1,772,525,424 Total 3,963,013,268 3,716,308,230 4,321,434,284 5,995,386,427 4,468,652,154 22,464,794,363 Number of students, MEPUFP (a) 99,295 76,302 87,414 98,189 91,022 452,222 Number of students, Survey (b) 163,240 81,276 106,138 130,856 90.003 571,513 (b/a) 164% 107% 121% 133% 99% 126% Cost per student, MEPUFP 39,912 48,705 49,436 61,060 49,094 49,676 Cost per student, Survey 24,277 45,725 40,715 45,817 49,650 39,308 Secondary Salary 2,453,797,326 1,086,585,458 1,694,983,514 1,567,017,871 1,344,761,755 8,147,145,924 Non-salary 124,875,463 59,320,799 96,908,035 154,330,148 80,409,129 515,843,574 Total 2,578,672,789 1,145,906,257 1,791,891,549 1,721,348,019 1,425,170,884 8,662,989,498 Number of students, MEPUFP (c) 42,060 11,120 19,670 17,519 15,737 108,661 Number of students, Survey (d) 77,890 13,475 27,211 17,066 15,986 156,938 (d/c) 185% 139% 138% 97% 100% 144% Cost per student, MEPUFP 61,309 85,039 91,098 98,256 89,425 79,725 Cost per student, Survey 33,107 61,001 65,852 100,864 89,151 55,200 Technical and Professional Salary 857,594,308 334,506,680 343,486,725 296,341,489 377,161,895 2,209,091,097 Non-salary 173,162,223 48,105,468 44,799,461 58,466,347 6,666,668 331,200,167 Total 1,030,756,531 382,612,148 388,286,186 354,807,836 383,828,563 2,540,291,264 Number of students, MEPUFP (e) 2,838 864 948 837 646 6,133 Number of students, Survey (f) 5,198 (f/e) 84% Cost per student, MEPUFP 363,198 442,838 409,585 423,904 594,162 414,200 University Salary 7,840,000,000 Non-salary 14,700,000,000 Total 22,540,000,000 Number of students, MEPUFP 8,101 Number of students, Survey - - - - 5,853 Cost per student, MEPUFP 2,782,373 Source: Expenditures from MEPUFP, 1995. Number of students from MEPUFP and 1994 Integrated Household Survey. Note: Number of students is for public schools only. 148 Annex 8 Table 8.2 Results of alternative methods of incidence analysis I... ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~....... ...._____**...:..:..::::.....*.* . . ............. ________________ .....................M m "r ~ : :. .: :: : :......:::::::~:~ Quintiles Male Female Total share Male Female Total share 1 3,762 534 2,044 9.41% 2.901 372 1,555 10.91% 2 5,039 1,690 3,272 15.03% 3,499 1,154 2,262 15.83% 3 5,684 3,027 4,262 19.62% 4,032 2194 3,048 21.38% 4 7,685 5,399 6,484 29.76% 5,130 3,695 4,376 30.60% 5 6,427 4,979 5,695 26.17% 3,446 2,641 3,039 21.27% Total 5,730 3,102 4,349 3,800 2,001 2,855 Quintiles 1 617 134 770 3.07% 401 87 234 3.50% 2 1,494 210 1,729 6.94% 888 123 484 7.22% 3 4,265 658 5,023 19.87% 2,259 389 1,384 20.67% 4 6,058 1,281 7,477 30.11% 3,402 747 2,007 29.89% 5 6,051 3,399 9,528 40.01% 3a294 1,918 2,598 38.73% Total 3,723 1,113 4,956 2,116 640 1,340 ...or .al Primr 'anS ecndar Quintiles 1 4,380 668 2,814 6.08% 3,302 459 1,789 8.52% 2 6,533 1,900 5,000 10.80% 4,388 1,278 2,746 13.08% 3 9,949 3,685 9,285 20.06% 6,561 2,583 4,432 21.12% 4 13,744 6,681 13,961 30.16% 8,532 4,441 6,383 30.42% 5 12,478 8,378 15,223 32.89% 6,740 4,559 5,636 26.86% Total 9,454 4,215 9,304 5,916 2,641 4,195 Method I wlas estimated using MEPUFP per student education subsidy and enrollments as reported in the 1994 Integrated Household Survey. b Method 2 was estimated using the MEPUFP per student education subsidy and enrollment from the 1994 Integrated Household Survey, which were deflated by region to equal total enrollments reported by the MEPUFP. Per Capita Expenditures in Education Per Capita Public Expenditures by Expenditure Group 4. The following discussion is based on annex 8, table 8.1 above. At the primary level, per capita expenditures are still more than twice as large for the richest quintile (GNF 4,088 per capita) as they are for the poorest quintile (GNF 1,768 per capita), although the fourth quintile receives the highest per capita expenditure of GNF 5,244. Given the age composition of the poorest groups (who tend to have more children), one would expect that the per capita expenditures for primary education would be larger among the poorer groups, reflecting the larger number of school-age children in these groups compared with richer groups. In Guinea, however, the extremely low enrollment rates among the poor means that they receive a much smaller amount of the public expenditure than do the richer groups, despite the age composition differences between the groups. At the secondary level, per capita expenditures are biased heavily towards the richest quintiles; the richest quintile receives per capita expenditures that are more than ten times larger than the expenditures for the 149 poorest quintile. Technical and professional education and university education almost exclusively favor the richest two quintiles. Per Capita Public Expenditures by Location, Gender, and Region 5. Even if the technical/professional school and university-level subsidies are discounted (since these types of services are usually offered only in urban areas), per capita expenditures for primary and secondary education are strongly biased towards urban areas. At the primary level, urban expenditures are almost twice as high as in rural areas (GNF 4,789 and GNF 2,795, respectively). At the secondary level, the difference is even more dramatic: GNF 3,031 per capita in urban areas and GNF 467 in rural areas. These differences are explained almost entirely by lower enrollments in rural areas. At the primary level, per capita expenditures for boys are roughly two times larger than for girls (GNF 4,601 and GNF 2,451, respectively). At the secondary level, it is a more than threefold difference (GNF 2,110 and GNF 638, respectively). 6. Guinea Forestire and Conakry have higher enrollment rates than do the other three regions (see annex 8, figure 8.1 below). At the primary level, per capita expenditures are distributed fairly equally among regions, although they are slightly higher in Conakry and Forest Guinea. A much more unequal regional distribution exists at the secondary level, where per capita expenditures are significantly higher in Conakry than in the other regions. Annex 8 Figure 8.1 Per capita distribution of public expenditures on education by region 6,000 5,000 -E 4,000 .secondary 3,000. E primary 2,000 1,000. 00 U LL Region How Well Targeted Are Public Expenditures on Education?7 7. Only the curves for primary, secondary, and all education lie above the Lorenz distribution of total educational expenditures per capita, indicating that they are distributed in a more pro-poor For a discussion on how expenditure effectiveness is calculated, see annex 6 on health. 150 way than are per capita educational expenditures. In this sense, overall education expenditures can be considered relatively pro-poor. Annex 8 Figure 8.2 Comparison of relative and absolute distribution of public expenditures on education: 1994 100 ....a 8 0 se condar ry 60 __ 40 _prim a_ry 20 1xen nditu re 2 tech/pref 0 20 40 60 80 100 cum ulative share of population Contribution of Households to Financing Public Education 8. Households spend an average of GNF 18,500 per student in public primary school and GNF 48,180 per student in public secondary school (annex 8, table 8.3). Households spend about equal amounts on boys and girls, but households in urban areas spend about two and a half times more than those in rural areas at the primary level and more than five times more at the secondary level. Annex 8 Table 8.3 Per student household expenditures on public primary and secondary education: 1993-1994 (GNF) kQiiintil&dGr6aip Brimnauy , -Seconday . Poorest 6,996 17,698 2 9,684 47,899 3 11,438 30,741 4 18,636 44,808 Wealthiest 30,325 57,581 Average 18,500 48,180 Urban 25,390 50,555 Rural 10,328 35,191 Male 18,151 46,626 Female 19,073 52,306 151 The Burden of Education Expenditures on Households Annex 8 Table 8.4 The burden of household spending on public education .dePer capitaneafod Percap1it edcation1 Educati eenitorsaaha.o Poorest 31,44 308 0.98% 2 56,85 952 1.68% 3 102,055 1,604 1.57% 4 188,408 4,513 2.40% Wealthiest 697,192 7,711 1.11% Average 215,113 3,013 1.40% Note: Household spending on education is for public education only. Source: 1994 Integrated Household Survey, preliminary results. 152 ANNEX 9 AGRICULTURE The Roots of the Current Mandate of the Ministry of Agriculture (MAEF): A Look at the Legacies of the Colonial Period, the First Republic, and the First Decade of the Second Republic 1. Colonial period legacy. Agricultural exports, especially pineapples and bananas, put Guinea on the world's economic map. While these exports have decreased drastically and, in the case of bananas, disappeared, MAEF's policy and conduct reflect a nostalgia for the booming agricultural export era. Indeed, agricultural export promotion is a top priority objective of agricultural policy in the Letter of Agricultural Development Policy (LADP), and is often reiterated during discussions with donors. This stated objective led MAEF to entrust the National Directorate of Agriculture (DNA) with the "responsibility" for the agricultural export component of the LADP and staff of at least forty civil servants. However, DNA lacks the material means and a clear strategy to implement the statutory mandate, which is itself questionable given the Government's private sector-led growth strategy. The new forestry legislation also reflects this period's legacy to a large extent, as it entrusts DNFF with the responsibility of managing gazetted state forests and protected areas as well as assisting and controlling all other privately-owned wood stands. 2. Legacy of the First Republic. The socialist era -- the State must do everything. The Sovkoz model of farming, heavy mechanization, and production brigades were the hallmarks of this era. While the production farms (fermes agro-pastorales d'arrondissement or FAPAs) were abolished in 1985, the personnel of the ministry did not undergo a concomitant downsizing until 1990. Although the ministry reduced its original roster by 9000 at that time, large numbers of former FAPA workers have been converted into contractual workers for low-land development (tdcherons aminagistes) and rural well-drilling (brigades source et brigade puits). These workers are solely contracted by the Ministry of Agriculture (National Directorate of Rural Engineering, DNGR, and National Rural Water Supply Service, SNAPE), seldom find contracts with private sector employees, and have been in this situation ever since the FAPAs were abolished. To illustrate, SNAPE has 60 permanent civil servants but as many as 250 contractual workers. The situation at the Institute of Agronomic Research of Guinea (IRAG) is similar. Second, during the First Republic, with its obsession for food self-sufficiency, a major emphasis was placed on training agricultural technicians and agronomists. Eight agronomy universities (Facultds d'Agronomie) were created, producing hundreds of graduates yearly who inflated the civil service roster of the Ministry of Agriculture.8 All new graduates became agronomists at Level A, the highest professional level of the agricultural ministry. As such, the graduates would not be stationed in the interior of the country for field work, which is reserved for lower level civil servants. Under the First Republic, the ministry had a legally binding agreement to hire all agronomists trained in Guinea and abroad. Clearly, the ministry was bound to face problems of overstaffing under these circumstances. To exacerbate the employment situation at MAEF, all the agronomists who were running the FAPAs returned to headquarters in Conakry (at To give an idea, in the early 1970s, Morocco (population 20 million) was producing about 50 agronomists (after a six-year training period) a year while Guinea (population 4 million) produced close to 1000/year (after a four-year training period). 153 DNA, primarily) when the FAPAs were closed. No longer able to absorb all of the new graduates, the Ministry stopped recruiting in 1986 and later, in 1987 closed down all agronomy universities altogether. The forestry sector was not considered a priority during this period. As a result, considerable work (inventories, trained staff, etc.) was lost. 3. Legacy of the first decade of the Second Republic. Since 1985, the Government has shown firm resolve to break with the past and make up rapidly for past decades of isolation and near backwardness. This meant high-paced investment in physical infrastructure especially, and in agriculture. This new approach met favorable terrain with donors, especially the bilaterals, more disposed to renew relations with Guinea now that the political and economic climate had become more favorable for externally-financed investment. Donors led the preparation and evaluation of projects they financed and even ran them sometimes -- because of low domestic capacity (most administrators were trained in the former Eastern European model of management in the former Soviet Union and Czechoslovakia). The recourse to large teams of long-term technical assistants also stems from this first decade of relatively low domestic capacity. 4. Government strategy for sectoral growth calls for providing incentives and support for private sector development, including the transfer of economic activity to the private sector, where it has sufficient capacity. In addition to abolishing the FAPAs, the Government has discontinued most force account production of agricultural goods and services. Progress has been made with: (a) DNGR, whose role in the implementation of the rural road programs is now confined to procurement matters and, in a few instances, to control of works;9 and (b) the National Directorate for Livestock (DNE), whose work force was reduced by two-thirds in 1989 and which actively sought privatization of the veterinary function (with private veterinarians and local animal health artisans, auxiliaires d'ilevage). By preparing and enacting the Code Foncier (land tenure code), virtually all the agricultural domain is now in private hands (whether individual firms or community). 5. One important MAEF area, however, where privatization and state divestiture have not been attempted is the drilling of boreholes for rural water supply. It remains under public service control. Only residual drilling contracts are tendered to large international firms and virtually no small local private enterprises participate. For low-land development and small-scale irrigation works, DNGR subcontracts with private individual technicians (tdcherons amenagistes); and for more complicated and larger irrigation and drainage works, it subcontracts with local small- and medium-sized firms. For maintenance of the construction, DNGR involves NGOs, private individuals (cantonniers), or villagers (actions communautaires). In the latter case, DNGR entices villagers to contribute to the construction as laborers or by providing building materials and food to the contractor workers. 154 Sectoral Performance and Contribution to Overall Growth Annex 9 Figure 9.1: Agricultural GDP Growth (1988-92) South Asia (excluding India) Sub-Saharan Africa Mali Senegal Guinea-Bissau E7 C6te d'voire Burkina Faso Guinea -3 -2 -1 0 1 2 3 4 155 Annex 9 Figure 9.2: Index of Selected Sectoral Performance Indicators and Public Expenditures in Agriculture (1988=100) 320 - 270- 220 170 - 120- 70 1988 1989 1990 1991 1992 1993 1994 - - - Real Agricultural PIP ------ Real Total Agricultural Budget - -0- - Rice Output (paddy) - Rice Import Volume - - - - Forest Cover --X-Area Harvested Rice Crop Yield Annex 9 Figure 93: Public Investment Program in Rural Development, 1991-94 National Services Others (including Rural 19% Infrastructure) Regional 33% Development Projects 11% "Fili6re" Projects Integrated Rural 23% Development Projects 14% 156 Rural Infrastructure 6. In the absence of a coherent national program, all major agricultural development projects conceived prior to 1989 had their own rural road rehabilitation and low-land development components. These components were implemented by force account, thus superseding markets, and contributing little to private sector development.'0 It is not known how much was spent on rural infrastructure under these different projects. However, in terms of physical output, it is estimated that no more that 250 km of rural roads were rehabilitated and 1100 ha of low land developed during the 1984-1990 period. Annex 9, table I shows the results of the rehabilitation program since 1990, its costs, and its maintenance expenditures.. Thus far, maintenance is financed by donors as per the credit agreements. Annex 9 Table 9.1: GUINEA - Rural Roads Rehabilitated 1990-1995 Length I km) tUSS muion) Ntainfenance (US5'6001 LipK to 9SQ <i -) 1990 260 0.9 -- 1991 370 1.1 -- 1992 960 4.2 145 1993 975 7.4 213 1994 1250 22.7 420 1995 1100 17.3 471 10 All major integrated rural development projects, such as PDR-Haute Guin6e, PDR-Basse Guinde, SOGUIPAH in Guinde Foresti6re, and Sud-Est Fouta Djalon share this feature. 157 Annex 9 Table 9.2: Public Expenditure in Rural Infrastructure by Source of Funding and Category of Expenditure, 1990-96 (in US$ millions) Source of Funding Category of Expenditure Training, Program IDA USAID CID KlW Other Govt. Total Civil Vehicles & Tech. 0 & M Road Other Total External Works Equipment Assistance, Maintenance Studies PNIR 43.8 33.0 - 8.5 0.8 10.2 96.3 56 6.2 25.1 6.6 0.8 1.6 96.3 Other 4.1 - 22.0 - 11.5 - 37.6 31.7 0.5 2.9 2.1 - - 37.2 Total 47.9 33.0 22.0 8.5 12.3 10.2 133.9 87.7 6.7 28 8.71. 0.8 1.6 133.5 Annex 9 Figure 9.4: Shares of Public Expenditures in Rural Infrastructure by Source of Funding and Major Categories By Source of Funding By Category of Expenditure Govt. Other 8% Road O&M External Training, Maintenance 7% Other KFW 36 ATech. Assistance, I% 1% 6% Studies 21% CfD I16% Vehicles & Civil 2S% Equipment Works 5% 65% 158 7. The National Rural Infrastructure Project (PNIR) helped reinforce the institutional and technical capacity of DNGR, to the extent that the directorate is now the sole coordinator and .implementing agency for all activities pertaining to road rehabilitation and low-land development in rural areas over the entire country. A Master Plan of rural roads under the aegis of the DNGR was drafted, identifying 8000 km of rural roads which needed to be rehabilitated as a priority. Funding for the rehabilitation of 5740 km has been secured, and works on 3492 km were completed by the end of 1994. Average annual expenditures on the rehabilitation of rural roads and bridge construction amounted to US$16 million, and cost per km varied between US$18,000 to US$25,000 which is high given the large number of long-term expatriate technical assistants initially involved in PNIR." More than 90% of the program in 1993 (and 98% in 1994) was financed by international donors. Annex 9, table 9.2 shows the aggregate public expenditures on rural infrastructure (excluding staff salaries) by source of funding and category of expenditure during the period 1990- 96. 8. Road works account for more than 60% of the total expenditures on rural infrastructures and are almost entirely financed by international donors. Technical assistance in the implementation of the rural infrastructure program is hefty, although it has been declining since the mid-term review of PNIR in 1993 from 23 long-term resident technical assistants before the review to only nine currently.12 9. The need to maintain the rehabilitated rural roads is estimated to increase from GNF 524 million in 1995 (for 661 km) to about GNF 3.0 billion by the year 2002 (for 2710 km). These estimates do not include the cost of maintaining 2393 km of rural roads that the rural community users are assumed to be in charge of. The willingness of community users to participate in maintenance was appraised before the construction of the roads. We need to know more and update it. Agricultural Extension 10. Prior to 1984, there was virtually no extension service for smallholders in Guinea. The few who received extension advice where associated with area-based rural development projects. These projects, however, had limited geographical coverage and their extension messages focused on a single crop. These services stopped once donor funding disappeared; no institutional development resulted from them. In 1984 the Ministry of Agriculture was entrusted with responsibility for agricultural extension but no national strategy was designed then. It is not known how much was spent on extension during the period prior to the launching of the IDA-financed National Agricultural Research and Extension Project (SDR 14.2 million). In its first years, the project suffered from a lack of counterpart financing and extension activities came virtually to a standstill. In addition, while in the other countries of the sub-region (e.g., Benin and C6te d'Ivoire) the rural roads program is largely spot improvement, in Guinea, it is full-fledged construction which costs twice as much. 12 By the time PNIR closes, expenditures category Technical Assistance and Training will have consumed a little less than 50% of total project costs. In view of the quality of rural roads rehabilitated by local small- and medium-sized enterprises and under the pilot program for rehabilitation using High Labor Intensive methods of the International Labor Organization at Dubreka, it is possible to reduce cost of rehabilitation substantially. 159 11. In 1992, after a redesigning of the credit agreement to tailor Government contribution to a sustainable level, the project picked up speed. Current discussions concerning the financial sustainability of agricultural services including extension are being carried out under the forthcoming Agricultural Services Project. Of particular interest is the issue of cost recovery in the long run for high value crops and farmers' organizations. The Government contribution to and foreign financing of public expenditures in agricultural extension for the period 1990-94 is in annex 9, table 9.3. A sizable amount of expenditures on agricultural extension undertaken infilibres, and integrated area development projects is not known and is not included in this table. Cost effectiveness of the various approaches being an issue, it is important to get data for comparison and to make informed recommendations. Annex 9 Table 9.3: GUINEA - Domestic and Foreign Financing of Agricultural Extension, 1990-94 (in GNF millions) .99 1991 1992 ..9 1994 TOTAL Counterpart Financing 243.3 200.0 400.0 150.0 113.7 1,107.0 Wage Bill for Civil Servants 170.0 1,557.1 1,648.5 1,650.6 1,656.7 7,682.9 Office Rental 16.6 16.6 16.6 16.6 16.6 82.8 Office Furniture 12.6 12.6 Memo Item Roster 650 811 785 786 767 Total Domestic Contribution 1,527.3 1,916.0 2,202.4 1,954.7 1,925.0 8,885.3 Foreign-Financed investment 946.0 1,173.7 1075.0 1,665,9 1,605.4 7,106.1 TOTAL EXPENDITURES 2,473.3 3,089.7 3,277.4 3,620.6 3530.4 15,991.4 Livestock Services 12. The department of livestock services administration (Direction Nationale de l'Elevage, DNE) underwent the most drastic restructuring of all MAEF's technical departments. The functional and economic structures of public expenditures on livestock services are described in annex 9, tables 9.4 and 9.5 below. Annex 9 Table 9.4: Functional Composition of Investment Expenditures in Livestock, 1992-94 (US$ '000) Programs 1992 1993 1994 Total Share 1992-94 (%) Average Studie 77 146 109 332 8 74 1 10 Animal Husbandry 566 425 318 1309 34.47 436 Animal Health 713 602 450 1765 46.48 588 Privatization and Herders' 165 129 97 391 10.30 130 Organizations Total 1521 1302 974 3797 100.00 160 Annex 9 Table 9.5: Economic Composition of Public Expenditures in Livestock Services, 1992-94 (in GNF millions) 199 1994 Toal Share 1992-94 1992 993 194 Toal'op 45erage C ul Ser% ice Wage Bill 1350 135) 1350 -1051: 86 13 5 Operating and Maintenance 291 146 226 663 14 221 Total 1641 1486 1576 4713 100 1571 161 ·’꽉―…안”·． Annex 9 Table 9.7 Volume of Foreign Assistance to Guinean Agriculture, 1995 (in GNF millions) ......... ... ......... ..... .... ... .................... M :4 ............ ......... ........... Multilateral IDA 105,123.00 credit EDI` 103,016.00 credit KfW 80,168.00 credit ADB/ADF 68,654.50 credit CfI) 48,860.00 credit IFAD 45,316.70 credit IDB 20,073.00 grant SFI) 7,247.20 grant OPEC 6,291.00 grant UNDP 4,405.10 credit UNCD1` 2,986.50 credit EIB 2,747.50 eredit LTNICEF 2,538.50 credit/grant ECOWAS 2,491.60 credit ABEDA 11,140.00 grant FAO 1,138.85 grant Subtotal 512,167.45 Bilateral USAID 56,214.00 grant Germany 16,569.10 grant Italy .7,602.05 grant FAC 6,866.90 grant orea 0.371 gran Subtotal 87,252.42 GRAND TOTAL 599,449.90 ExchangeratesasofAttgust25,1995: 1DM=667.I4GNF;1DTS=1,480.7GNF;1ECU=1,260.36GNF;1FF= 196.2S; 1 LI = 0.6 1 GNF; I UC = 1,53 9GNF; I USD = 995.5GNF. Source: MAEF/BWA, Budget dInvestissement Public du Secteur Agricole, p. 10. 163 ANNEX 10 THE IMPACT OF THE GARAFIRI/KALETA PROJECT ON PUBLIC EXPENDITURES Introduction 1. Guinea's long-standing ambition has been to take advantage of its considerable hydropower potential and achieve energy self-sufficiency. The cost of the Garafiri dam has recently been estimated at US$200 million over 6 years -- varying from 10% to 22% of the annual investment budget between 1996-2000. The cost of the associated Kaleta power plant is estimated at around $240 million. The Bank has decided not to participate in the financing of the project based on: (i) the equalizing discount rate (EDR) (as calculated by the Bank) having been deemed too low essentially because the project would be premature; and (ii) the project's risks have been deemed too important, in particular in view of the current situation in the power sector. The Bank made its decision known to the Government in early 1995. Financing Details 2. Because financing details are unavailable, it is assumed here that the original expenditure schedule will stay the same. The equalizing discount rate (EDR) of the combined Garafiri/Kaleta project compared with thermal power plants has been estimated by the Bank at 9%. The Government-owned electricity company (ENELGUI) will be the owner and the manager of the project. Analysis of the impact of the project on public finances is based on the figures provided by the Government to the Bank in February 1995. The analysis will be updated as new information on financing is available. In the absence of information on lending conditions, several assumptions have been made about ENELGUI's financial situation and lending conditions. Garafiri's EDR 3. Based on the Bank's estimate of Garafiri's EDR (9%) and on conventional wisdom about sectoral investment efficiency, public resource allocations will become inefficient as Garafiri crowds out such programs as girls' primary education or child immunization. This problem would be compounded by the Government's limited capacity to estimate and compare public investment programs on the basis of conventional performance criteria (cost-effectiveness, externalities, and so forth). 4. If undertaken now, the construction of the Garafiri/Kaleta project would create excess capacity in the sector. And it is likely that overconsumption will continue, due mainly to low collection rates and the theft of electricity. These costs may increase further as the Government's limited capacity to pay for electricity consumption pushes up its arrears. All such costs would have to be passed on in the form of electricity prices, thus penalizing law-abiding consumers. A rough estimate of this loss would be about $0.04/KwH. The cost increase for the Government (as a consumer) would be between GNF 3 billion and GNF 4 billion annually. 164 Prospects for ENELGUI 5. Approximately 50,000 people are currently registered electricity customers in the Conakry/Kindia region and the Government forecasts a 7% growth in demand for electricity which will require additional investment by ENELGUI to bring the distribution network in line with Garafiri/Kaleta's production capacity. The implementation of the Garafiri/Kaleta project, even if done efficiently, will cause undue strain on ENELGUI's financiai and human resources which are badly needed in the short term in view of the difficult situation in the power sector. In addition to uncertainties about the evolution of demand and ENELGUI's investment capacity, the number of clients is likely to grow more slowly than the 7% projection, given the high costs of electricity (including costs induced by inefficiencies) an large losses; in effect, the majority of the population, including the poor, would remain barred from access to electric power. 6. It is assumed that a significant part of foreign financing will be made either to ENELGUI with or without (as in the case of the CFD) the Government's guarantee or to the Government, which will on-lend to ENELGUI. This process raises a formatting issue pertaining to how the project is presented in the TOFE: external financing for the project should not appear in the FINEX; foreign borrowing should not accrue to the Government's debt; or on-lending to ENELGUI should appear under net lending in the TOFE. 7. Projections by Bank staff of ENELGUI's financial parameters for the next five years found that the company is likely to remain in serious financial straits at least during the construction of Garafiri (until year 2000) and will not be able to assume local costs. Therefore -- and in the absence of further information on financing conditions at this point -- all local costs (except Garafiri's recurrent costs for O&M) appear under Government financing in annex 10, table 10.1 under the following assumptions: the majority of loans are Government-guaranteed; ENELGUI will not have the financial capacity to support costs until year 2000, and only very limited capacity thereafter; and overall financing conditions include a mix of grants and loans under market conditions, generating an average 6% interest and five-year grace period. Clearly, additional information on financing conditions and ENELGUI's capacities must be sought before the issues discussed in the text are finalized. 8. Although the Garafiri/Kaleta project will have an impact on both on the Budget National de Diveloppement (BND) and the current budget (for debt service), the locally-financed budget is viewed in the text as a whole, since resources are fungible. Impact of Garafiri/Kaleta on the Efficiency of Public Finance Management 9. The foremost problem with realizing the dam and the electrical power plant is that it will drain local resources for projects. Unless the Government improves its domestic resource mobilization efforts dramatically in the next few years, the construction of Garafiri/Kaleta will have serious consequences on the Government's capacity to secure financing for its contributions to FINEX projects, as well as on its own BND capacity (annex 10, table 10.1). Because the Government's priority expenditures are expected to go to the construction of the dam/power plant, the drain on the ]BND would total approximately 10% of the annual BND budget between 1995 and 2000 under the high-growth expenditure scenario, and about 20% under the low-growth expenditure scenario (chapter 1, table 1.1). If the drain is calculated on the basis of the average amount of BND disbursements in the past three years (55%), the drain on domestic resources would roughly double under the two scenarios, to 20% and 40%, respectively. This budgetary depletion would further slow the implementation rate of the PIP and affect the efficiency of public investment accordingly. 165 Annex 10 Table 10.1 Impact of the GarafirilKaleta project on public expenditures (GNF billions; US$1=GNF 1,000) 1995 1996I19 19 1999 20 GARAFIRI Foreign financing 0.00 33.05 48.31 54.24 50.85 Government financing o/w 5.08 4.66 9.75 9.75 9.74 11.19 IDC 2.54 7.63 7.63 7.62 BND' *5.08 2.12 2.12 2.12 2.12 Interest' 11.19 ENELGUI financing (average recurrent cost) 3.00 KALETA Foreign financing 0.00 16.44 30.81 47.17 Total Government financing o/w 4.72 10.17 IDC 2.72 8.17 Interest (6%) BND v 2.00 2.00 Total Financing 5.1 37.7 58.1 80.4 96.1 71.5 Scenario 1: tax effort and expenditure real growth limited to 6.9% G+K as % FINEX budget 0% 11% 15% 21% 24% 14% G+K as % BND budget 10% 7% 12% 11% 13% 8% as % of total PIP 1.6% 10.4% 15.0% 19.3% 21.9% 15.5% Scenario 2: no tax effort and expenditure real growth limited to 2.3% G+K as % FINEX budget 0.0 12.2% 17.4% 24.9% 28.1% 15.9% G+K as % BND budget 10% 9.0% 18.3% 17.9% 26.0% 37.5% as % of total PIP 1.9% 11.6% 17.5% 23.7% 27.6% 20.1% a new IDA estimate. b BND covers road construction (Garafiri only) and civil works. c 6% of construction cost financed by FINEX. This cost should be assumed by ENELGIJI but it is assumed here that (a) ENELGUI's financial situation remains difficult after 2000 and (b) all loans are guaranteed by the Government. d IMF/Govemment projections. 10. Under existing macroeconomic constraints that imply an overall cap on the level of public investment, the project will require an important tradeoff in investment allocations. For the BND, which will be hardest hit by the constraints, the tradeoff will radically shift resource allocations, with consequent effects on the financing capacity of priority sectors. For the FINEX, to the extent that the macroeconomic constraints are binding and some donors might finance Garafiri/Kaleta rather than other projects, the construction will increase allocations to the energy sector to the detriment of other sectors. If social programs were to suffer directly from the construction of Garafiri/Kaleta, the impact on poverty would be obvious. In addition, the project would have an indirect redistributional effect by perpetuating restricted access to electricity by a very small portion -- the wealthiest one -- of the population. 166 ANNEX 11 THE INSTITUTIONAL ENVIRONMENT AND THE BUDGETARY PROCESS Evaluation of the Institutional Structure: Much Shuffling Since 1990 1. The first public budget in the history of Guinea was prepared in 1988. From 1988 to 1990, the then Ministry of Economy and Finance was responsible for preparing both revenue and expenditure provisions, including the PIP. A Government reorganization in late 1990 separated the various responsibilities of the ministry into two entities: the Ministry of Finance, with responsibilities for the treasury and public accounting, including the preparation of the current budget and implementation of both the current and investment budgets; and the Ministry of Plan and Cooperation, with responsibility for economic management strategies, data collection, PIP preparation, and management of external grants. Decision making control has rested at the Executive level and the ministers have maintained close working relationships with the President, thus influencing the evolution of central institutions and the definition of their mandate. 2. After the 1990 reorganization, the National Directorate of Treasury, the National Directorate of the Budget, and the National Directorate of Procurement were established at the Ministry of Finance under clear mandates. The public debt, a key management function, was established only as a division within the Directorate of the Budget. The Customs Office, while part of the Ministry of Finance, is managed autonomously. For the new Ministry of Plan, the National Directorate of Public Investments (DNIP) was strengthened further with assistance from the donor community. The DNIP was organized along sectoral lines, with responsibilities for evaluating investment projects in the respective sectors, recommending their inclusion in the PIP, following up on their implementation, and proposing the level of PIP budget allocations that would enable projects to be implemented efficiently. Despite significant levels of technical assistance, it has not strengthened its capacity to evaluate projects and follow up on their implementation. Within the Directorate of the Budget at the Ministry of Finance, the Division of Investment Expenditure (DDI) was organized along the lines of the DNIP to follow up on investment expenditures. Another directorate within the Ministry of Plan, the National Directorate of Cooperation (the DNCI), was made responsible for negotiating and following up on external grants. 3. In principle, these separate structures were to establish a close level of coordination for arriving at realistic PIP and recurrent budget allocations that were consistent with macroeconomic and sectoral strategies, priorities, previous performance on investment targets, and overall resource constraints. In practice, these structures operated much on their own, each trying to preserve and expand its area of responsibility. This was especially true of the DNIP, which also became involved in securing external financing for PIP projects and implementing the investment budget. The DNIP's involvement slowed down the public procurement process, making it almost unmanageable and expensive. Government officials estimate that as much as 25% of contracts costs are associated with the risk of doing. business in Guinea. 4. Another structure within the Ministry of Plan -- the National Directorate of Economics (DNE) -- was to play an important role in budget preparation, with responsibility for preparing macroeconomic policy objectives and growth perspectives, and thus providing the macroeconomic 167 framework for the budget. The various divisions of this Directorate received large amounts of technical assistance from the donor community. But their ability to produce the necessary input for PIP preparation after the end of technical assistance was always an issue. Internal organizational problems, poorly-trained staff, and the dearth of operating resources were important impediments. 5. A major Government reorganization at the end of 1992 changed these institutions again, as the two ministries were remerged into one: the Ministry of Finance and Plan. But the organization responsible for grant management -- the DNCI of the former Ministry of Plan and Cooperation -- was carved out and annexed to the Ministry of Foreign Affairs. This separation of the DNCI from the PIP process has had serious repercussions: it has shut down coordination between the DNCI and DNIP and the Directorate of the Budget, and several projects that are financed by grants and managed directly by donors have been excluded from the PIP budget. 6. The merger of the Ministries of Finance and Plan attempted to correct the institutional rivalries that existed among the procurement, budget, treasury, and public investment directorates. An internal reorganization was to follow the creation of the new ministry. Its objective was to eliminate duplicative structures and excessive bureaucratic layers and to improve the organizational hierarchy, thus simplifying the budget preparation and implementation processes. The Division of Investment Expenditure (DDI) of Finance and the DNIP of Plan were to become one. Unfortunately, the reorganization never took place, and each structure continued to conduct business as in the past. Public procurement became a major problem, involving the Government implementing agency, the procurement directorate, the public investment directorate, and the Executive. To address the prevailing institutional rivalries, a decree by the Minister of Finance and Plan in late July 1994 gave DNIP the authority to approve PIP budget expenditure requests, and to commit expenditures to prefects and chiefs of diplomatic missions for decentralized budgets. Current Budget Procedures 7. Wage payments are managed directly by the DNB and its Wage Division at the Ministry of Finance, and by the DNT in coordination with the Ministry of Civil Service Reform, which submits a list of personnel to be paid that is then matched with records maintained by the Wage Division. The various directorates and divisions direct payments to the line ministries, and distribute salary payments to staff. In the past, this process was subject to abuses. The recent effort by the Ministry of Civil Service Reform and the Ministry of Finance to update personnel files in Conakry required that staff receive their salary payment in person. The process has enabled the authorities to verify that individuals are indeed posted to the particular unit; individuals who do not claim their salary for a period of three months are eliminated from the wage bill. This exercise should be extended to the rest of the country and instituted as a permanent feature of wage bill management. Another wage bill item that is open to abuse by central and line ministries consists of expenditures for paying nonpermanent personnel or for "regularizing" newly hired personnel. Public pensions are managed directly by the DNB and the Treasury, and payments require personal claims from beneficiaries or their family. 8. Approval of nonwage recurrent expenditures require that credit for contracts to render work, goods, or service to line ministries be opened at the beginning of the budget year, or after the budget has been revised at midyear. Line ministries or implementing agencies through submission initiate the payment process by submitting to the DNB a request for the specific amount under the budget 168 credit. After commitment of the amount, the line ministry submits a request for payment, which includes all documentation necessary for verifying the authenticity of the expenditure. The documentation is reviewed by the relevant division of the DNB, which, after providing a visa, submits it to financial control. The Division of Financial Control checks the legal and financial aspects of the transaction and advises the Director of the Budget. Only after verification does the budget director submit the transaction for payment by the DNT. Within that directorate, the request and supporting documents are first passed to the expenditure division for accounting registry, and are then sent to the central payments office. The chief officer of the payments office signs the check for payment and sends it to the treasurer for a second signature. The check is then sent to the central payments office for deposit in the account of the enterprise. The Expenditure Approval Process 9. Normal payment procedures are long and complex. As with the current budget, budget credits are open for expenditures allocated in the BND. They form the basis for signature of purchase bonds or contracts by line ministers. Until recently, a request for authorization to commit expenditures was forwarded by the line ministry or implementing agency to the Minister of Plan and Cooperation. The request, accompanied by supporting documentation, was reviewed by the relevant sectoral staff of the DNIP. Sectoral staff then forwarded the documents to programming staff at the DNIP, which in turn indicated their position about the expenditures. The Minister of Plan then drafted a letter to the Minister of Finance to indicate agreement with the proposal. Authorization to commit and verify the expenditure was then requested by the implementing agency to the DNB and more particularly to the DDI. After review, the DDI indicated its conformity with the documents and forwarded them to the financial control office for verification of legal and financial aspects, and then to the budget director for approval of commitment and verification. The DDI again received the documents after signature and forwarded them to the DNT and to the central payments office for control. The documents were then forwarded to the treasurer, who decided whether the expenditure should be paid. If authorized by the treasurer, the documents were then sent to the expenditure division of the DNT to be registered, and to the central payments office for signature of the payment, and to the treasurer for a second signature. The payment returned to the central payments office, which then checked the availability of money in the Central Bank accounts, and thereafter for deposit in the account of the enterprise. The process took about six months. This cumbersome, lengthy process is changing as the ACGP begins to become more involved in expenditure approval (annex 12). 10. The Loi des Finances provides for a unified budget, consisting of three components: (1) revenue, (2) current expenditures, including public debt, personnel expenditures, operating expenses, and transfers, and (3) investment expenditures. The investment budget itself consists of two items: externally financed projects (FINEX), and the BND, which includes the Government's contribution to the financing of FINEX projects, plus investments undertaken and financed solely by the state. But because the external account deficit is financed largely with grants, grants-financed projects that are managed directly by donors are excluded from the investment budget and are reflected only in the balance of payments. Thus, the public investment budget only partially reflects total public investment in Guinea. The Ministry of Finance (MOF) is responsible for preparing revenue projections and the current budget. The Ministry of Plan and International Cooperation (MOP) is responsible for preparing the investment budget. The entire budget is inclusive in the sense that it 169 accounts largely for Government revenue; some irregularities are associated with expenditures that surpass budgetary allocations. 11. The line ministries provide inputs for the preparation of both budgetary components at the beginning of the budget cycle; donors provide estimates of expected expenditures for externally- financed projects for the investment budget. 170 ANNEX 12 THE ADMINISTRATION AND CONTROL OF LARGE PROJECTS (ACGP) Introduction 1. Concerned about the progressive slowdown in public investment, during the new Government reorganization at the end of 1994, the Executive chose not to address institutional complexities, but to create a new agency -- the Administration and Control of Large Projects (ACGP) -- to help get projects off the ground. Under the direct responsibility of the Presidency, the ACGP has received a broad mandate that includes three primary responsibilities: * The preparation of any given project of more than GNF 1 billion. In coordination with the line ministries, the ACGP hires consultants to carry out feasibility studies, and in cooperation with the Ministries of Plan and of Finance it seeks and negotiates investment financing. It also prepares procurement bid documents for the Interministerial Procurement Commission for Large Contracts; evaluates the technical components of bids on behalf of the Commission; submits technical reports to the relevant financing institutions; negotiates and signs contracts; follows up on the progress of works and the delivery of equipment; and participates in securing the level of annual PIP budget expenditures that are necessary to reach project objectives. * The supervision of operations for restructuring or privatizing public enterprises. In coordination with the Ministry of Finance and the Line Ministry, the ACGP prepares the technical proposals for privatization projects and submits these to the Executive for decision. After authorization of the restructuring/privatization, ACGP supervises the bidding and evaluation process in coordination with the ministries, and presents the results to the Council of Ministers for approval. The ACGP is also represented in the Board of Directors of each public enterprise, thus ensuring the effectiveness of its role in the privatization/restructuring process. * The implementation of projects of less than GNF 1 billion for decentralized authorities or regional communities. The ACGP ensures the technical reception of completed projects in coordination with the ministry in charge of decentralization and the beneficiary community. 2. Because the, organization and overall ethnic representation of the Guinean public administration in Government are more similar to those in Cameroon, it may be useful to review the experience in both C6te d'Ivoire and Cameroon (annex 12, box 12.1). While the DGTx in C6te d'Ivoire was initially able to accelerate the level of investment, it was done at a tremendous cost -- centralizing decision making further and considerably weakening public administration. It also overemphasized large projects to the detriment of smaller, socioeconomically attractive projects. In Cameroon, the DGTC was also unsuccessful, having failed to meet its objectives. As a result, public investment came to a standstill and DGTC had to be dissolved. 171 Annex 12 Box 12.1 COte d'7voire: DGTx and Cameroon: DGTC The organizational structure of the DGTx in C&te d'1voire, supported by a large amount of expatriate technical assistance. substituted for the public administration's preparation and implementation responsibilities. As a result. line ministries and public institutions were severely weakened. It was able to influence the content of the PIP in such a way that preference was given to large projects, sometimes with poor economic rates of return. The priority given to large projects became a stumbling block for the implementation of smaller projects with greater socioeconomic impacts. The DGTx has recently been reorganized and downsized. The DGTC in Cameroon was created in the image of the DGTx. In creating it, the Government sought to alleviate the problems of a large and ineffective public administration and the consequent bottlenecks in the implementation of Cameroon's public investment program. Despite open opposition by line ministries and public enterprises, the DGTC was created as an office of the President, under a large expatriate technical assistance contract to provide the expertise necessary for meeting institutional objectives. The DGTC encountered sustained resistance by all public administration agencies in the process of works. equipment. and services procurement. As a result, public investment almost came to a standstill. The amount spent on DGTC tectmical assistance. investments, and operations proved to be unjustified. DGTC was recently dissolved. 3. Although ACGP management has stated its intention to build an institution in the fashion of the DGTx in C8te d'Ivoire (but with less external expertise, and smaller), the experiences with such a structure in the two countries outlined above give reasonable doubt as to ACGP's capacity to successfully overcome the size of its mandate. It currently has 48 professional staff with experience in most sectors, but it expects to expand to 80 specialized staff. Based on discussions with the various line ministries and public enterprises, the perception is that ACGP will represent yet another structure whose budgetary interests are to be considered, rather than a mechanism for simplifying the process. 4. It may be too early to predict ACGP's future performance. But it is evident that its broad mandate crosses over into the responsibilities of line ministries and of public and decentralized institutions, and has further concentrated decision making at the Executive level. It will also make the budgetary process much more difficult, running counter to institutional strengthening objectives sought by Government in the context of donor-financed assistance. The recent staffing of the ACGP has deprived key line ministries of their most competent and experienced staff, weakening these institutions further. The creation of the ACGP under the aegis of the Presidency will also cause concerns about accountability. In view of ACGP's direct linkages with the Executive Office and the fact that the new Minister of Financial and Economic Control is the President of its Board, the question arises whether the ACGP will be open to scrutiny from financial control institutions. Policy Options 5. The very broad mandate of ACGP deserves special consideration. It should be emphasized that this institution is directly involved in every phase of the investment budget, including in the implementation of small projects for decentralized or regional structures. ACGP also manages the privatization process. Even when taking into consideration that it has hired the most experienced and technically competent staff of the entire public administration, its mandate is too large to be managed comfortably by any one institution. The question that the Guinean Administration should 172 ask is: does ACGP have a role to play, and if so, what should this role be? The Government may consider these questions in terms of the budget process. There are, however, some inescapable realities. First, ACGP is one more institution added to the structures mandated with budget planning and implementation. Devolution of responsibilities to the Ministry of Finance would appear necessary. Second, ACGP reverses the stated Government objective of decentralization through re- centralization of responsibilities for regional investment. This may not be sustainable over the medium term in light of Guinea's emerging democratization process. In analyzing ACGP's third mandate -- privatization -- and upon examination of the adequacy of specialized skills within ACGP, the Ministry of Finance, and line ministries, ACGP would appear be the most qualified to hold this responsibility. 6. Nevertheless, one option could be to give ACGP responsibility for the privatization process. If this were done, consideration would need to be given to: (a) formally redefining its role to cover only privatization of public enterprises; (b) reducing its staff; and (c) utilizing foreign advisory services for implementation of specific privatization targets, leaving internal processes and overall supervision to ACGP. 173 г г CATALOGUERS/FILE DRAFT CONFIDENTIAL Report No: 15147 GUI Type: ER