Report No. 40066-HT Republic of Haiti Public Expenditure Management and Financial Accountability Review (PEMFAR) Improving Efficiency of the Fiscal System and Investing in Public Capital to Accelerate Growth and Reduce Poverty January 25, 2008 Poverty Reduction and Economic Management Sector Unit Latin America and Caribbean Region Document of the World Bank and the Inter-American Develoment Bank, co-produced with the Government of Haiti AND ACRONYMS ABBREVIATIONS AAN AutoritC Aeroportuale Nationale - National Airport Authority AIDS Acquired lmmune Deficiency Syndrome APN Autorite Portuaire National (National Port Authority) BRH Banque de l a RCpublique d’Ha7t1- Haitian Central Bank CAE Country Assistance Evaluation CAMEP Centrale Autonome Mttropolitaine d’Eau Potable - Metropolitan Water Authonty CAS Country Assistance Strategy CARICOM Caribbean Community CCI Cadre de CoopCration Interimaire - Interim Cooperation Framework CDD Community Dnven Development CCRIF Caribbean Catastrophe Risk Insurance Facility CEM Country Economic Memorandum CFAA Country Financial Accountability Assessment CIDA Canadian International Development Agency CMEP Conseil de Modernisation des Entreprises Publiques - Council for Modernisation o f Public U t i l i t i e s CNMP Commission Nationale des Marches Publics -National Commission for Public Procurement CNFL4 Commission Nationale pour l a Reforme Administrative - National Cornmission on Administrative Reform COCCI ComitC Conjoint de coordination de l a mise en oeuvre et du suivi du C C I - Joint Committee for the Implementation and Monitoring o f the ICF CONATEL Conseil National des TClCcommunications - National Council for Telecommunications CPAR Country Procurement Assessment Report CPI Consumer Price Index CSCCA Cour Supkrieure des Comptes et du Contentieux Administratif - Supreme Audit Institution CSA Country Social Assessment DAA Departement des Affaires Administratives - Department o f Administrative Affairs DCPE Direction de l a PIannification et de l a CoopCration Exteme - Directorate of planning and external cooperation DDE Dtpartement des Etudes - Regional Education Departments DDR Programme de DCsarmement, DCmobilisation et RCintCgration - Disarmament, Demobilization and Reintegration Program DGI Direction GCnkrale des Imp& - Tax Administration Authority DPG Development Policy Grant DR Dominican Republic DSA Debt Sustainability Analysis EDH ElectricitC d’Ha7ti - National electricity company EFA-FTI Education For A11 - Fast Track Initiative EGRO Economic Governance Reform Operation EPCA Emergency Post Conflict Assistance EU European U n i o n FER Fonds d’Entretien Routier - Road Maintenance Fund FIAS Foreign Investment Advisory Service FY Fiscal Year GDP Gross Domestic Product GNI Gross National Income GOH Government o f Haiti HIPC Heavily Indebted Poor Counties Initiative HN Human Immunodeficiency Virus HTG Haitian Gourde IADS International Agncultural Development Service ICA Investment Climate Assessment ICF Interim Cooperation Framework ICT Information and Communication Technologes IDA International Development Association IDB Inter-American Development Bank IFAD International Fund for Agncultural Development IFC International Finance Corporation IGF Inspection GCnnCrale des Finances - General Finance Inspectorate IMF International Monetary Fund I-PRSP Interim Poverty Reduction Strategy Paper JSAN Joint Staff Advisory Note LAC Latin America and Caribbean Region LIC L o w Income Country LICUS Low Income Countries Under Stress MARNDR Ministkre de I’ Agriculture, des Ressources Naturelles et du DCveloppement Rural - Ministry of Agriculture and Natural Resources MDA Ministry Departments and Agencies MDG Millennium Development Goals MDRI Multilateral Debt Relief Initiative MEF Ministkre de 1’Economie et des Finances - Ministry of Economy and Finance MENFP Ministkre de 1’Education Nationale et de l a Formation Professionnelle - Ministry o f Education and Vocational Training MJPS Ministkre de l a Justice et de l a SCcuritC Publique - Ministry o f Justice and Public Security MINUSTAH UN Mission for the Stabilization of Haiti MPCE Ministkre du Plan et de l a Cooperation Exteme - Ministry o f Planning and External Cooperation MTPTC Ministkre des Travaux Publics et des Transports et Communications - Ministry o f Public Works, Transports and Communications MTEF M e d i u m Term Expenditure Framework NEPO National Education Partnership Office NGO Non-Governmental Organization NPV Net Present Value ODA Official Development Assistance PEFA Public Expenditure and Financial Accountability program PEMFAR Public Expenditure Management and Financial Accountability Review PER Public Expenditure Review PFM Public Financial Management PIP Public Investment Program PNH Police Nationale d’Haiti - National Police PRGF Poverty Reduction and Growth Facility PRSP Poverty Reduction Strategy Paper RGCP Reglement G6ntrale de l a Comptabilit6 Publique - General Rules o f Public Accounting SDR Special Drawing Rights SMP Staff Monitored Program SNEP Societ6 Nationale des Eaux Potables - National Utility for Potable Water SSA Sub-Saharan A h c a SYSDEP Sysdme d’Informatisation des DCpenses - Expenditure Management System TB Tuberculosis TSS TransitionaI Support Strategy TELECO Telecommunications d’HaTti - Telecommunications utility UCREF Unit6 de lutte contre l e blanchiment de I’argennt - Anti-Money Laundering Unit ULCC Unit6 de Lutte Contre l a Corruption - Anti-Corruption Unit UNAIDS The Joint United Nations Programme o n H I V / A I D S UNDP United Nations Development Programme USAID United States Agency for International Development VAT Value-Added Tax ACKNOWLEDGMENTS T h i s Public Expenditure Management and Financial Accountability Review (PEMFAR) i s a joint Inter-American Development Bank (IDB)/World Bank and Government o f Haiti report. The PEMFAR i s a product o f a year o f teamwork and numerous interactions, and dialogue between I D B N o r l d Bank staff and many people and institutions in Haiti: the Government o f Haiti, the private sector, Non-Government Organizations (NGOs), academia, bilateral and multilateral donors. T w o key events were organized t o launch the PEMFAR. I n December 2006, following an internal IDBrWorld Bank discussion o f the PEMFAR concept note, the team organized a brainstorming session with the main donors involved in Haiti’s development t o identify k e y issues and themes to be developed in the PEMFAR. In January 2006, the PEMFAR team organized jointly with the Government o f Haiti a mission t o agree o n the scope and methodology o f the PEMFAR as w e l l as to deepen the understanding o f issues to be covered in the report. The outcomes o f these two important events help to lay the foundations and the strategic directions of the PEMFAR report. The team would like to thank H i s Excellency Daniel Dorsainvil (Minister o f Economy and Finance, MEF, Haiti) and the Government team for the excellent collaboration throughout the preparation o f the PEMFAR report and the valuable advice and inputs provided to complete the report. The report was completed under the leadership o f Mauricio Carrizosa (Sector Manager, LCSPE). The team also benefited from the advice o f Roberto Tarallo (Sector Manager, LCSFM), Enzo de Laurentiis (Sector Manager, LCSPT), and Errol Graham (Senior Economist, LCSPE). The team i s particularly grateful t o Professor Pierre-Richard AgCnor (Hallsworth Professor o f International Macroeconomics and Development Economics, University o f Manchester, and co- Director, Center for Growth and Business Cycle Research, England) for his technical guidance, invaluable inputs, quality control o f the report, and suggestions throughout the preparation o f the report. The task team leader and author o f the P E M F A R report i s Emmanuel Pinto Moreira (Senior Economist, W o r l d Bank). The IDB team leader o f the report i s Susana Sitja (IDB). The PEMFAR core team includes Sandra Bartels (IDB), Fily Sissoko (Senior Financial Management Specialist, W o r l d Bank, LCSFM), Patricia Macgowan (Senior Procurement Specialist, W o r l d Bank LCSPT), Susana Sitja (IDB), MClanie Xuereb-de Prunele (Assistant Professor University o f Paris I1 and consultant W o r l d Bank), Roberts Waddle (E.T. Consultant, W o r l d Bank, LCCHT). The team l e d by Ernmanuel Pinto Moreira and Susana Sitja, consisted o f Gilles Damais (Consultant, IDB), Robert Cauneau (Consultant, IDB and World Bank), M a r k Paolleti (Consultant, IDB and World Bank), Hernan Pfluecker (Consultant, W o r l d Bank), Hagop Angaladian (Consultant, W o r l d Bank), Stephanie Kuttner (Consultant, W o r l d Bank, SDV), and Gilles Gavreau (Consultant, W o r l d Bank). Valuable contributions were also provided by Mathurin Gbetibouo (Resident Representative, W o r l d Bank, LCCHT), Chingboon Lee (Sector Leader, W o r l d Bank, LCSHD), Ahmadou Moustapha Ndiaye (Lead Financial Management Specialist, W o r l d Bank, LCSFM), Joana Godhino (Senior Health Specialist, W o r l d Bank, LCSHH), Nicolas Peltier (Senior Infrastructure Economist, W o r l d Bank, LCSTR), Samuel Carlson (Senior Education Specialist, W o r l d Bank, LCSHE), Peter Holland (Operations Officer, W o r l d Bank, LCSHD), and M a r y M o m s o n (Country Officer, W o r l d Bank, LCC3C). Peer reviewers were Ctlestin Monga (Lead Economist, DECVP), Blanca Moreno-Dodson (Senior Economist, PREMVP) and E m i l e Finateu (Senior Financial Specialist, AFTFM). Haiti: Public Expenditure Management and Financial Accountability Review Table o f Contents EXECUTIVE SUMMARY .............................................................................................................................. i PART I :ECONOMIC AND FINANCIAL MANAGEMENT CONTEXT ................................................ 1 CHAPTER 1:MACROECONOMIC AND FISCAL CONTEXT ............................................................. 3 1 . 1 STRUCTURE OF HAITI’S ECONOMY AND OF PUBLIC FINANCE .............................................. 2 1.1.1. Economic Structure. Growth. and Poverty ............................................................................................... 2 1.1.2. Public Finance Structure ......................... ................................ ............. 4 1.2 FISCAL POLICY AND ECONOMIC 0 -2006 ........................................................ 14 1.2.1 Fiscal Performance and Improved Economic Outcomes .................................................................... 14 1.2.2 Debt relief and Fiscal Space Issue ......... ........ ..... ...........16 1.3 POLICY RECOMMENDATIONS ............ .................................................................. 21 CHAPTER 2 :PUBLIC EXPENDITUREREVIEW: BROAD TRENDS AND PATTERNS ............... 25 2.1 ANALYSIS OF PUBLIC EXPENDITURE BY FUNCTIONAL CLASSIFICATION ...........26 2.1.1 Shifi in Budget Allocation toward Economic and Social Sectors ...................................................... 26 2.1.2 Analysis o f Actual Expenditures ...30 2.1.3 International Comparison of Fun ..................................... 34 2.2 FACTORS DETERMINING THE STRUCTURE OF PUBLIC EXPENDITURE ............................ 34 2.2.1 Level o f Domestic Resources ............................................................................................................. 34 2.2.2 Influence of External Financing ......................................................................................................... 35 2.3 ANALYSIS OF THE PUBLIC INVESTMENT PROGRAM .............................................................. 36 2.3.1 Structure o f the Public Investment Program....................................................................................... 36 2.3.2 Influence o f Donor Financing on the Public Investment Program ..................................................... 37 2.4 BUDGET EXECUTION ..........................................................................................................................39 2.4.1 Broad Trends o f Budget Execution .................................................................................................... 39 2.4.2 Execution o f the PIP ........................................................................................................................... 40 2.5 POLICY RECOMMENDATIONS ......................................................................................................... 41 CHAPTER 3 : COUNTRY FINANCIAL ACCOUTABILITY ASSESSMENT ...................................... 43 3.1 BACKGROUND ...............................................................................................................................44 3.2 LEGAL AND INSTITUTIONAL FRAMEWORK ....................................................................... 46 3.3 BUDGET PREPARATION ............................................................................................................. 47 3.4 BUDGET EXECUTION .................................................................................................................. 48 3.5 ACCOUNTING AND FINANCIAL REPORTING ...................................................................... 48 3.6 DEBT AND CASH FLOW MANAGEMENT ............................................................................... 49 3.7 INFORMATION SYSTEMS ........................................................................................................... 49 3.8 INTERNAL CONTROLS ................................................................................................................ 50 3.9 EXTERNAL CONTROLS AND LEGISLATWE SCRUTINY ................................................... 50 3.10 INSTITUTIONAL FRAMEWORK AND PROCUREMENT MANAGEMENT CAPACITY 50 3.11 PROCUREMENT OPERATIONS AND MARKET PRACTICES ............................................. 51 3.12 INTEGRITY OF THE PUBLIC PROCUREMENTSYSTEM ................................................... 52 3.13 POLICY RECOMiMENDATIONS ................................................................................................. 52 PART 1 1: PUBLIC EXPENDITUREREVIEW IN PRIORITY SECTORS ............................................ 57 CHAPTER 4 :ECONOMIC SECTOR EXPENDITURE REVIEW ........................................................ 59 4.1 AGRICULTURAL SECTOR .......................................................................................................... 61 4.1.1 Background and sector objectives .................................... 4.1.2 Structure and trends in agricultura ........................................................ 4.1.3 Public Investment Program (PIP) ....................................................................................................... 66 4.1.4 Weaknesses, Issues, and Challenges Public Expenditure Management In Agriculture Sector 4.1.5 Policy Recommendations ......................................................................................................... 4.2 INFRASTRUCTURE SECTOR ..................................................................................................... 72 4.2.1 Background and sector objectives ........................................................................................... 72 4.2.2 Trends and Structure o f Infrastructure Expenditures .......................................................................... 74 Haiti: Public Expenditure Management and Financial Accountability Review 4.2.3 Public Investment Programs in the Infrastructure Sector ................................................................... 80 4.2.4 . Weaknesses Issues, and Challenges in Public Expenditure Management in Infrastructure Sector ... 81 4.2.5 Policy Recommendations ....... 84 CHAPTER 5 :SOCIAL SECTOR EXPENDITUREREVIEW ............................................................... 87 5.1 EDUCATION SECTOR .......................................................................................................................... 88 5.1.1 Background and Sector Objectives ...........88 5.1.2 Trends and Structure o f Education Expenditures ............................................................................... 90 5.1.3 Public Investment Programs in the Education Sector ......................................................................... 95 5.1.4 Identifying Weaknesses, Issues, and Challenges in Budget Process and Public Expenditure Management ................................................................................................................................................... 102 5.1.5 Policy Recommendations and Prionties ..... .............. 5.2 HEALTH SECTOR ................................................................................................................................105 5.2.1 Background and sector objectives ................... ......................................................................... 105 5.2.2 Trends and Structure o f Health Expenditures.. ......................................................................... 108 5.2.3 Public Investment Programs in the Health Sector ............................................................................ 114 5.2.4 Weaknesses, Issues, and Challenges in Public Expenditure Management 5.2.5 Policy Recommendations .......................................................................... CHAPTER 6 :JUSTICE AND SECURITY EXPENDITURE REVIEW .............................................. 120 6.1 BACKGROUND AND SECTOR STRATEGY AND OBJECTNES ............................................... 120 6.2 TRENDS AND STRUCTURE OF JUSTICE AND SECURITY SECTOR EXPENDITURES 123 ...... 6.2.1 Trends in Allocations ................... ................ ................. 6.2.2 Intra-Sectoral Allocations o f Exp ............................................. 124 6.2.3 Economic composition o f expenditure ............ ............................................... 6.2.4 Projected Trends in Police Wage ............................................. 129 6.2.5 Budget Execution Trends ................................................................................................................. 130 6.3 PUBLIC INVESTMENT PROGRAMSIN THE JUSTICE AND SECURITY SECTOR 131 .............. 6.4 WEAKNESSES, CHALLENGES AND RECOMMENDATIONS .................................................... 132 6.4.1 Weaknesses and Challenges ......................................................................................................... 132 6.4.2 M a i n Policy Recommendations ......... ...... . 134 PART 1 11: ADDRESSING GROWTH AND POVERTY CHALLENGES: A MACRO-MODELING APPROACH ................................................................................................................................................. 138 CHAPTER 7 :ASSESSING THE IMPACT OF FISCAL POLICY ON GROWTH AND POVERTY: A MACROECONOMIC FRAMEWORK ................................................................................................ 140 7.1 DESCRIPTION OF THE KEY FEATURES OF THE H A I T I MACRO-MODEL ............. 140 7.1.1 The Supply Side ............................................................................................................................... 140 7.1.2 Consumption and investment ............ . 144 7.1.3 Composition o f public spendi ......................... 145 7.1.4 Taxes and the government budget constraint ................................................................................... 146 7.1.5 Public capital: quality indicators .............. .............................................. 147 7.2 CALIBRATION .............................................. .............................................. 147 CHAPTER 8 : ADDRESSING GROWTH A T T D POVERTY CHALLENGES: FISCAL REFORMS AND AID REQUIREMENTS ..................................................................................................................... 152 8.1 BASELINE SCENARIO, 2007-2015 ..................................................................................................... 153 8.2 POLICY EXPERIMENTS ..................................................................................................................... 153 8.2.1. Increase in Public Investment........................................................................................................... 154 8.2.2 Spending reallocation ... 8.2.3 Tax Reform ........................... ................... 157 8.2.4 Increase in security spending............................................................................................................ 158 8.2.5 A composite fiscal pack ANNEXES .................................................................................................................................................... 165 ANNEX 1 ...................................................................................................................................................... 167 ANNEX 2 ...................................................................................................................................................... 173 Haiti: Public Expenditure Management and Financial Accountability Review ANNEX 3 ...................................................................................................................................................... 179 ANNEX 4 ...................................................................................................................................................... 188 ANNEX 5 ...................................................................................................................................................... 196 ANNEX 6 ...................................................................................................................................................... 205 ANNEX 7 ...................................................................................................................................................... 208 ANNEX 8 ...................................................................................................................................................... 211 Tables Table 1. Composition o f GDP at factor cost, 2000-06 ............................................................................. 2 Table 2. Income Per Capita, Nominal GDP, and Growth Rates, 2000-06 ............................................... 2 Table 3. Composition o f Government Revenue in Percent o f GDP, 2000-2006 ..................................... 4 Table 4. Shares o f recurrent spending categories in total recurrent, FY2000-06 (in percent) Table 5. Composition o f Government Expenditure, in Percent o f GDP, FY 2000-06 ... Table 6. Overall Fiscal Balance (HTG Million), FY2000-2006 ................ Table 7. Structure o f Pledges and M a i n Donors Areas o f Intervention, 2004-07 (US$ Table 8. Donor Development Financing. Commitments 2004-07 v. Disbursements as o f September 2007 (in US$ Million) ....................................... Table 9. ........................... Table 10. otherwise indica Table 11. Table 12. General Trends in Budget Allocations, FY2002-07 (in real HTG Million) Table 3. Broad Trends in Budget Allocations to Priority Sectors, FY2002-07 (real H T G million) ......28 Table 4. Broad Trends in Budget Allocations to Priority Sectors, FY2002-07 (in real HTG Million).. 29 Table 5. Geographical Distribution o f Government Budget by Department (In nominal HTG Million, FY2006lO7) ............................................................................................................................. 30 Table 6. Total Spending by Broad Categories (real HTG Million), FY2002-06 ................................... 30 Table 7. Functional Distribution of Actual Total Expenditures-Payment basis (in real H T G million)3 1 Table 8. Functional Distribution o f Actual Public Recurrent Expenditures-Payment basis (real HTG million) ...................... .................. ..................................... 32 Table 19. Shares o f “Comptes Co for SeIecte ty Sectors (in percent), FY2002-05 .............................................................................................................................. 33 Table 20. Comparison o f Spending o n Priority Sectors. Inpercent o f total budget allocations (excluding debt service) ............................................................................................................................ 34 Table 2 1. Public Investment Program FY2004-05, FY2005-06, and FY2006-07 (inmillion o f real gourdes unless otherwise indicated) .................................................. Table 22. Execution Rates in Percent, FY2002-06 ...... Table 23. Budgetary allocations (in real HTG millions Table 24. Budgetary Allocations Structure (in real HT ..................................... 63 Table 25. Structure o f MARNDR’s Operating Budget (allocations in real terms HTG millions), FY2005-2007 ....... Table 26. Trends in MARNDR’s Expenditures, (in Table 27. Detailed Expenses by Budget L i n e (FY 2005-2006) (Inreal HTG thousand) ......................... 66 Table 28. Detailed Expenses by Budget L i n e (FY 2002-2003) (Innominal HTG thousand). 66 Table 29. Shares in Budget Execution (In real HTG million) .................................................................. 66 Table 30. Investment Budget (PIP), yearly allocations (Inreal HTG million unless otherwise indicated) ........................................................................................................... Table 3 1. Comparison o f Budgets Requested and Allocated (in nominal HTG Table 32. Generated Funds in Ministry o f Agriculture (In real HTG thousand) Table 33. M a i n currents account in Ministry o f agriculture .................................................................... 69 Table 34. Network Infrastructure in Haiti 1990-2004 ................. .................. ........ 72 Haiti: Public Expenditure Management and Financial Accountability Review Table 35. Allocations to the Ministry o f Public Works, Transports and Comniunication (MTPTC) (In million o f real gourdes), FY2001-07 ...................................................................................... 75 Table 36. Infrastructure Sector. Share o f selected items in Recurrent Budget and their execution rate (In percent), FY2002106 ............................................................................................................... 78 Table 37. Infrastructure Sector. Budget Execution (In M i l l i o n o f real gourdes), FY2001106 ................. 79 Table 38. Infrastructure Sector. Recurrent Budget Execution (In M i l l i o n o f real gourdes,), FY2001106 79 Table 39. Resources for PIP in the Infrastructure Sector. Comparison to other Priority Sectors, FY2004105 and FY2005106 (In million o f nominal gourdes) ................................................. 80 Table 40. Donors’ financing under ICF (in million $US unless otherwise indicated), 2004-06 pledges and June 2004-December 2005 realizations ........................................................................... 8 1 Table 41. Allocations to Education Sector (In real H T G million), FY2002107 Table 42. Composition o f Spending, (In real HTG million), FY2002-07. ........ Table 43. MENFP Budget Execution (In real HTG million), FY2002-06............ Table 44. Resources for PIP in the Education Sector. Comparison ro other Prio FY2006-07 (In real HTG million) .......................................................................................... 96 Table 45. Status o f Disbursementversus Commitments, as o f December 3 I ,2005 for all types o f interventions (under ICF and others, in thousand o f US$) ..................................................... 96 Table 46. Donors’ Commitments and Disbursements under I C F and Other Interventions (US$ thousands, unless otherwise indicated) ................................................................................... 97 Table 47. Commitments (2004-07) and Disbursements (as o f December 2005) by Donor, Framework and Areas o f Intervention and Axis .................. 99 Table 48. Donors’ Conlmitments to Financing School Year 2004-05 (in thousand US$) ..................... 101 Table 49. Health Indicators. Comparison Haiti and regionaI averages (LAC, SSA, and LIC) Table 50. Allocations to Health Sector (real HTG million), FY2001/06 ................................. Table 5 1. Financing o f Public Investment Plan, FY2004-05 to FY2006-07 .......................................... 112 Table 52. Health Sector. Budget Execution (real HTG million), FY20021 112 Table 53. Health Sector. Recurrent Budget Execution (HTGT million), F ............................ 1 13 Table 54. Health Sector. Share o f selected items in Recurrent Budget and execution rate (inpercent), ............................. 114 Table 55. 115 Table 56. Donors’ financing under ICF (in million December 2005 ..................................................................................................................... 1 16 Table 57. Allocations to Justice and Security Sector (Inreal HTG million), FY20021 Table 58. Recurrent Budget Allocations to Justice/Security Sector (In real HTG mill Table 59. Evolution o f the intra-sectoral allocation o f recurrent expenditures FY2005-(In real HTG million - excluding aid project, unless otherwise indicated) ................................................. 125 Table 60. Evolution o f the composition o f expenditure (in percent of total exp., excl. aid project), FY2001106 .................................................. Table 61. Justice and Public Security. Credits and exp FY2007 in real HTG thousand - Excluding Table 62. Justice and Public Security. Projected Salaries Expenditure Table 63. Projected JusticeiSecurity Salaries Expenditure and Total Government Salaries, FY2007-09 (in M i l l i o n o f Constant US$). ........... Table 64. Intra-allocation o f credits and expenditures, and execution (In real HTG million), FY2004106 .......................... .......................................................................... Table 65. stice and Security Sector. Average over FY20 real HTG million) ................................................................................................................. 132 Haiti: Public Expenditure Manaqement and Financial Accountability Review Figures Figure 1. Real G D P Growth (1995-05, in percent) ................................................................................... 3 Figure 2. ODA received by H a i t i 1990-2005 (Total net, in constant U S $ million) ................................ 10 Figure 3. Broad Structure o f Budget Allocations in Percentage o f Total Budget (average over FY2005- 07) 26 Figure 4. Shares o f Executive Branch Components in Percentage o f Total Budget FY2004-07 27 Figure 5. Broad structure o f the PIP, i n Percent o f Total Resources ....................................................... 3 6 Figure 6. Agriculture Value Added in 2004" (percent o f total) ............................................................... 61 Figure 7. Economically Active Population in Agriculture in 2004 (percent o f total) Figure 8. Share o f Agriculture in Total Budget in selected countries, 2000-2005 ..... Figure 9. Public Expenditure in Education (percent o f GDP) H a i t i and Selected Comparable Countries, 2006 91 Figure 10. Shares o f Financing o f Education Sector ............................................................. ............... 9 1 Figure 11. Non-Public Sector Enrollment Share in Primary Education (oercent) Figure 12. Per capita total expenditure on health in US$ in LIC in 2003 ........................................... 109 Figure 13. Total Health expenditure in LIC in 2003 (percent o f GDP) Figure 14. Private health expenditure in selected LIC in 2003 (percen 110 .............................. Figure 15. Public health expenditure in selected LIC in 2003 (percent o f GDP) ............................... 110 Figure 16. Allocation Distribution, FY2004-05 to FY2006-07 .......................................................... 1I 1 Figure 17. Figure 18. Figure 19. Figure 20. Government Budget--Overall Structure ................. ........ ................................ ........... ....... 146 ~ Figure 2 1. Baseline scenario, 2007-1 Figure 22. Higher Total Public Inves Figure 23. Higher Total Public Investment and Higher Efficiency of Public Investment, 2007- 15, (Deviation f r o m the Baseline scenario) ....... Figure 24. Reallocation o f spending t o he Figure 25. L o w e r collection cost, higher direct tax rate, 2007-15 (Deviation from the Baseline scenario) 157 Figure 26. L o w e r collection cost, higher d nt in infrastructure, 2007-15 (Deviation f r o m the Baseline scen Figure 27. Higher security spending, 2007- Figure 28. Higher security spending and Elasticity o f Security Spending, 2007-15 (Deviation f r o m the Baseline scenario) .. ...................................................., .............................. ................ 158 Figure 29. Combined shock with lower collection cost, higher direct tax and security spending, 2007- 15 (Deviation f r o m the Baseline scenario) Figure 30. Combined shock with higher expenditure, 2007-15 (Deviation f r o m the Baseline scenario) .............................................................. 159 Figure 3 1. Combined shock with lower collection cost, higher security spending, and direct tax incresing later, 2007-15 (Deviation f r o m the Baseline scenario) ......................................................... 160 Figure 32. Combined shock with lower collection cost, higher security spending, elasticity o f security spending, and direct tax incresing later, 2007-15 (Deviation f r o m the Baseline scenario) 160 Boxes B o x 1. T a x Reforms That Work. The Case o f Ghana and Uganda: What Lessons for Haiti? Box 2. Revenue-Enhancing Measures Introduced since 2004 ................................................ Box 3. Creating Fiscal Space: Old W i n e in N e w Bottles ........................... 20 Box 4. K e y Characteristics o f the Infrastructure Sector ........................................................................... 74 Box 5. Pre-requisite for a Multi-Year Sectoral MTEF............................................................................. 83 Box 6. The Road Maintenance Fund (Fonds d'Entretien Routier, FER) 84 Haiti: Public Expenditure Management and Financial Accountability Review Box 7 . Institutional and Legal Framework 90 Box 8 . Strategy for the Health Sector in H a i t i....................................................................................... 108 Box 9 . M a i n Features o f the Justice and Security Sector ....................................................................... 122 Box 10. H o w Public Infrastructure Can Foster Growth:Lessons f r o m Recent Studies 142 Haiti: Public Expenditure Manaqement and Financial Accountability Review EXECUTIVE SUMMARY I.SUMMARY OF’ M A I N MESSAGES OF THE REPORT H A I T I MADE GOOD P R O G R E S S OVER THE P A S T THREE Y E A R S BUT MAJOR C H A L L E N G E S REMAIN TO ACCELERATING GROWTH AND REDUCING POVERTY 1. After the lost decade 1994-2004, marked b y political instability and economic decline, H a i t i has reformed significantly and revived growth, especially in the past three years. Macroeconomic policies implemented since mid-2004 helped restart economic growth, reestablish fiscal discipline, reduce inflation and increase international reserves. Financial sector stability has been maintained though weaknesses have emerged. Significant progress was also achieved in the implementation o f economic governance measures, mainly in the area o f legal framework, core public institutions and financial management processes and procedures. Notably, basic budget procedures were restored, the public procurement system strengthened, and anti-corruption efforts stepped up. Efforts were also made to improve efficiency and transparency in the management o f public enterprises. This wave o f reforms l e d to renewed confidence and translated into higher growth. Real GDP i s estimated to have grown by 2.3 percent in FY2006, implying an increase o f about 0.6 percent in per capita GDP, compared to -0.2 percent in FY2005. The successful implementation o f i t s stabilization program helped H a i t i benefit from a three year IMF-PRGF supported program. In addition, in Kovember 2006, H a i t i qualified for debt relief under the Enhanced Heavily Indebted Poor Countries (HIPC) Initiative by reaching the decision point under the initiative. 2. The Government’s development strategy articulated under the Extended I n t e r i m Cooperation Framework Document (ICF) presented at the July 25,2006 Donors’ Conference in Port-au-Prince and in the I n t e r i m Poverty Reduction Strategy Paper (I-PRSP) centers around four main pillars: (i) strengthening political governance and promoting dialogue and reconciliation; ( ii)improving economc governance and modernizing the State; ( iii) promoting economic growth, including by maintaining a stable macroeconomc framework; and (iv) improving access to quality basic services, particularly for the most vulnerable groups. In this context, the Preval-Alexis Government intends t o move forward w i t h an ambitious agenda to revive the economy, improve the quality and availability o f basic services and modernize the state. The Government’s development agenda has received strong support from the donor community, forcefully stated during two 2006 donor conferences: July 25 in Port-au-Pnnce (Haiti) and November. 30, in Madrid (Spain). In addition, during the donor meeting held on March 23, 2007 in Washington D.C. (USA), the Donors reaffirnmed their support to Haiti’s development program. 3. These recent political and economic developments open a window o f opportunity to break with Haiti’s turbulent past and create the sound foundations for strong and sustained economic growth and poverty reduction. Yet, as Haiti enters a new phase o f i t s history, the country faces daunting challenges, mainly related to: (i) uncertain security conditions; (ii)limited resource base; (iii) poor physical and institutional infrastructure; (iv) limited human capital; and (v) weak financial management systems. Along these lines, the Haiti C E M identifies inadequate infrastructure, political instability, inefficient bureaucracy and education as key binding constraints t o economic growth. 4. I n such an environment, the development challenge o f more dynamic growth in order to reduce poverty requires bold policy actions across a broad spectrum covering various areas o f Government interventions to: (i) improve security; (ii)expand and improve the quality o f the infrastructure base; (iii)expand the economic base and (iv) enhance human capital. But because o f Haiti’s scarce resources, prioritizing Government interventions i s critical to ensure that public resources are allocated to their best uses. This calls for reforms to improve efjciency o f public spending. However, public expenditure reforms would not b e enough to decisively put H a i t i o n a i Haiti: Public Expenditure Manaaement and Financial Accountabilitv Review strong and sustained growth path unless they are complemented by revenue-enhancing measures. This implies that the country design a coniprehensivefisccl reform package. ACCELERATING GROWTH AND REDUCING POVERTY WILL REQUIRE BOLD POLICY ACTIONS W I T H A STRONG EMPHASIS ON FISCAL REFORMS 5. From 2004 to 2006, Haitian authorities moved quickly in implementing fiscal adjustment policies and structural measures, backed by an IMF-Staff-Monitored Program (SMP), and two successive Emergency Post-Conflict Assistance (EPCA Iand 1 1) as w e l l as the Bank’s Economic Governance Reform Operation I(EGRO I). The Government tightened expenditure controls, mainly through the reduction in the use o f discretionary accounts: the cornptes courants. Quick revenue- enhancing measures led to an increase in Government revenue (see Chapter 1). 6. Yet, the growth-poverty response of these “quick fix” fiscal adjustment policies has been limited. M o r e importantly, the Haiti macro-model shows that if the current macroeconomic trends - which are not bad - were to be maintained, the prospects o f increasing growth would not b e good and the Millennium Development Goals (MDG) o f halving poverty by 2015 would not b e achieved even with foreign aid/GDP ratio o f 5 percent (see Chapter 8). The poverty rate w i t h RavaIlion’s adjusted elasticity would decrease f i o m 55 percent in 2007 t o 52.8 percent in 2015. The model also shows that a compositefiscalpackage would lead to an increase in the growth rate o f real GDP per capita by 0.7 percentage points o n average (deviation fiom the baseline). As a result, the poverty rate (with a growth elasticity o f -1 .O) decreases by 2.6 percentage points in 2015. This means a decline in the poverty rate from 55 in 2007 t o 50.2 in 2015. 7. A major policy lesson from this experiment i s that strong and sustainable growth depends on the scope and quality of the fiscal reforms. Fiscal reforms should target a broad-based fiscal package, which aims at expanding the fiscal space and improving efficiency in the allocation o f public spending. This package would combine: (I) an increase in total public investment, ( ii)a reallocation o f public spending to investment; (iii) a crease in the effective indirect tax rate, (iv) an increase in direct tax rate; (v) an increase in security spending, and (vi) a reduction in collection costs. 8. However, even with the best fiscal reforms, it i s unlikely that Haiti will achieve in the near future the high growth rates required to significantly move the country out o f the poverty- trap. The pay-off o f reforms in general and fiscal reforms in particular in terms o f higher growth performance and lower poverty tends to materialize only gradually. The Haiti macro-model shows that the proposed fiscal package tends to have positive impact o n growth and poverty over time. Foreign aid could play a catalytic role to foster fiscal reforms and help accelerate growth in the short and medium-term. ATTRACTING FOREIGN AID TO REINFORCE FISCAL REFORMS AND BOOST ECONOMIC GROWTH 9. Haiti i s unlikely to accelerate growth and move out of the poverty trap solely on the basis of its own resources. The growth experience over the past few years shows that growth rates barely pulled ahead o f the population growth rate o n a sustainable basis. Growth was volatile, partly reflecting political instability and aid f l o w volatility (see Chapter 1). 10. More aizdpredictable aidflows are needed to boost economic growth and help overcome some of the vicious dynamic circles that lock Haiti in a low-growth/high poverty equilibrium, and improve living standards. During the past decade, aid flows t o Haiti were volatile (see Chapter 1). This led to unpredictable budgetary revenues, limited spending in particular o n infrastructure projects, and fiu-ther constrained growth. The procyclical rzature o f foreign aid in H a i t i over the past decade also restricted the efficiency o f policy tools t o respond t o business cycles. A growth and poverty reducing strategy in Haiti w i l l require more stable aid flows to ensure a continuous .. 11 Haiti: Public Expenditure Manaqement and Financial Accountability Review implementation o f fiscal reforms, and a public investment program. A forward looking approach of development challenges facing H a i t i i s conditional on predictable Government resources. 11. But huge aid flows tend to face issues of limited absorptive capacity o f sectors and risks of wastage of public resources in Haiti. The priority sectors’ budget increased sharply over the past three years, reflecting the large increase in the overall Government budget (see below). However, the execution o f government spending over the past three years was slow, partly as the result o f huge problems o f absorptive capacity in the sectors (micro-issue), lack o f cash f l o w planning and monitoring (macro-institutional issue), and poor budget planning at the central and sectoral levels (micro-macro issue) (see Chapter 4 and 5). IMPROVING ECONOMIC GOVERNANCE AND EFFICIENCY IN THE U S E OF PUBLIC RESOURCES 12. Effective management of aid flows i s critical to ensure that their potential growth- enhancing, poverty-reducing and human development-improving effects do materialize. Improving accountability, transparency, and efficiency in the use o f public resources i s crucial t o ensure that public investment translates into accumulation o f capital and growth. Increasing efficiency o f public investment in Haiti i s directly related t o the ability o f the Government o f H a i t i to improve economic governance. To this end, completing the implementation o f the economic governance reform agenda launched in 2004 is critical to create sound foundations for accountability and transparency o f the use o f public resources. However, as the country moves toward i t s medium-term economic program, this first generation o f economic governance reforms would need t o be complemented by a second-generation o f governance reforms focusing o n in-depth procurement reforms and actions to fight corruption, improve the rule o f l a w and advance judiciary reforms to decisively firm up the ground for efficiency o f public investment: a pre-requisite for higher growth rates. 13. Haiti’s macro-model shows that improved efficiency of public investment (reflecting improved economic governance) leads to higher growth rates and lowers poverty. Simulation results show that the growth rate o f output per capita relative to the baseline value rises t o 0.6 in 2015, whereas it was only 0.2 when efficiency o f public investment i s not accounted for. The higher growth rate o f GDP per capita translates into a lower poverty rate. With a growth elasticity o f - I .O, the drop in the poverty rate reaches 2.5 percentage points by 2015 (relative to the baseline), compared to 1.1 percent drop in the case where efficiency considerations are not taken into account. 14. While increased and stable aid flows are much needed, H a i t i cannot - and, indeed, should not - build i t s growth and human development agenda solely on the basis of an expected increase in foreign aid. The Government must also mobilize domestic revenue. A gradual increase in government domestic revenue would reduce the reliance o n aid flows and increase ownership in the implementation o f the growth and poverty reduction strategy. The issue o f financing the growth and poverty reduction strategy raises the issue o f mutual accountability o f the Government and donors in mobilizing the resources needed t o effectively implement the proposed growth and poverty reduction strategy. ... 111 Haiti: Public Expenditure Management and Financial Accountability Review 11. THRUST, OBJECTIVES, SCOPE AND STRUCTURE OF THE REPORT 15. The main objective o f this Public Expenditure Management and Financial Accountability Review (PEMFAR) i s t o help policymakers in Haiti design an agenda o f policy actions t o accelerate growth and reduce poverty (MDGs). T h e PEMFAR i s a joint response f r o m the Inter-American Development B a n k (IADB) and the W o r l d B a n k to a demand statement formulated b y the Government o f H a i t i t o strengthen the analytical underpinnings o f the ongoing PRSP (also named “National Strategy for Growth and Poverty Reduction”), with the aim o f centering Haiti’s next round o f economic governance reforms and i t s overall agenda around growth and poverty reduction. 16. The strategic focus o f the P E M F A R o n issues o f linkages between public finance, growth and poverty i s motivated by the development challenges facing Haiti, the ongoing international thinking o n strategies for achieving higher growth rates and reducing poverty, and the Government’s ongoing PRSP. 17. The report proposes an exercise integrating the analysis o f a Public Expenditure Review (PER), a Country Financial Accountability Assessment (CFAA) and a Country Procurement Assessment Report (CPAR), referred to here as Public Expenditure Management and Financial Accountability Review (PEMFAR). The PEMFAR provides a diagnosis o f existing systems and capacity in public expenditure management and financial accountability as w e l l as the inter-sectoral and intra-sectoral allocation o f expenditures for key priority sectors, including agriculture, education, health, infrastructure, and justice and secunty. Based o n the findings o f analytical work, the PEMFAR identifies priorities for a medium-term action plan for reforms in public expenditure and financial management. This action plan consists o f a concrete set o f recommendations for priority reform measures and possible expenditure reallocations as w e l l as for required human and financial resources to implement them. The proposed action plan has been prepared jointly with and adopted by the Government. The plan prioritizes and sequences the proposed measures and coordinates the actions and interventions o f the various actors. Each component o f the report focuses o n a key question: P a r t I (Economic and financial management context): are Haiti’s fiscal policies and financial management sound to create the foundations for high growth and reduce poverty? P a r t I1 (Public expenditure review): are allocations o f public expenditure and their execution in priority sectors broadly in line with Government’s development priorities? P a r t I11 (Macroeconomic framework): can fiscal reforms and increased aid flows foster economic growth and reduce poverty? 18. The structure o f the report reflects this design. Part Ipresents the economic and financial management context o f Haiti. Part I1 focuses on the analysis o f sector expenditure reviews covering the five I-PRSP priority sectors - agriculture, infrastructure, education, health, and justice and security. Part I11presents Haiti’s macroeconomic framework and assesses the impact o f fiscal reforms on growth and poverty reduction. Aid requirements are also calculated. 19. Certain aspects o f public finances have been omitted or not sufficiently developed in the report due to the programmatic approach of this PEMFAR. They will be addressed in future work (PEMFAR I1 and 111). They include: (1) Tracking the impact o f public expenditure o n beneficiaries (public expenditure trackmg surveys); (ii) fiscal implications o f decentralization; (iii)geographic iv Haiti: Public Expenditure Management and Financial Accountability Review allocation o f spending; (iv) gender issues; and (v) human resource management and technical assistance requirements. V Haiti: Public ExDenditure Manaaement and Financial Accountability Review 111. SUMMARY OF KEY FINDINGS OF THE REPORT H A I T I ’ S MACROECONOMIC AND FINANCIAL MANAGEMENT CONTEXT I S NOT S O U N D ENOUGH TO PROMOTE GROWTH AND REDUCE POVERTY 20. Despite recent improvements Haiti’s macroeconomic a n d fiscal context remains weak. With a l o w resource base a n d j x e d and quasi-fixed expenditure, the Government has little room to maneuver to use the budget as an effective policy tool to influence economic and social outcomes. 21. International support p r o v i d e d t o the country since 2004 has helped the authorities implement adjustment policies a n d structural reforms. A s the result, H a i t i reached the Decision Point under the Enhanced HIPC. Debt relief granted to Haiti provides s o m e j s c a l space. However, resources freed from debt service relief and associated fiscal space i s relatively limited. Efforts should be made t o expand the tax base and improve eficiency in the use o f public resources to enlarge the fiscal space and increase the Government’s room to maneuver. Moreover, the growth and poverty response o f improved policies has yet t o materialize. There i s a need to design afiscal reform package, the execution o f which would result in higher growth rates and lower poverty. 22. Haiti’s budget structure is dominated by the predominance o f the “executive branch”, w h i c h accounts f o r 95 percent o f t o t a l allocation on average over FY2002-07. The allocation o f expenditures i s highly centralized. About 72 percent o f the FY2006-07 budget was allocated t o the central government while only 28 percent was granted to the ten departments. Moreover the distnbution o f resources between departments seems not to reflect the poverty incidence as the poorest departments were granted less resources in FY2007 budget than the richest ones. Under these circumstances, catching-up o f poor departments may thus be difficult as the budget does not appear to b e an instrument t o address the differences between departments. 23. O v e r the past five years, there has been a shift in expenditures away f r o m the “political sector” t o the “economic” and “social” sectors, reflecting the authorities’ recent efforts t o align expenditures t o economic a n d social objectives as w e l l as the r e t u r n o f donors’ financing t o support investment programs since 2004. The bulk o f the increase in allocations t o “economic” and “social” sectors was directed to growth-enhancing and poverty reducing sectors, including agriculture, transport and communications, education, health, and justice and security. T h i s i s a good budgetary outcome. However, t h i s also signals the fact that the strong influence o f donor financing o n expenditure choices has not been offset by a ‘tfungibility effect”, i.e. by a reallocation o f the Government’s o w n funding to non-priority sectors, as donor financing for prionty sectors became available. Moreover, the structure o f Haiti’s public spending, in particular the investment budget i s mainly determined by foreign aid flows, w h c h are volatile. Planning the execution o f the investment budget thus becomes difficult. 24. T h e Government has made significant efforts to improve transparency a n d accountability in the expenditure management in t h e context o f adhesion to governance reforms. There has been a gradual decline in the use o f comptes courants as a source o f execution o f spending in priority sectors since 2004. However, volatility o f aid flows, limited absorptive capacity, and poor budget planning have resulted in l o w execution rates. D E S P I T E RECENT PROGRESS, H A I T I ’ S COUNTRY FINANCIAL ACCOUNTABILITY AND PROCUREMENT S Y S T E M S FACE S I G N I F I C A N T W E A K N E S S E S AND CHALLENGES 25. H a i t i has made significant progress in strengthening fiscal discipline a n d i m p r o v i n g the efficiency o f i t s p u b l i c financial management (PFM) system in the last three years, in a relatively vi Haiti: Public Expenditure Management and Financial Accountability Review difficult context. A l o n g with achieving macroeconomic stabilization, the Government has strengthened the budget preparation and execution process, increased the transparency o f budget information, strengthened budgetary oversight and intensified i t s anti-corruption efforts. It has adequately reflected in the last fiscal years budget, the policy priorities defined in the Interim Cooperation Framework (ICF) and the Interim Poverty Reduction Strategy Paper (I-PRSP). The budget classification system has been improved both o n the expenditure and the revenue sides. The new budget classification i s based o n the administrative and economic nature o f expenditures, and comes relatively close to meeting international standards. 26. T h e automated financial management system, SYSDEP and i t s progressive implementation in Ministries Departments and Agencies (MDAs) have improved the budget execution process. Financial Comptrollers as w e l l as public accountants are being recruited to strengthen the MDAs and accelerate the budget execution process. The audit reports for fiscal year 2001102 and 2002103 were also completed and transmitted by the Auditor General (Cour Supirieure des Comptes et du Contentieux Administratif- CSCCA) t o parliament. 27. Despite these improvements, there remain significant weaknesses and challenges. A s a result o f the weaknesses (see below), the overall scoring o f the Country, under the PEFA and procurement indicators i s relatively l o w in most o f these areas. 28. T h e budget is not forward-looking. T h e links between sectoral policies and priorities and the budget are weak. The budget i s only based o n projections o f activity during the previous year and i s not rooted in a medium-term expenditure framework. There are a number o f off-budget operations. The profusion o f off-budget accounts and activities severely constrains MEF’s ability to control resource allocation and public spending. A large part o f public spending i s channeled through ministries’ “own resources” collected by the MDAs, and special accounts o f the Treasury. 29. Budget execution i s affected mainly b y the lack of cash flow planning and monitoring and the weak capacity of the line ministries. Budget releases to the MDAs are made based o n the ceilings fixed by the Constitution (one-twelfth o f the annual budget appropriation) and not the effective cash f l o w needs. As a result MDAs have difficulty in executing their activities in the absence o f a budget release based o n their real cash f l o w needs and this arbitrary constraint discourages procurement planning and hampers implementation o f multi-year contracts. The budget execution process i s also laclung a manual. The existing manual i s outdated and does not reflect the recent changes in the Public Financial Management (PFM) system. The deployment o f financial comptrollers to support the MDAs, in accelerating the budget execution process, i s moving slowly. Budget allocations approved for regional departments are made available with significant delays resulting in slow implementation o f activities at that level. 30. F o r m a l coordination mechanism to link aid policies, project and programs to the Country’s priorities and budget needs to be established. A large part o f externally financed expenditures i s executed outside the budget, with donors using their o w n implementation arrangements. This results in poor information f l o w between the spending ministries and the MEF and the MPCE, poor coordination between the development partners and the Government, the lack o f a clear framework for the execution o f those expenditures and the lack o f database related to project based assistance. 31. A sound accounting system, including clear standards and a related automated information system, still needs to be developed. A new accounting framework was adopted recently. It i s expected that a sound accounting system w i l l soon be in place in the Treasury department, remedying one o f the major weaknesses o f the PFM system in H a i t i over the last years. To support the implementation o f this framework, the accounting module o f the computerized Government expenditures management system (Systkme d ‘Irformatisation des Dkpenses - SYSDEP) needs to be developed. Moreover, the link between the different automated financial management software and that used by the MEF to report o n Revenue i s not automatic. In addition to the risk o f errors from data vii Haiti: Public Expenditure Management and Financial Accountability Review transfer, the current system cannot generate a clear and comprehensive reporting o f the financial situation o f the Country. 32. Despite a relatively comprehensive institutional and legislative framework, the internal and external oversight o f the budget is s t i l l not effective given the lack of capacity o f the Institutions. Internal controls are very weak o r not yet functioning. External controls are limited by the capacity o f the Supreme Audit Institution (CSCCA). The role o f the parliament has been limited to the approval o f draft appropriation laws during the past few years. Budget Review Acts as w e l l as annual audit reports o f the CSCCA, when available, were not subjected to parliamentary oversight. 33. The adoption o f the new procurement l a w i s a necessary f i r s t step but not sufficient, by itself, to institutionalize the recent procurement reforms. M u c h still needs to be done in terms o f fleshing out the procedural details o f the reforms and the CNMP’s ability to enforce them in order to ensure that all procuring agencies adhere to the new requirements o f the legal framework. 34. A major challenge to the successful implementation o f the PFM reforms i s the inadequate quality and quantity o f human resources, which has been a primary impediment to public sector efficiency in Haiti over the last years. S I G N I F I C A N T PROGRESS H A S BEEN MADE TO ALLOCATE MORE RESOURCES TO PRIORITY SECTORS, YET EXECUTION OF SPENDING H A S BEEN S L O W 35. Over the past three years, allocations o f resources to p r i o r i t y sectors dramatically increased, mainly driven by huge flows o f external aid provided by the donors into the investment budget since the re-engagement o f donors in 2004. The allocation to the agriculture sector almost doubled in real terms between FY2002-04 and FY2005-07 while that to the infrastructure sector more than tnpled in real terms over the same period. At the same time, the allocation o f public expenditure to the education and health sector increased remarkably by more than 30 percent in real terms o n average over the period FY2002-07 while i t doubled for the health sector over the same period. However, at about 2.5 percent o f GDP, the levels o f allocations to the education sector s t i l l compares poorly w i t h the L o w Income Countries (LICs) average (3.2 percent o f GDP) and even by sub-Saharan Afnca standards (3.4 percent o f GDP). Along the same lines, at 6 percent o f total Government budget o n average over FY2005-07, allocations to the health sector stand below comparable countries: Madagascar (8.4 percent); Dominican Republic (1 0.3 percent); Burkina Faso (10.3 percent); and Zambia (13.2 percent). The justice and secunty sector also recorded a large increase in i t s budget, which nearly doubled in real terms between FY2002-04 and FY2005-07. Yet, the allocations can be considered as relatively l o w since justice and secunty often account for a much larger share o f between 20-30 percent o f public expenditures, especially in post-conflict countnes and fragile states. 36. Leaving aside the issue o f i t s relative levels compared t o international standards, the allocation o f resources to p r i o r i t y sectors i s not the major problem o f public finance in Haiti. The m a j o r issue i s the execution o f spending, in particular the execution o f the investment budget. Less than 30 percent o f the agriculture and infrastructure overall budget was spent o n average between FY2004-06. About 75 percent and 60 percent, respectively o f the overall budget allocated to the education and health sectors were spent on average between FY2004-06. A s a result, the sectors have little influence o n economic and social outcomes. 37. L o w execution rates reflect the combination o f three factors: (i) limited absorptive poor budget planning at the central level (macro-issue) and ( capacity (sectoral issue); (ii) iii) lack of cash flow planning and monitoring (macro and institutional issue). But i t i s also symptomatic of weaknesses in public expenditure management at the sectoral level (which m i r r o r weaknesses at the central level). First, budget preparation does not follow any specific timetable, and budget proposals have not been backed up by a sectoral strategy. They are seldom substantiated by clear ... Vlll Haiti: Public Expenditure Management and Financial Accountability Review justification and spending targets. The result is that proposals f r o m the line ministries often do n o t obtain the MEF’s agreement. In fact, for most o f the sectors, budget allocations are j u s t the renewal o f allocations o f the past year with a factor increase based on neither clear justification nor spending targets. The preparation process i s also affected by the centralized approach adopted within the Ministry, which results in little involvement o f the Ministry’s directions and entities. Second, budget execution in the sector suffers from the lack o f detailed and accurate data. Aggregate execution figures may hide differences in budget items and detailed information o n execution i s scarce and incomplete. Also, the lack o f clear spending targets makes it difficult t o judge the effectiveness o f budget execution in terms o f achieving the objectives set. Only recently has the Government o f H a i t i prepared a list o f poverty reduction items in the context o f the H I P C tracking mechanism. The challenge n o w i s to convert this l i s t into specific budget lines t o ensure the effectiveness o f the tracking and monitoring exercise. Third, governance issues are also a major constraint t o effective expenditure management most notably in the education and health sectors. Lack o f clearly defined procedures, transparency and accountability in the use o f public funds often result in poor expenditure management in these sectors. Fourth, another issue o f the expenditures management at sector level concerns the capacity o f the line ministries to manage better their o w n human resources. THE PREPARATIONOF THE PUBLIC INVESTMENT PROGRAM (PIP) I S A GOOD STEP TOWARD A STRATEGIC BUDGET PLANNING. YET, SIGNIFICANT CHALLENGES REMAIN TO MAKE I T AN INTEGRATED BUDGET TOOL FOR GROWTH AND POVERTY REDUCTION 38. Haiti has prepared a three year PIP. This document which uses a back-of-envelop calculations methodology and a top-down approach i s an important step toward a programmatic budgeting approach. However, several issues would need to be addressed for the PIP t o become an integrated budget planning tool. First, the PIP does n o t account for linkages between various sectors. It has limited links with the sectoral strategies and i s not the product of a coordinated exercise involving all line ministries. Second, aid requirements are determined by totaling resource needs o f each sector. Third, the PIP does not express the development needs o f H a i t i as stated in a coherent poverty reduction strategy such as the I-PRSP. In fact, its l i n k s with the I-PRSP are n o t clearly defined. Finally, the document does not adopt a forward-looking approach, which focuses o n growth and poverty reduction. 39. The real challenge now i s to design a full-fledged integrated PIP, which will adopt a bottom-up approach and account for the linkages between various sectors. The lack o f clearly defined sectoral strategies with spending targets makes the design o f the new PIP more challenging. I t also raises the issue o f the right sequencing o f the preparation o f the PIP. In any event, a pragmatic approach i s needed in the context o f urgent development challenges facing Haiti. ix Haiti: Public Expenditure Management and Financial Accountability Review IV. TOP 7 STRATEGIC POLICY PRIORITIES FOR EFFICIENT USE OF PUBLIC RESOURCES TO ACCELERATE GROWTH AND REDUCE POVERTY IN HAITI 40. Based o n the key findings, the PEMFAR recommends the following seven policy priorities to improve efficiency in the use o f public resources to accelerate growth and reduce poverty. These priorities will be discussed with the Government o f Haiti and the donor community. Expanding fiscal space by simultaneously accelerating the implementation of revenue-enhancing measuYes and improving efficiency of public spending. 41. On the one hand, H a i t i has a l o w resource base. On the other hand, fixed and quasi--xed expenditures take up an important part o f Government revenue. However, there i s scope for increasing the fiscal space b y implementing revenue-enhancing measures and reallocating expenditures in the budget. On the revenue side, direct and indirect tax revenues are much lower than other l o w income countries (LICs). On the expenditure side, savings could be achieved by gradually reducing Government transfers to public entities. 42. Recommended Policy Actions Accelerate implementation of the revenue-enhancing measures agreed upon with the international community (IMF-FAD). In the short-tern, authorities should seek to reinforce tax administration and broaden the tax base. In the longer term, authorities should consider gradually increasing some o f the tax rates that are below regional averages (see IMF’s recommendations). The strategy should also include measures to establish effective customs control through ports o f entry other than Port-au-Pnnce. A comprehensive r e f o r n plan for revenue administration should be prepared in the near future. Measures t o raise revenue could include: (i) eliminating tax incentives in the investment code; (ii) strengthening taxation o f h n g e benefits, for instance by eliminating their deductibility o n the employer side; ( iii)comprehensive coverage o f personal capital income in the tax net; (iv) increasing the holding tax o n property above some relatively high threshold; and (v) increasing specific excise taxes at least in line with the inflation, and increasing existing ad valorem rates.’ Actions to strengthen the existing tax administration could include (i) improving collection procedures; (ii) developing audit plans and procedures; and ( iii) reorganizing along functional lines. 0 Improve the efficiency o f public spending to “kick-start’’ and “sustain” growth to alleviate poverty. Efficient use o f public resources requires that the Government redesign i t s policy transfer to the public entities and alleviate its financial burden. F o r instance, over the past six years, central Government transfers to EDH have increased to 1.3 percent o f GDP, or 8.8 percent o f total Government expenditure. During 2006, with rising international petroleum prices, electricity production in Port-au-Prince declined by about one-third, while fiscal transfers remained broadly unchanged. The savings obtained by reducing transfers to public entities could be directed to produchve spending, which includes: infrastructure, education, health, and security. Efficient use o f public resources requires that the Government focus on these sectors. ’ At 10 percent, the VAT i s at the lower end o f the range among low-income countries. However, increasing the VAT could raise concerns regarding the negative impact on the poor. Thus, the PEMFAR team does not recommend this measure. X Haiti: Public Expenditure Manaqement and Financial Accountability Review Higher growth, in turn, would generate increased fiscal resources to finance productive spending, further bolstering the dynamism o f the Haitian economy. Designing an aggressive policy of attracting and effectively programming grantflows 43. Revenue-enhancing measures would take time to raise Haiti’s revenue-GDP ratio to international levels. In the short term, Haiti would continue to depend on foreign aid to finance its development priorities. However, because o f debt sustainability considerations, contracting additional debt obligations might not be feasible and does not appear t o be a reliable policy option. The Government should thus put in place an aggressive policy o f attracting foreign grants. However, the success o f this policy will depend on the Government’s capacity to improve budget execution so that public resources are effectively used. 44. Recommended Policy Actions 0 Prepare a medium-term strategy for attracting grant financing. The strategy should first identify the financial needs o f the country t o achieve the MDGs. This should be based o n the integrated full-fledged PIP, which would be costed (see below). The sectoral strategies would help provide a costing o f the needs. The strategy should be developed in close collaboration with the donors in the context o f the financing o f the Government’s ongoing medium-term reform program and the PRSP. In light o f the tremendous needs o f H a i t i and the limited resources available, prioritizing the needs would be critical. Focus should thus be o n growth-enhancing and poverty- reducing sectors, including infrastructure, education, health, a p c u l t u r e and justice and security. 0 Set up an institutional mechanism in charge o f grant financing. This mechanism could be created at the level o f the Ministry o f Economy and Finance (MEF). I t could take the f o r m o f the creation o f a unit within the MEF, which would have specific responsibility for grant financing. T h i s mechanism would allow tracking o f currently available grant resources and potential future resources. The successful implementation o f such a mechanism will closely depend on coordination with donors. 0 Establish a “buffer reserve fund” to cope with abrupt reduction in grant financing. The volatility o f aid flows considerably disrupted the implementation o f policy reforms over the past decade in Haiti (see Chapter 1). Setting up a buffer reserve fund would help reduce the negative effects o f aid volatility. The modalities o f this fund (level o f resources to be included in the fund, remuneration, commitment o f the resources, etc.) should be discussed with the donor community. Accelerating financial management and procurement reforms 45. The successful implementation o f PFM reforms requires that the Government takes specific policy actions. 46. Recommended Policy Actions 0 Complete the legal framework for financial management by: (a) supplementing the Decree on the preparation and execution o f budget laws with a clear and exhaustive reiteration, in a specific section, o f all the broad budgetary principles, and (b) incorporating the concept o f performance and results based management in the framework. xi Haiti: Public Expenditure Manaqement and Financial Accountability Review Complete the legal framework for public procurement by: adopting the new Procurement L a w in essentially the same form as the current draft; (b) enacting all application texts needed by the Procurement Law; (c) publishing updated standard bidding documents, including standard administrative and technical clauses (CCAG and CCTG) in the form o f decrees with detailed provisions t o cover the execution phase o f contracts as well as the award phase; and (d) coordinating C N M P and CSCCA reviews/controls and to prevent any conflict-of-interest. Improve budget preparation by: (a) reviewing the sectoral strategies and priorities, linking them with the budget and moving gradually toward an MTEF; (b) strengthening the budget and procurement plans o f the MDAs; (c) improving existing budget classifications; (d) refining the definition o f budget lines containing poverty reduction expenditures; (e) preparing a list o f all off- budget operations and incorporating them into the budget; and (f) preparing an appendix o f revenue and expenditure forecasts for autonomous funds. Improve budget execution by: (a) accelerating the recruitment and deployment o f financial comptrollers in MDAs; (b) completing the survey o f salary arrears; and (c) elaborating and implementing a new accounting and financial framework for expenditures. Improve accounting and financial reporting by: (a) accelerating the reduction o f the use o f comptes courants; (b) including the details o f non-requisition expenditures and comptes courants in the annual financial statements; and (c) clearing the backlog o f State General Accounts and General Account Balance. Improve debt and cash flow management by: (a) appointing a public accountant at the treasury department (TD) to manage debt; (b) establishing an information-sharing circuit between TD and the debt directorate; and (c) establishing a cash planning and monitoring committee. Improve the information systems by: (a) elaborating and implementing a Financial Management Development Master Plan; and (b) improving SYSDEP. Strengthen internal controls by: (a) accelerating the deployment o f financial comptrollers in the M D A s ; (b) defining a harmonized framework for the conduct o f internal controls within technical ministries; (c) hiring and training the General Finance Inspectors; and (c) setting forth in a regulatory text the modalities for using the results o f inspection mission and their publication. Strengthen external controls and legislative scrutiny by: (a) implementing the capacity building program o f CSCCA; (b) developing an action plan to clear the backlog o f audit reports (fiscal years 2003/04; 2004/05 and 2005/06); and (c) transmitting t o Parliament the draft Budget Review Act. Reinforce the institutional framework and procurement management capacity by: (a) reassigning the C N M P dispute resolution function to an independent body such as the CSCCA and expanding C W ’ s activities to include comprehensive data collection and procurement audits; (b) strengthening the procurement pIanning function; (c) mobilizing the resources o f both the C N M P and the civil service training center to improve procurement slulls o f civil servants through expanded training programs; and (d) developing a program o f introductory procurement training for other public sector stakeholders as w e l l as private sector operators and civil society organizations. Improve procurement operations and market practices by: (a) limiting contracts awarded through single source and/or direct contracting to a maximum o f 5 percent o f the total value o f xii Haiti: Public Expenditure Management and Financial Accountability Review contract awards, (b) discouraging the use o f restricted tendering in favor o f national and international tendering (appel d’offves ouvertl, (c) using the existing CNMP Web page t o advertise all government bidding opportunities and increase international advertising to broaden the government’s access t o foreign suppliers, (d) ensuring that contract awards are made public through the C h W Web site and/or publications, (e) requiring procuring entities t o submit t o CNMP, at regular intervals, data o n contract awards below the CNMP review threshold, (0 discontinuing the practice o f awarding contracts based o n “trois pro fonnas’’ and replace i t with either an open national tendering process or the use o f competitively awarded “indefinite delivery contracts” or “standing offer arrangements”, (8) ensuring the quality o f bidding documents, including technical specifications, prepared by procuring entities by instituting periodic prior review o f bid documentation by the CNMP, (h) enhancing the training available to procuring entities on bid evaluation procedures, and (i) developing guidelines for contract administration and building capacity in CSCCA for dispute resolution. 0 Improve integrity o f the public procurement system by: (a) instituting internal quality control in procuring agencies t o promote internal audit o f procurement, develop preventive measures and increase capacity and efficiency; (b) building capacity in the CNMP and/or the CSCCA to organize and conduct external (“independent”) procurement audits; (c) eliminating the conflict- generating activities (issuing an opinion and approving contracts ex-ante) from the CSCCA’s mandate in order t o allow the CSCCA t o focus on their audit activities; (d) empowering the CSCCA to handle appeals as part o f a more robust system for dealing with protests and complaints from suppliers; (e) reactivating the CNMP Web site with more comprehensive information and advertisement o f bidding opportunities; (0 improving the working relationshrp between the CNMP and the ULCC to develop measures to prevent corruption in government contracting; and (g) developing a working relationship between the CSCCA and ULCC to establish a secure mechanism to allow the public at large to report fi-audulent, corrupt or unethical behavi.or without fear o f reprisal. Designing a full-fledged public investment program, focusing on investment in infrastructure 47. The successful implementation o f Haiti’s PRSP requires that the Government prepare a full- fledged and integrated PIP. 48. Recommended Policy Actions 0 Starting in FY2008, the Government should prepare a new PIP, which should adopt a bottom-up approach and be based on sectoral strategies. The new approach should allow line ministries to prepare their sector PIP, adequate to identify investment required, recurrent expenditure anticipated and aid flows required. The sectoral PIPs should then be put together t o produce a Government integrated and multi-year PIP. I t i s critical that the integrated PIP account for the linkages between the sector and not be a simple adding-up o f sectoral PIPs. There i s thus a need to coordinate among various ministries and spending agencies t o establish a coherent, core PIP. Aid requirements might thus be lower and should be identified at the outset o f the design process for the integrated PIP. The integrated PIP should also be a rolling exercise allowing some flexibility t o account for resource constraints. The design o f the integrated PIP should be done in the context o f the revised PRSP (three years fi-om now) given the current timetable to complete the ongoing PRSP. 0 Secure the financing o f the integrated PIP. This would require that all the resources possible are added. In other words, resources freed from debt relief should be complemented by other sources o f financing: the issue o f additionality ofresources. I t should also involve xiii Haiti: Public Expenditure Manaqement and Financial Accountability Review the channeling o f resources through the Government’s budget to ensure that Government has full control o f all resources allocated to execute the PIP. This in turn, implies improved effectiveness, transparency, and accountability in budgetary operations. Allocating more resources to the priority sectors and expanding sector absorptive capacity 49. M o r e resources should be provided to the priority sectors to enable them to deliver qualitypublic goods. However, their capacity to execute huge flows o f resources should b e strengthened. 50. Recommended Policy Actions 0 Increase allocations o f resources t o the education sector, with 50 percent o f t h e resources t o p r i m a r y education. The allocations o f resources t o the education sector should reflect the EFA targets and be based o n the sector MTEF. However, because of the large share o f the non-public sector in the education sector, increasing allocation t o the public education sector would not have m u c h impact unless there i s a strong partnership with the non-public sector. Access to non-public schools stands out as an area where this partnership could b e explored. A possibility that could b e considered i s the provision o f subsidies to poor students currently un-enrolled in school so they could attend n o n public primary schools free o f charge. 0 Increase the allocations t o the health sector, with a focus o n the delivery o f basic health services. A first step for the sector i s to complement the current strategy with specific spending targets and define clearly budget allocation priorities. 0 Increase the allocations t o t h e infrastructure sector by focusing on investment spending w h i l e securing t h e financing o f r e c u r r i n g costs associated with new investments. This means that the sector should first prepare a full-fledged costed and integrated infrastructure strategy with sub-sectoral actions plans, which are linked to spending targets. 0 Increase the resources provided t o the agriculture sector by focusing on allocations t o specific items, including fuel and lubricants, fertilizers, plant material and seed, hydraulic works, insecticides, etc. Increase allocations t o t h e justice/security sector so as t o gradually bring the allocations t o international levels observed in post-conflict countries and fragile states. T h i s requires that the sector develops a hll-fledged integrated strategy with specific spending targets. 0 Augment the capacity o f sectors t o absorb resources by reinforcing technical, human a n d administrative capacity. A specific focus should be o n strengthening the sectors’ ability t o comply with procurement procedures. Designing a one-year technical assistanceprogram in quantitative macroeconomic modeling 5 1. Quantifying the impact o f fiscal reforms (in general) and public spending (in particular) requires developing a quantitative macroeconomic framework, w h i c h accounts for the linkages between public investment, growth and poverty. 52. Recommended Policy Actions Prepare a comprehensive technical assistance program, which identifies the technical needs in basic macroeconomic modeling in the Ministry o f Economy and Finance (MEF). xiv Haiti: Public Expenditure Management and Financial Accountability Review Install a multi-sectoral team, comprised o f a few technical staff members o f the MEF, Ministry o f Planning, and various ministries, including education, health, infrastructure and justice and security. 0 Exploit the ongoing preparation of the P R S P to reinforce the synergies between the proposed technical assistance in macroeconomic modeling and technical needs to complete the PRSP. Improving human resource management 53. The successful implementation o f the public finance reforms requires that institutions be adequately staffed with appropriate qualified and motivated personnel. 54. Recommended Policy Action Strengthen the capacity o f the Human Resource Unit. This will enable the Unit to achieve i t s mandate notably by establishing and implementing a new framework for human resource management, including qualifications, slulls assessment and a capacity building program, developing a transparent and merit-based procedure for n e w recruits and promotions. 0 Design a capacity building program t o reinforce human resources for improved budget management. This program should be designed in close collaboration w i t h donors to secure i t s financing. xv h % 0 3 YI . Y cd 0 cd e $ Y 0 % ’ V !i I B e, E .3 0 > Y X m e, $ 5: i?2 cu 0 !3 .r( Y cd 0 0 3 Q 2 g . 0 . . . . . 0 . . H .. E a 3 m s 2 VI 2 0 v) .I v) e, 0 2 E 2. .I .3 A 2 & . . a L3 v) - G W - 1 P % 3 .9 2 .3 L e, h 2 .* Y .* 3 & % bn c1 - .I e, 0 0 Y e v) e, .* a 0 5 - 0 e, cd E a $ a rci 0 g i2 a 3 m 5 Ui . 2 e e, g Y . 3 3 .3 m 0 5 Y E 5 a f$ .* 3 k b rc 0 x i% 3 U m U .3 3 .3 e, Y 3 M % a) G a Y M a 0 s 0 P % 9 e E Y P V, v) % E e, $- E Y 2 v) ? 3 f$ Y z w 4-r 0 M e, U 2 '3 e, E 4-r 0 x .d X x x Haiti: Public Expenditure Management and Financial Accountability Review 1 Haiti: Public ExDenditure Manaaement and Financial Accountabilitv Review 2 Haiti: Public Expenditure Manaqement and Financial Accountability Review CHAPTER 1 : MACROECONOMIC AND FISCAL CONTEXT 1.1. This chapter reviews the macroeconomic and fiscal context o f Haiti. I t first describes the economic and public finance structure o f Haiti, focusing o n economic growth and the structure o f Government revenues and expenditures. I t then discusses recent fiscal performance and improved economic outcomes. Finally, the chapter analyses the issue ofJiscal space associated with recent debt relief provided t o Haiti in the context o f the Enhanced HIPC Decision Point. 1.2. The k e y findings o f the chapter can be summarized as follows. First, Haiti’s economy i s undiversified and largely dominated by the agriculture sector, which makes up more than one- quarter o f the country’s GDP. Growth performance over the past decade has been below regional averages and also below the performance o f Afkican countries. Partly as a result, Haiti has remained the poorest country in the Latin America and Caribbean region and amongst the poorest in the world. Second, the public finance structure i s characterized by a l o w revenue base. At 10 percent o f GDP, Government revenues are among the lowest in the world. At the same time, the Government faces fixed and quasi-fixed expenditures, including wages and salaries, interest payment o n its debt, subsidies and transfers t o state-owned enterprises and welfare financial obligations. Put together, this leaves the Government with little room to maneuver and to use the budget as a policy tool t o influence economic and social outcomes. Third, in the face o f its l o w resource base and huge development challenges, Haiti has relied o n external assistance t o finance i t s development needs. Aid flows, provided by a limited number o f donors, tended t o be volatile, reflecting political developments over the period from 1990 to2004. In addition, while large amounts of resources were committed under the ICF, disbursement o f aid flows has been slow. Fourth, since 2004, fiscal policy and economic outcomes have improved as the Government moved quickly in implementing adjustment policies and structural reforms supported by the donor community. However, the growth and poverty responses o f improved policies are yet to materialize. Fifth, improved macroeconomic stability and successful implementation o f its reforms program helped H a i t i reach the Decision Point o f the Enhanced HIPC Framework. Debt relief granted to Haiti provides some fiscal space. However, resources freed-up from debt service relief and associated fiscal space, are relatively limited. Efforts should be made to expand the tax base and improve efficiency in the use o f public resources to enlarge fiscal space and increase the Government’s r o o m t o maneuver. 1.3. Policy recommendations center on: (i) accelerating implementation o f macroeconomic stability policies and structural reforms; (ii) implementing revenue-enhancing measures and a more aggressive policy t o attract foreign grants; (iii) strengthening the management o f external debt t o maintain debt sustainability; (vi) improving efficiency o f public spending t o “luck-start” and “sustain7’ growth to alleviate poverty; and (v) designing and implementing a multi-year and integrated public investment program t o be supported by foreign aid flows. 3 Haiti: Public Expenditure Management and Financial Accountability Review 1.4. Political and economic instability, low growth and high levels o f inequality and poverty have been persistent challenges confronting Haiti. The impact o f protracted political conflict, withdrawal o f international economic support, natural disasters and weaknesses in economic governance and management has been severe. Real income per capita has declined o n average by 2 percent annually over the past 20 years. With more than h a l f o f the population subsisting o n less than US$1 a day, H a i t i i s the poorest country in the Latin America and Caribbean region and among the poorest countries in the world. There are marked inequalities in access t o productive assets and public services*, which tend to hamper physical and human capital accumulation, thereby perpetuating l o w growth and poverty. 1.5. The return to democracy in 2004 brought about hope. F r o m 2004 to 2006, the Transitional Government made significant progress in improving governance and transparency in public sector operations, and restoring economic progress. The Alexis-Preval Government that took office in June 2006 after free and open elections moved quickly to restore fiscal order and to elaborate a poverty reduction strategy. There have been notable improvements both in terms o f economic growth and fiscal performance. Real GDP grew by 1.8 percent in FY2005 and further by 2.3 percent in FY2006 compared with a contraction o f 3.5 percent in FY20043. A modest increase in Government revenue and a decline in public expenditures reduced Government’s deficits. Increased donor confidence and the resumption o f external assistance helped finance a significant increase in capital expenditures. 1.6. Nevertheless, Haiti’s poor infrastructure base, limited natural resource base, limited human capital, weak financial management systems, as well as uncertain security conditions remain binding constraints to growth and poverty reduction. A more diversified economic base, improved core infrastructure and major efforts in fiscal management are essential for sustained growth and poverty reduction - the Government ’ s overarching development objectives. Moreover, effective use o f domestic resources and external assistance will be essential to meet the country’s daunting challenges and create conditions for a long-term economic recovery and improvement in social sector services critical t o the success o f the country’s poverty reduction strategy. Nearly h a l f o f national income goes t o the richest 10 percent o f the population. The year 2004 was characterized by the armed conflict that ended the Aristide government and by a process o f establishing the transition government. 4 Haiti: Public Expenditure Management and Financial Accountability Review 1.1 Structure of Haiti’s Economy and o f Public Finance 1.1.1. Economic Structure, Growth, and Poverty 1.7. H a i t i i s essentially a rural and undiversified T economy. Agriculture generates more than one quarter o f GDP and employs two-thirds o f the economically active population. Industry The small amount o f available Services 55 55 55 55 57 57 58 technology, the l o w level o f Source: Government Statistics, World Bank Database and staff capitalization of Haitian Calculations. producers, the insufficiency o f *Estimate of February2006. infrastructure in rural areas, the poor condition of agricultural Table2. Income Per Capita, Nominal GDP, and Growth roads, as well as land insecurity Rates. 2000-06 have limited I agriculture’s potential. About 80 percent o f 2000 2001 2002 2003 2004 2005 2006* GN1percapita(US$ 490 460 430 390 370 400 480 small farmers cannot satisfy the Atlas Method) basic food needs o f their families GDP(ln 79 8 85.7 94.0 119.7 140.3 168.0 196.2 and rely o n several other sources billion of gourdes) of revenue, including Nominal GDP (in 3.8 3.6 3.5 2.9 35 43 4.7 remittances, the sale o f seasonal billion of dollars)’ manpower in other regions o f the RealGDPGrovlth(in 0.9 -1.0 -0.3 0.4 -3.5 18 2.3 country and in the Dominican percent) Republic, and small business. Source: Government Statistics, World Bank Database and staff The service sector accounts for Calculations. about 60 percent o f GDP and *Estimate of February 20 employs about 40 percent o f the wo&force. Activities in this sector are largely concentrated in retail and wholesale trading, exports and basic public services. Industry, which i s undiversified, has been flat and accounts for 16 percent o f GDP (see Table 1). I t comprises mainly small-scale local manufacturing, construction and electricity. 5 Haiti: Public Expenditure Management and Financial Accountability Review 1.8. Growth Figure 1. Real GDP Growth (1995-05, in percent) performance over the past decade has been lackluster, and below regional averages and performance of African countries. When compared t o Afkican countries o f the same level o f development, Haiti fares below the slow growing African countries and far below the sustained growing African economies (see Table A1.2 in Annex 1). Its average growth rate o f 0.9 percent Source: World Bank Database and Staff Calculations. over 1995-2004 i s 11 times lower than that o f Rwanda (the fastest growing African country), 2.6 percentage points lower than that o f Niger (a slow growing country) and below that o f Central A h c a n Republic (a post-conflict and fkagde state). I n addition, growth has been very volatile as year-to-year changes in Haiti’s growth rates have been caused largely by uncertain secunty conditions, volatile donors’ assistance, and weather conditions and their effects on economic activity (see Figure 1). Despite the respectable growth levels achieved during the 1970s and recently 2005 and 2006, per capita income today i s only about US$480, nearly the same level as in 1990. 1.9. Partly as a result of poor growth performance, H a i t i has remained the poorest country in the Latin America and Caribbean region and amongst the poorest in the world. The 2006 United Nations Human Development Index ranked H a i t i 154th out o f 177 countries. About 54 percent o f Haiti’s population lives below the US$1 a day poverty line and 78 percent below US$2 a day (2001 data).4 Social indicators are also abysmal. Although adult illiteracy decreased from 60 percent in 1990 t o 52 percent in 2003, it remains the highest in the Latin America and Caribbean region. Only 55 percent o f children 6-to12 years o f age are enrolled in school. Quality i s also a major concern. Food deprivation and limited access t o health care, due to poor infrastructure and lack o f qualified personnel and drugs, has resulted in dire health conditions for Haiti’s poor. Haiti also faces a high incidence o f HIV/AIDS: U N A I D S estimates that 5.6 percent o f the adult population has HIV. 1. I O . Ensuring strong and sustained growth to reduce widespread poverty and improve social conditions stands out as a critical issue facing Haiti. I t requires continued improvement in security conditions, more and predictable aid flows, a large public investment program t o rehabilitate and expand infrastructure, and efficient allocation and execution o f Government resources, a critical challenge for Haiti. 4 Estimates based o n household surveys suggest that poverty and inequality rates m a y have n o t changed substantially over the last two decades. Part of the explanation could be that while G D P per capita declined, consumption levels were maintained by remittances which have accelerated since the mid-1990s. See H a i t i Country Economic Memorandum, 2006. 6 Haiti: Public Expenditure Management and Financial Accountability Review 1.1.2. Public Finance S t r u c t u r e I .I .2.I . Government revenue structure 1.11. About 95 percent of Haiti’s total Government revenues are derived from taxes, Table 3. Composition of Government shared between taxes on domestic activity Revenue in Percent of GDP, 2000- (taxes on goods and services and income taxes), 2006 which account for about 65 percent o f total revenue, and taxes on international trade, which make up 30 percent. At 10 percent o f GDP in 2006, government revenues are among the Total revenue 8.3 9.6 lowest in the world (see Table 9 below). Tax Current revenue 8.3 9.6 revenue in Haiti i s much Eower than the levels Domestic taxes 5.8 6.3 observed in other low-income countries because o f Customs duties 2.2 2.7 the country’s narrow tax base, l o w yield from Other current revenue 0.3 0.5 income, excise and trade taxes, as well as weak Transfers from public ... ... enterprises tax administration. Table Io f Annex 1 shows that Grants 0.2 3.0 the VAT tax rate i s one o f the lowest in the region. Budget support ... 0.9 In addition, collection issues hamper tax Project grants ... 2.1 performance. For instance, with tax rates similar Source: Government and IMF Statistics, 2007. to Haiti, Guatemala and Paraguay obtain larger collection percentages; the Dominican Republic with a lower rate, i s able to collect more. The Corporate Income Tax (CIT) rate i s close to regional standards but revenues from the CET represent only 0.85 percent o f GDP, much lower than the 2 percent collection in low-income countries. . Moreover, Haiti grants numerous fiscal concessions and tax holidays are in place. For Haiti’s investment code grants fiscal and customs’ holidays o f between 15 and 20 years. 1.12. Given Haiti’s current revenue structure and weaknesses in the tax administration system, the priority for the authorities i s to reinforce the tax administration and broaden the tax base. Revenue-enhancing measures introduced since 2004 have resulted in only a modest increase in the ratio o f revenue to GDP. However, there i s a potential for catching-up. Further increase in revenue would depend on the ability o f the authorities to tackle the issues o f weak tax collection, excessive tax exemptions and a narrow tax base. A 2005 IMF-FAD report has highlighted several channels that the Government could explore for increasing the tax/GDP ratio.5 M M o r e recently, the I F has provided policy recommendations to increase revenue performance in Haiti. Along these lines, in the short-term, the focus should be o n improving tax administration and broaden the tax base. In the longer term, the authorities should consider raising some o f the tax rates to regional levels. The strategy should also include measures to establish effective customs control through ports o f entry other than Port-au-Prince. A comprehensive reform plan for revenue administration should be prepared later. Measures to raise revenue could include: (i) eliminating o f tax incentives in the investment code; ( ii)strengthening taxation o f h n g e benefits, for instance by eliminating their deductibility o n the employer side; (iii) comprehensive coverage o f personal capital income in the tax net; (iv) increasing holding tax on property above some relatively high threshold; and (v) increasing specific excises, at least in line with the inflation, and 5 International Monetary Fund. Haiti Selected Issues. April 29, 2005. 7 Haiti: Public Expenditure Manaqement and Financial Accountability Review increasing existing ad valorem rates.6 Actions t o strengthen the existing tax administration could include (i)improving collection procedures; (ii) developing audit plans and procedures; and ( iii) reorganizing along functional lines. The extent t o which these reforms can translate into a higher revenue/GDP ratio in H a i t i i s not known as the potential impact o f these suggested reforms has not been quantified. I t requires a general equilibrium framework, to account for interactions between taxation, consumption, and growth. However, the experience o f countries such as Ghana and Uganda suggests that well-managed tax reforms can raise tax revenues significantly (see B o x 1 below). Box 1. Tax Reforms That Work. The Case of Ghana and Uganda: What Lessons for Haiti? In Ghana, the tax revenue/GDP ratio increased by almost 10 percentage points from the early 1980s to the late 199Os, reaching i t s current level o f about 16 percent o f GDP. In Uganda, the revenue/GDP ratio rose from about 7 to about 11 percent during the 1990s. These gains were underpinned by policy reforms that included: (i) lowering o f top marginal income tax rates for persons and corporations to reduce incentives for tax evasion; ( i i ) broadening the tax base - for example by eliminating tax incentives for corporations and including fringe benefits i n the personal income tax base; and ( iii) expanding the base for indirect taxes, most notably the VAT. On the administration side, reforms included: (i) the creation o f autonomous tax and customs administrations; ( ii) reorganization o f the tax and custom administration departments along functional lines (payment, enforcement, audits, rather than b y type o f tax); (iii) computerization; and (iv) improvement o f taxpayer registers. 1.13., Although we do not know with much certainty how much o f an impact these measures have had on growth (and thus on the tax base) in these countries, it i s worth pointing out that growth effects will depend in general on how the increased revenue itself i s In that sense, ensuring that tax reforms “work” requires a careful analysis o f thei behavior (saving and investment decisions) as w e l l as the allocation o f revenues productive and unproductive spending components. The “net growth effect” o f tax reform also ’ depends on spending allocation. 1.14. Notwithstanding these remarks, in the short-term, the Government has taken the commitment to gradually increase direct and indirect tax revenues to raise yields closer to the levels in other low-income countries. The authorities’ strategy will initially focus o n establishing effective customs control through entrypoints other than Port-au-Prince. I .1.2.2. Trends in Central Government Expenditures 1.15. Broad structure and trend of expenditure. T h e composition o f government expenditures i s dominated by a large share of recurrent expenditure. They accounted for more than 73 percent o f total expenditure leaving less than 30 percent for capital expenditure over FY2000-06. However since 2004, there has been a gradual shift toward more capital expenditure in Haiti’s budget, reflecting mainly the increase in donors’ financing o f capital expenditure. The share o f capital expenditure in total expenditure increased gradually over FY2000-06: fi-om an average o f 22 percent in FY2000-03 t o 32.4 percent (10 percentage point increase) in FY2004-06. In the meantime, the share o f recurrent spending shrunk from an average o f 78 percent o f total expenditure in FY2000-03 t o 67.6 percent in FY2004-06. Yet, accounting f o r about 9 percent o f GDP on average over FY2004-06, recurrent spending still dominates the expenditure structure 6 At 10 percent, the VAT i s at the lower end o f the range among low-income countries. However, caution should be taken regarding the increase in the VAT rate given the regressive nature o f t h i s tax. 8 Haiti: Public Expenditure Management and Financial Accountability Review while capital expenditure (excluding foreign-financed outlays) has been relatively modest at less than 5 percent o f GDP over the same period (see Table 5 below). 1.16. Economic composition of spending. T h e past six years were characterized b y continued decline in the share o f government wage bill and an increase in other current expenditure (see Table 4). The share o f wages and salaries in total current expenditure declined from an average o f 44.3 percent in FY2000-03 to an estimated average o f 35.6 percent in FY2004- 0 6 (see Table 4 below). At about 3.2 percent o f GDP in FY2006, the government wage bill i s below the levels o f 6-7 percent o f GDP for other low-income c ~ u n t r i e s The . ~ decline in the wage bill was particularly pronounced after 2000, reflecting the freeze o n nominal wages in the public sector. However, the public sector wage bill takes up a significant portion o f the Government overall budget in Haiti, reflecting very l o w overall revenue and expenditure levels. I n FY2006, total spending o n wages and salaries was 32.2 percent o f central government current revenues and 34.1 percent o f government current expenditures, ratios higher than the average for PRGF-eligble countries (about 28 percent and 26 percent, respectively). A comparison o f public sector salaries to those o f the private sector reveals that public sector salaries are significantly lower than private sector averages. A recent survey conducted by the Prime Minister’s office showed that employees in higher skill managerial positions earn 40 percent o f their private sector counterpart’s salaries, while employees in lower s l u l l positions earn 70 percent. The l o w level o f public sector wages raises doubts about the govemment’s ability t o attract and retain quality personnel. M o r e generally, the small size o f the c i v i l service raises questions about the ability o f the government to deliver essential public services, including health, education. infrastructure and assurance o f public order. I t also raises the critical issue o f generating more domestic revenue to eventually finance a more efficient and better paid civil service. 1.17. Interest payments on the debt account for 8-18 percent of total recarrend. spending over the period FY2000-06 (see Table 4). Together with wages and salaries, interest payments make up about 45 percent o f total government recurrent expenditure on average over FY2004-06. These “$xed and quasi--xed expenditure” (about 8.2 b i l l i o n gourdes, or US$194.2 m i l l i o n in FY2006)8 imply that the government has little scope t o shift expenditure from one category to another. At the same time, capital expenditure i s mainly foreign-financed (see below). Put together, this leaves the Government with little room to maneuver and t o use the budget as a policy tool to influence economic and social outcomes, unless foreign aid i s geared toward achieving the Government’s own public investment program (PIP). In other words, the current structure o f Haiti’s budget provides little r o o m to create fiscal space. The issue o f fiscal space i s discussed below in the context o f HIPC and MDRI debt relief. 7 I t was estimated that for PRGF eligible countries, the average central government wage bill represents about 6-7 percent o f GDP. See IMF Haiti: Selected Issues. 2005. 8 The estimated exchange rate at end-2006 i s US$1=42 gourdes. 9 Haiti: Public Expenditure Manaqement and Financial Accountability Review 1.18. The share of expenditure on goods and services in total current expenditures declined from a n average of 38.4 percent in FY2000-03 to 4* Shares Of recurrent Vending categories in total recurrent, 31.6 percent in FY2004-06 (see Table 4). This reflects the Government's recent efforts to contain current expenditure and improve expenditure management procedures. I t i s worth mentioning that the Government o f Haiti made significant efforts to reduce the proportion o f expenditures channelled through discretionary ministerial accounts, so-called comptes c o u r a n t ~ . ~This has signalled the Government's strong commitment to improve expenditure 'Ontro17 and tackleSources: Government statistics, IMF, and Staff weaknesses in expenditure management. calculations. 1.19. I n the meantime, subsidies and transfers have been increasing since 1999, reflecting subsidies to state-owned enterprises and increasing welfare obligations o f the Government. Their share in total current expenditures tripled from an average o f 7.8 percent in the period FY2000-03 to about 24 percent in FY2004-06 (see Table 4 above). Some expenditure overruns (mostly electricity-related transfers) resulted in a further increase in this category o f expenditures, with i t s share rising t o 30 percent o f total current expenditures over the past two fiscal years." I t i s worth noting that the electricity sector has been a significant drain o n Government resources, through funds provided to the state electncity company EDH, payments for private electricity generation and purchases o f fuel for generators. Over the past six years, central Government transfers to EDH have increased to 1.3 percent o f GDP, or 8.8 percent o f total Government expenditure. During FY2006, with rising international petroleum prices, electricity production in Port-au-Prince declined by about one-third, while fiscal transfers remained broadly unchanged. H o w t o reallocate Government expenditure from transfers and subsidies to public sector enterprises toward spending in key areas such as health, education, infrastructure, and security remains a challenge for the Haitian authorities. 9 T h e percentage o f non-salary current expenditures disbursed through the comptes courants was 6 2 percent over the period October 2003-March 2004. However, this percentage was reduced to less than 10 percent in FY2005106. loSubsequently, the government amended the budget to contain expenditures. 10 Haiti: Public Expenditure Manaqement and Financial Accountability Review 1.20. Capital expenditure i s dominated by externally financed expenditures (see Table Table 5. Composition of Government 5). About 75 percent o f capital expenditure was Expenditure, in Percent of GDP, FY financed by foreign assistance during 2000-06 FY2005/06. Government’s capital spending was volatile, reflecting in part the difficulties in establishing a multi-year public investment program under the conditions o f unstable external financial assistance.” 1.2 1. Government expenditures o n priority sectors. Chapter 4, 5, and 6 provide a detailed analysis o f the trends o f government expenditures in these sectors. At this point, i t i s Transfers and subsidies 0.7 2.3 Capital expenditure 2.5 4.4 worth noting that government expenditure on education, health, infrastructure, and security are 10w.I~Budget allocation for the Sources: Government statistics, IMF, and Staff education sector stood at about 2.5 percent o f calculations. 1 Includes statistic discrepancy. GDP in FY2006/07, the lowest in the LAC region (average o f 5 percent of GDP) and l o w by sub-Saharan Africa standards as w e l l (average o f 3.9 percent o f GDP). l3The l o w level o f direct government spending in the social areas combined with inadequate physical infrastructure explains why H a i h has some o f the poorest social indicators in the world. At less than 2.7 percent o f FY2006/07 GDP, allocations t o the health sector are the lowest o f the L A C region, below the regional average o f 3.3 percent o f GDP. However, considerable efforts have been made by the Government to increase the allocations to social. sectors in the FY2007 budget (see Chapter 5 below). Public infrastructure, especially roads, has suffered from lack o f maintenance over the years. According t o the ICF findings, only 5 percent o f the network i s considered in good condition and 15 percent in acceptable condition. The Ministry o f Public Works, Transport and Communications in charge o f the road network does not have the required resources - both human and financial - to ensure adequate road maintenance. Road maintenance expenditures have been falling dramatically over the past years so that less than 10 percent o f the network i s maintained o n a regular basis. 1.1.2.3. Fiscal Balance and Structure of Financing 1.22. On the one hand, fiscal revenues are too low to finance much-needed public services and ensure fiscal sustainability. On the other hand, the expenditure structure i s dominated by “fixed and quasi-fixed’’ incompressible spending, which does not allow much maneuver room to the Government. Haiti faces a structural Jiscal balance problem. The overall fiscal deficit has continuously deteriorated over the past decade, reflecting poor fiscal management and the difficulty to match scarce resources with pressing social needs. At an average o f 4.3 percent o f GDP over FY2004-06, fiscal deficits (excluding grants) have been higher than the averages observed in LIC. However, recent efforts to increase revenues, contain expenditures, combined 11 I t i s worth noting that authorities currently are preparing a public investment program that w i l l center o n supporting infrastructure and will be coordinated with donor assistance. ‘* The l o w level o f Government expenditures o n education and health should be interpreted with caution since Haiti traditionally has relied on more than many other low-income countries o n private fmancing and private sector provisions o f social services such as education and health. l3 See Education For All in Africa: Paving the Way for Action 2005, UNESCO BREDA. 11 Haiti: Public Expenditure Management and Financial Accountability Review with large flows o f foreign grants resulted in improved fiscal deficits (including grants) over the past three years (see Table 6 below). 1.23. Because of the difficulty in accessing foreign financing, the banking sector served as the main source of finance to the Government. Monetary financing o f fiscal deficits has been a common practice. Over 2000-2003, on average more than 90 percent o f the fiscal deficit (excluding grants) was covered by the central bank financing, which has resulted more often in inflationary pressure^.'^ Subsequently, the bulk o f financing o f fiscal deficits has shifted away from domestic financing toward external financing. Project loan financing covered on average more than 40 percent o f the fiscal deficit during the period FY2004-06, reflecting the return o f donor assistance t o Haiti after an economic embargo that began in 2000. Meanwhile, Central Bank financing o f fiscal deficits has been eliminated for the first time since 1999. The authorities’ current fiscal strategy aims at balancing the need to avoid central bank financing o f fiscal deficits, while at the same time providing adequate resources for investment and poverty reduction. 1.24. Securing sufficient and timely budget support is, therefore, critical to ensure T a b l e 6- O v e r a l l Fiscal B a l a n c e (HTG financing o f expenditures. Given the l o w revenue-to-GDP ratio, and constraints o n monetizing fiscal deficits, the Government might suffer from severe cash shortfalls when expected I disbursements are delayed. The need t o improve Including grants -2,696.3 -881 aid predictability and avoid delays in Excluding grants -2,926.0 -7,210.3 ’ disbursements i s therefore critical t o ensuring Excluding grants and NA -1,494 7 proper execution o f the budget. At the same time, externally financed projects Domestic financing 2,567.5 707 the Government must strengthen revenue BRH 2,700 828 projections underlying budget preparation and the Banks -74.5 - 1 17 cash management system. It should consider O.W. establishing precautionary cash balances that Other non-bank -5 .O -40.0 could be used as bridgeJinancing in case o f delays financing Arrears (net) -53.0 35.7 in disbursement o f budget support. The lack o f External financing 129 1,383.7 predictability in aid flows i s not only the result o f Loans 468.5 2,878.7 problems o n the donor side but also o n the Amortization -6,24.3 -1,338.7 Government’s side. Strengthening domestic Source: Government of Haiti, IMF and Staff management o f aid operations, including rigorous Calculations. monitoring o f implementation o f policy reforms, combined with greater predictability o f aid flows (based o n a well-designed multi-year program o f investment outlays), could help ensure that disbursements are made o n schedule. 1.25. Haiti needs ambitious fiscal reforms to support i t s development. A broad-based development strategy would require substantially larger amounts o f government resources than are currently available t o expand physical infrastructure and improve access to social services in the areas o f health and education. To this end, Haiti’s need to increase i t s domestic revenue i s substantial. Along the lines o f recent IMF work, this would involve reforming the tax administration and broadening the base o f the corporate and personal income taxes, and improving 14 Average inflation increased from 8.1 percent in 1999 to 16.5 percent in 2001 and hrther to 26.7 percent in 2003. 12 Haiti: Public Expenditure Manaqement and Financial Accountability Review tax and customs administration. In the medium-tern, emphasis should be placed on tax and customs administration in the provinces, and possibly raising excise rates. However, to “kickstart” and “sustain” economic growth will need a large increase in aid targeted at core infrastructure, to finance the PIP - Tax reform will pay greater dividends in a growing economy. On the expenditure side, Government expenditure would need to be subject to much tighter prioritization, scrutiny and control. Specifically, the trend toward lower spending on wages and higher spending on other current expenditures would need to be reversed. In addition, a shift toward a substantial expansion o f Government spending focused on capital expenditure would be needed. I .I .2.4. Development Financing and External Assistance Issues 1.26. I n the face of i t s low resource base and huge development challenges, H a i t i has relied on external assistance to finance i t s development needs. T h e pattern o f aid flows provided to Haiti has mainly been dominated by multilateral aid, which accounts for 28 percent o f total ODA over the period from 1990 to 2005. Multilateral assistance to Haiti was largely provided by IDB and the World Bank which together accounted for more than 40 percent o f multilateral ODA to Haiti. The United States, Canada and France were the major bilateral providers o f ODA over 1990- 2005, together accounting for 82 percent o f bilateral assistance. 1.27. Aid flows, provided by Figure 2. ODA received by H a i t i 1990-2005 (Total net, in a limited number of donors, constant U S $ million) tend to be volatile, reflecting in I I part political developments (see _ ” “ ” ” “ ” Figure 2). T h e massive inflow o f foreign aid under President Aristide in 1990-91 was followed by an embargo under the military regime (1992-94). Total ODA inflows (on disbursement basis) dropped from an average o f US$227 million during 1990-91 +in million USD (constant) +in % of GDP to US$141 million on average Source: OECD, IDS database, DAC over 1992-93. Subsequently, international assistance was revived from 1993 to 1995. It then declined from 1995 to 97 and resumed slightly in 1999. But aid flows declined after the disputed 2000 elections and have resumed since 2004 following the resumption o f democracy and political stability. 1.28. I n 2004, the Transition Government defined a development framework, the I n t e r i m Cooperation Framework (ICF) and associated financing needs.” Subsequently, the 15 The I C F covers the period July 2004 - September 2006 and consists o f four axes: (i) strengthening political governance and promoting national dialogue; ( ii)strengthening economic governance and contributing to institutional development; ( iii) promoting economic recovery; and (iv) improving access to basis services. T h e total financing needs for the implementation o f the I C F over the period July 2004 - September 2006 were estimated to be approximately US$1.37 billion. This amount does not include the financing urgent humanitarian needs presented in the Haiti Flash Appeal (US$ 35.1 million) as well as the additional needs resulting from the 2004 floods in the South-eat o f the country. The internal resources available from the national budget for the I C F activities were estimated approximately at US$127 million. 13 Haiti: Public Expenditure Manaqement and Financial Accountability Review Government detailed i t s main development priorities in the Extended I C F and in the Interim Poverty Reduction Strategy Paper (I-PRSP) completed in September 2006.16 In response t o the Government’s successful implementation o f i t s reform agenda, aid commitments have started increasing since 2004.” Table 7 summarizes the structure o f total assistance pledged over the period 2004-07 and the main areas o f donor interventions. Pledges exceeded financing needs over 2004-06 as defined in the ICF. The five main donors contribute the bulk o f financing to Haiti. T w o o f them, U S A and Canada, accounted for about 86 percent o f financing needs over 2004-06. In terms o f pledges, the top five accounted for about 80 percent o f total pledges for the period 2006- 07. They cover all four priority areas o f Haiti’s growth and poverty reduction strategy spelled out in the extended I C F and the I-PRSP. The lion’s share o f donor interventions i s directed for access to basic services (Pillar 4), which accounts for more than 30 percent o f total pledges over the extended ICF period 2006-07. This reflects donors’ response t o finance the Government’s Program o f Social Appeasement (Programme d’ Apaisement Social), which aims t o provide quick responses t o meet pressing social and economic needs in disadvantaged and conflict-prone areas. Likewise, external resources available were estimated at US$3 15 million. This leaved a financing gap o f US$924 million. See ICF 2004-2006 Summary Report. July 2004, p.4 1. l6The extended ICF was presented at the July 25,2006 Donors’ Conference in Port-au-Prince and covers the period July 2006 - September 2007. The extended I C F and the I-PRSP reaffirmed the continuation o f the axes and the strategic objectives o f the 2004 ICF. 17 Total financing requirement for the extended ICF period (July 2006 - September 2007) i s o f US$1.8 billion, o f which more than two thirds already covered by both domestic and external funding, leaving a gap o f US$544. At the Conference, the donors pledged about US$750 million for the period July 2006 - September 2007. 14 Haiti: Public Expenditure Management and Financial Accountability Review Table 7. Structure of Pledges and M a i n Donors Areas o f Intervention, 2004-07 (US$ Million) USA 634 210 844.2 46.3 28.0 Pillar 4 - Access to Basic Services (13 1.9) Pillar 1 - Political Governance (41.4) IDB 263 150 413 30.1 20.0 Pillar 2 - Economic Governance (47.5) Pillar 3 - Economic Growth (73.39) Pillar 4 - Access to basic services (15) EU 273 58 330.8 19.9 7.7 Pillar 4 - Access to Basic Services (58) Canada 535 107 642 39.1 14.2 Pillar 1 - Political Governance (54) Pillar 4 - Access to Basic Services (27) IDA 131 68 201.6 9.6 9.1 Pillar 4 - Access to Basic Services (30) Pillar 2 - Economic Governance (23) Pillar 3 -Promoting Economic Growth ( 15) Sources: Haiti Interim Strategy Note and Authors’ calcu8ations. Total financing needs for the implementation of the ICF over the period 2004-06 were estimated at US$1.37 billion. Total aid pledged for the extended ICF was estimated at US$750 million. 1.29. I D A financing mainly has focused on economic governance (pillar 2) and basic social services (pillar 4), which account for more than 75 percent of its pledges of financing for 2006/07. IDB financing has mainly focused on economic growth (pillar 3) and economic governance (pillar 2), which account for about 80 percent o f i t s pledges o f financing for 2006/07. In support o f economic governance and economic growth, the IDA envelope for FY0718 includes a development policy grant (DPG) to support the second phase o f the Government’s economic governance reform program (Economic Governance Reform Operation 117 US$23 million), a rural water and sanitation project (US$5 million), a project to support the implementation o f the Education for All (EFA) initiative, and a Risk Financing Project (US$9 million).” T h e IDB has committed in FY2007 to finance US$50 m i l l i o n in grants, US$12.5 million for the first loan o f a Public Resource Management Strengthening Program; US$25 million for a Productive Roads Rehabilitation Program; and US$12.5 million for an Agriculture ’*The IDA FY07’s envelope o f the amount of US$68 million i s the third exceptional IDA allocation under both IDA 13 and IDA 14 guidelines f o r H a i t i f o r the period FY05-07. Under IDA 13 and 14 rules, H a i t i received exceptional allocations of US$75 m i l l i o n in FY05 and US$63 m i l l i o n in FY06. For FY08, Haiti’s IDA allocation will b e structured in line with performance as per the usual performance-based allocation process o f IDA. A US$6 m i l l i o n Electricity Loss Reduction Project was approved by the B o a r d in early FY07 by using IDA allocations carried out from FY06. 15 Haiti: Public Expenditure Management and Financial Accountability Review Intensification Program. In FY2006, the IDB approved projects totalling US$86 million, divided between US$35 million for economic governance; US$36 m i l l i o n for economic growth and US$15 million for basic services. 1.30. Disbursement of aid pledged remains an issue as the experience of the implementation of the ICF confirms. As o f September 2006, about US$968 m i l l i o n had been mobilized to implement the interventions under the ICF and the extended ICF. T h i s represents about 90 percent o f the total U S $ l . 1 billion pledged during the July 2004 Donor Conference but less than 50 percent o f total estimated US$2.1 billion pledged by major donors throughout the ICF implementation period (See Table 8 below). However, this general trend might hide disparities among donors and across projects. While more than 60 percent o f EU and about 55 percent o f USA’s commitments have been disbursed as o f September 2006, less than 25 percent o f IDB’s commitments have been mobilized. Yet, these ratios might be high by regional standards. For instance, IDA disbursement ratio o f 38.1 percent i s quite high compared t o international standards o f 33.2 percent and 24.1 at the regional and bank-wide levels, respectively. Budget support operations tend to be disbursed quicker. The Economic Governance Reform Operation I (EGRO I ) was fully disbursed and closed in March 2006 as was the Fiscal Reform (PBL I ) by IDB that was fully disbursed and closed in September 2005 and a second PBL that was approved in July 2005 and has disbursed over 75 percent o f i t s resources. On the other hand, most o f the projects approved during the period July 2004-September 2006 s t i l l are at an early implementation stage, as many o f them experienced delays averaging six months in the declaration o f effectiveness, following the political transition.*’ 3.31. T h e low level o f disbursements i s related to the fact that some agencies disburse gradually and in line with progress in the implementation o f the programs they support. Delays in disbursements also reflect poor performance as w e l l as weak institutions and limited an capacities. Performance can improve quickly t o the extent that the government’s commitments are implemented on time and within the timeframe agreed upon with donors. By contrast, building institutions and capacities i s B gradual and difficult process, which requires strong and sustained commitment from the Government and the donors as w e l l as predictability in aid flows. Haiti faces the difficult challenge o f meeting its pressing development financing needs while at the same time building strong domestic institutions to mobilize much needed external resources. Table 8. Donor Development Financing. Commitments 2004-07 v. Disbursements as o f September 2007 (in USS Rlillion) Donors Commitments Oisbursenicnts t!nciisbursed as of Disbursement 2001-07 As of September Septeniber 2006 Rates (in YO) 2006 (1) (2) (3) ( 2 M 1) USA 844.2 451.3 386.9 54.2 IDB 591.5 146.5 445.0 24.8 EU 330.8 208.0 122.8 62.9 IDA 201.6 80.0 121.6 39.7 Canada 642.0 26.6 615.4 4.1 Others 76.2 52.7 23.5 69.2 Total 2,044.3 971.1 1,073.2 47.5 Sources: Government ICF and Staff calculations. ~ *’ See Haiti Interim Strategy Note. World Bank December 2006. 16 Haiti: Public Expenditure Manaqement and Financial Accountability Review 1.32. I n a weak institutional setting, huge aid flows might face absorptive capacity issues, with potential destabilizing macroeconomic effects. Creating or expanding the technical and human capacity to manage aid flows and macroeconomic policies thus becomes a priority for Haiti. Human resource management and capacity building i s a cross-sectoral issue that this PEMFAR covers. I t recommends that donors provide a full-fledged capacity building assistance to Haiti. 1.33. Channeling more funds through the national budget i s crucial to ensure coherent management of public resources and enable the government to explore and account for dynamic policy-tradeoffs between alternative allocations of government resources. I t i s also critical to maintain the short-term objective o f macroeconomic stability to improve the investment climate and general confidence in public policies, and the long-term goal o f strong and sustained growth for poverty reduction. However, the prerequisites for more budget support including transparency and accountability in public finance management remain a concern. Chapter 3 discusses these issues and provides policy recommendations. 1.2 Fiscal Policy and Economic Outcomes, 2004-2006 1.2.1 Fiscal Performance and Improved Economic Outcomes 1.34. I n 2004, Haitian authorities inherited a deteriorating macroeconomic situation. Real GDP contracted that year by 3.5 percent, resulting in a fall in per capita GDP by more than 5 percent. Fiscal deficit (including grants) worsened to 3.5 percent o f GDP in 2003. The Government also accumulated large external and domestic arrears. Prior to 2004, the Government had at times operated without approved budgets or with budgets approved late into the fiscal year. 1.35. The authorities moved quickly in implementing fiscal adjustment policies and structural measures over the past three years, backed by an M F - Staff-Monitored Program ( S M P ) , and two successive Emergency Post-Conflict Assistance (EPCA I[ and 11) as well as the Bank’s Economic Governance Reform Operation I(EGRO I ) and two IDB Policy Based Loans (PBL Iand 1 1) totaling US$50 million. The reform program sought to restore macroeconomic stability, improve public finance management, and boost economic growth. Since 2004, the economic outcomes o f the reform program have been satisfactory. 1.36. Government revenue (excluding grants) increased f r o m 9.0 percent o f GDP in FY2003 to an estimated 10.0 percent in FY2006, as a result of a series o f revenue-enhancing measures (see B o x 2). However, Haiti’s revenue performance remains weak and i t s revenue-to- GDP ratio i s one o f the lowest within the L A C region and even by sub-Saharan Afiica standards. At 10.0 percent o f GDP, Government revenues are 9 percentage points lower than that o f Burundi (post-conflict country), 6 percentage points lower than the West A h c a Economic and Monetary Union average and 6-7 percentage points lower than that o f Benin or M a l i (see Table 9). 17 Haiti: Public Expenditure Management and Financial Accountability Review Table 9. Government revenue (excluding grants) Haiti and Selected Comparable sub-Saharan Burundi 20.3 21.1 20.1 20.0 19.1 Central Africa 10.8 7.7 8.1 8.2 8.7 Republic Benin 16.3 17.0 16.4 16.5 16.7 Burkina Faso 11.3 12.1 12.8 12.3 12.4 C6te d’Ivoire 17.9 16.9 17.5 17.1 18.0 Guinea-Bissau 15.3 15.2 17.2 17.6 19.8 Mali 15.9 16.4 17.4 17.9 17.2 Niger 10.6 9.9 11.2 9.7 11.3 Senegal 17.9 18.2 18.5 19.4 19.7 Togo 12.3 17.0 16.8 15.7 16.1 IWAEMU 15.4 15.7 16.2 16.0 16.5 Sources: IMF, African Department Database; World Economic Outlook; and staff estimates. Estimate as of February 2006. 1.37. The Government tightened expenditure controls, mainly through reducing use of discretionary accounts. In 2006, recurrent expenditure was kept broadly in line with its 2002 level o f less than 10 percent o f GDP. With increased revenues and tighter expenditure controls, the central Government overall deficit (including grants) was reduced from 3.5 percent o f GDP in FY2003 t o 0.9 percent in FY2006. This has largely eliminated recourse t o central bank financing o f the Central Government deficit. 1.38. The substantial fiscal adjustment helped reduce inflation and supported the recovery o f growth. End-of-period inflation fell from 38 percent in FY2003 to an estimated 12 percent in FY2006. However, this rate o f inflation i s s t i l l high relative to comparable l o w income countries. It i s six times higher than the West Afncan Economic and Monetary Union (WAEMU) average (2.0 percent) and Mali, five and three times higher than those o f Burluna Faso and Benin, respectively. Annual GDP growth increased to 1.8 percent in FY2005 and i s expected t o increase further t o 2.3 percent in 2006 after a decline o f -3.5 percent in 2004. N e t international reserves (NIR) have increased, raising import coverage fi-om 11’4 months o f imports o f goods and services in FY2003 to an estimated 1.7 months in FY2006. 1.39. Progress has also been achieved in the implementation o f structural and economic governance measures, notably in the areas of budget formulation, execution, and reporting as well as public procurement and public enterprise management and road maintenance. Chapter 3 o f the report provides an overview o f the public finance management reforms undertaken since 2004 and the remaining challenges. 1.40. The successful implementation o f i t s reform program enabled Haiti to be granted a Poverty Reduction and Growth Facility (PRGF), and to reach i t s Enhanced HIPC Decision Point. 18 Haiti: Public Expenditure Management and Financial Accountability Review Box 2. Revenue-Enhancing Measures Introduced since 2004 T h e following reforms have been undertaken since 2004: Internal revenue Income tax: Establishes a new individual income tax, raising the minimum taxable wage from 20,000 to 60,000 gourdes; a new corporate tax, a proportional rate o f 30 percent i s imposed on the net income o f companies and other corporate entities; for individuals residing in Haiti, tax assessments are based on residence and, for nonresidents, o n income generated in Haiti; access i s provided to information held by banking institutions on taxpayers being audited. Special stamp duties: Establishes a special value-added tax o f 0.02 percent o n all invoices and receipts issued by the DGI. Business volume tax: Establishes a simplified regime for individuals subject to income tax who maintain business volumes below 1,250,000 gourdes. Taxpayer register: Establishes a tax registration number, the “Tax Identification Number” (NIF); as a single central computerized taxpayer register, to b e supplied with data from the manual registers held at different locations around the country and will gradually encompass all operations pertaining to each taxpayer. Includes an amendment to tax legislation so as to require use o f the NIF on all tax documents. Establishes appropriate sanctions for fraudulent use; and modifies how the National Identify Card (CIN) i s administered by tax services. Tax code rationalization: Eliminates the tax stamp o f .01 percent on invoices and receipts; eliminates the special 12 gourde duty on invoices and receipts issued by the DGI (substituted by Special Stamp Duties above); and abolishes duties and fees o n imported alcohol and tobacco. Legal framework The L a w o f 15 July 1996 was amended to provide for bonuses to be paid to officials who monitor and impose fines for infractions o f customs rules and laws, to facilitate and accelerate clearance o f goods through customs. 0 Customs Code: A new Customs Code has been drafted, to be i n line with modem codes on the matter, and that includes valuation o f goods in accordance with GATT Chapter VII. The new Code has been submitted to Parliament. Institutional and administrative framework Regulatory Framework N e w organizational charts prepared for the administrative departments; a manual prepared containing descriptions o f functions; agents trained in the new control techniques; monitoring measures adopted; personnel designated for the new services; and a monitoring committee designated. Training area: An AGD official has studied customs techniques at the World Customs Organization (WCO); two officials were trained at a 10-month course at the French Customs School in Tourcoing; three officials took a n i n e month customs training course at the Morocco Training Center. Information Systems: AGD has begun to implement Sydonia World, a computerized management system covering most foreign trade procedures. This system processes manifests, customs declarations, accounting, transhipment and exemptions. 1.2.2 Debt relief and Fiscal Space Issue 1.41. Because o f i t s good policy track record over the past three years, Haiti reached the Decision Point under the Enhanced HIPC Initiative’s export window in November 2006. Thus, H a i t i was granted debt relief estimated at US$140 million in NF’V terms (US$212.9 million in nominal terms) as o f end-September 2005, w h i c h reduced the level o f the debt to a more 19 Haiti: Public Expenditure Management and Financial Accountability Review sustainable 150 percent debt to export ratio.21 H a i t i would qualify for Multilateral D e b t R e l i e f Initiative (MDRI) f r o m IDA u p o n reaching i t s completion point.22D e b t r e l i e f associated with MDIU, i s estimated at about U S $ 2 4 3 million in NPV terms (US$464 million in n o m i n a l terms) assuming that H a i t i reaches i t s completion point by end-September 2008.23In addition, several Paris Club members have indicated their intention to provide additional r e l i e f b e y o n d the HIPC Initiative. Table 10. External Debt Service, 2006/06-2014/15 (In USS million unless otherwise indicated) 2005/ 20061 20071 ZOOS/ 20061 20101 2Olll 20121 20131 20141 Average 06 07 08 06 10 11 12 13 14 15 2005l06 2014t15 Reduction in debt serviceasaresult ... 13.3 6.5 15.1 11.2 5.3 5.0 5.2 50 5.6 8.0 of HIPC assistance’ Reduction in debt service as a result 0.0 0.0 0.0 9.3 14 1 19.8 19.7 19.6 19.5 20.5 12.2 of MDRI assistance’ Reduction in debt service as a result of additional 00 0.0 2.4 5.3 5.4 5.3 5.3 5.3 5.3 4.8 3.9 Source: Enhanced Heavily Indebted Poor Countries (HIPC) Initiative Decision Point Document. World Bank, October 2006. ,The reduction in debt service i s measured as the difference between the projected debt service after full use o f traditional debt relief and debt service after the application o f H P C relief MDRI assistance applies only to the World Bank and starts after the completion point (September 2008) Assumes that MDRJ has no impact on Haib’s new borrowng over the projection penod. 1.42. Debt service relief after HLPC and MDRI relief i s estimated at about US$22 million on average over 2005/06-2014/15 and about US$26 million on average over the same period when bilateral relief beyond the HIPC i s considered (see Table 10 above). T h e reduction in debt service would provide some JiscaE space in Haiti’s budget. M a x i m i z i n g the poverty impact o f resowces made available by debt r e l i e f would require that the Government ensure that these resources are allocated to productive spending. T h e Government has recently defined a l i s t o f pro- poor expenditures. Inthe short-term, there i s a need to reinforce this l i s t and translate i t s items into 21 Within the framework o f the Enhanced HIPC Initiative, the level o f indebtedness is considered unsustainable when the N P V o f debt to exports o f goods and services i s above 150 percent. As o f end- September 2005, Haiti debt in N P V terms, after full application o f traditional debt r e l i e f mechanisms, i s estimated at US$928 million, equivalent to 177 percent o f exports o f goods and services (above the HIPC Initiative threshold). ** The assistance provided by IDA would cover all outstanding debt disbursed from IDA prior to end- December 2003. Haiti i s not expected to have eligible debt for MDIU debt r e l i e f from the IMF. Haiti i s scheduled to repay all eligible debt to the IMF (i.e. debt that was outstanding to the IMF before December 3 1,2004) b y December 2006. 23 W i t h the delivery o f MDRI assistance, Haiti’s NPV o f debt-to-exports ratio would significantly fall, remaining within the 90-100 percent range over the period 2006-25. This assumes that MDRI has no impact on Haiti’s new borrowing over the projection period. 20 Haiti: Public Expenditure Management and Financial Accountability Review specific budget lines. The ongoing preparation o f the PRSP offers the opportunity to define such a mechanism that could ensure these expenditures are protected against unpredictable shortfalls in government revenue. This could be done through the design o f a cash allocation plan. 1.43. But the notion of pro-poor expenditure i s not clear. F o r instance, spending on roads might b e more “pro-poor” than direct transfers to low-income households because they foster investment and growth (and thus an increase in income o f the poor, even with no change in income distribution) and have a more lasting impact o n poverty reduction than transfers. A better distinction should be between productive spending (which enhance growth: infi-astructure, education, health, the latter two affect productivity; see Chapter 7 B o x on infrastructure) and non-productive spending (or not directly productive; transfers for instance). F r o m that perspective, the critical issue i s not t o protect some categories o f spending that are deemed “pro-poor”, but rather to protect a core PIP, which will foster growth and help to reduce poverty (see Chapter 8 macro-model and policy simulations). In this context, designing a core integrated PIP agreed upon with the donor community, which identifies foreign resources available should reduce in the short-term, the negative effects o f volatility o f aid flows that Haiti has undergone since the 1990s. In the medium-to-long term, i t should firm up the foundations for a growth and human development strategy. 1.44. Fiscal space associated with debt service relief depends closely on the Government’s ability to accelerate i t s reform agenda and reach the Completion Point o f the Enhanced HIPC Framework by September 2008.24 A delay in reaching the completion point would result in revenue loss. Table 11 below shows projected revenue loss if the completion point i s delayed. For instance, Haiti would forgo an estimated US$48.5 million (equivalent t o more than FY2006-07 recurrent budget for health and more than 60 percent o f that o f education sector) in debt relief in the event o f a one year delay in reaching the completion point. 24 See Assistance to the Republic o f Haiti Under the Enhanced Heavily Indebted Poor Countries Debt Initiative. World Bank. Report No. 37394-HT. October 25,2006. 21 Haiti: Public Expenditure Management and Financial Accountability Review 64.5 68.4 71.1 76.8 78.6 83.0 88.0 93.9 100.5 before debt relief HIPC debt relief 26.8 20.2 21.2 7.1 1.2 0.6 0.7 0.7 0.6 Debt service after 37.1 48.1 49.9 69.7 77.3 82.3 87.3 93.3 100.0 HIPC 2008 completion 37.6 52.0 57.2 56.9 56.5 56.3 51.0 point relief Debt service after HIPCandMDIU 37.7 48.1 12.3 17.7 20.1 25.5 30.9 36.9 43.0 (2008) Cumulative cost of delaying the 48.6 107.7 166.1 223.6 280.8 337.8 395.3 completion point Source: Enhanced Heavily Indebted Poor Countries (HIPC) InitiativeDecision Point Document. World Bank and IMF staff estimates, M a y 2007. 1.45. Because o f the nature and the limited amplitude o f debt relief, fiscal space resulting from debt relief will provide a small boost to resources and will need to be supplemented with an increase in aid to support the Government’s public investment program. The resources saved from debt relief could prove useful if they are invested properly. However, given their magnitude, the impact on growth and poverty will be limited unless Haiti finds additional means to expand fiscal space and finance an ambitious public investment program. At the same time, there i s some potential to expand the fiscal space both at the expenditure and revenue levels provided that the reallocation o f expenditures in the budget i s done efficiently and revenue-enhancing measures identified above are implemented. B o x 3 below summarizes the issues associated with fiscal space and the ways through which a typical LIC such as Haiti could create fiscal space. Regarding expenditure efficiency and reallocation, i t involves reducing waste in public spending, particularly in the case o f public investment. Regarding tax revenues, efforts to broaden the tax base and strengthening tax administration (through, for instance, a reduction in collection costs and measures to reduce tax evasion) might result into additional revenues, which can be used to finance productive spending. Foreign assistance i s needed to strengthen the Government’s efforts to increase domestic revenue. However, the tendency o f foreign grants t o be highly volatile raises the issues o f sustainability o f medium and long-term development programs hinging on the risks o f shortfall in grant financing. 22 Haiti: Public Expenditure Management and Financial Accountability Review I Box 3. Creating Fiscal Space: Old W i n e in N e w Bottles? Definition. Fiscal space can b e defmed as “the availability o f budgetary r o o m that allows a government to provide resources for a desired purpose without any prejudice t o the sustainability o f a government’s financial p o s i t i ~ n . ~ ’ ~ ~ W a y s to create fiscal space. How can low-income countries create fiscal space? Essentially, there are three ways to make additional resources available for productive government spending, without compromising the sustainability o f public finances in the medium term: (i) through improvement i n expenditure efficiency and reallocation o f spending, ( iithrough higher tax revenues, and ( ) iii) through greater domestic o r foreign financing at relatively l o w cost. Expenditure efficiency and reallocation of spending: A first option t o create fiscal space i s by reducing waste in public spending, particularly on public investment. T h i s can b e achieved by improving governance in a broad sense, that is, with tighter management and greater accountability. Another option i s t o reduce “unproductive” expenditures. However, care must b e taken in classifying expenditures as “unproductive.” In an environment such as H a i t i where crime and violence deter individuals from saving and investing, spending o n security should retain a high priority - despite the fact that i t might n o t appear t o b e prima facie directly productive. Similarly, attempting t o create fiscal space by cutting spending o n maintenance might b e an ill-advised strategy; i t might “work” in the short term (inthe sense that consequences for the quality o f infrastructure might n o t b e immediate), but only at a greater economic cost in the long term. Higher tax revenues: I t i s possible that by broadening the tax base and strengthening tax administration (through, for instance, a reduction in collection costs and measures to reduce tax evasion), additional revenues can b e raised to finance productive spending. In many poor countries, however, the capacity to raise resources through taxation remains limited by the very fact that incomes are l o w t o begin with; and if the components o f spending that are being contemplated for expansion require large o r lumpy investments (as is often the case for infrastructure), relying solely o n higher taxation t o create fiscal space might n o t b e feasible. Low cost domestic o r foreign financing: Resources can b e borrowed o r received in the form o f foreign grants. In many p o o r countries, the capacity t o borrow domestically i s often limited by the size o f the financial sector.26 A s for foreign borrowing, unless it i s o n highly concessional terms, it might adverse implications for the sustainability o f public debt. Debt relief, by contrast, frees resources that were previously earmarked for debt service operations, and might in principle provide a significant boost t o outlays in some specific productive spending ~ a t e g o r i e s .An ~ ~ alternative financing option i s foreign grants. However, what has become clear in recent years i s that grants tend t o display a high degree of volatility.28 T h i s i s particularly problematic if a country must finance a multi-year public investment program that requires a sustained i n f l o w o f resources; an unexpected shortfall could derail the program and annihilate its potential effects o n growth and fkture tax revenues.29 Another potential problem with 25 Peter S. Heller, “Understanding Fiscal Space,” IMF Policy Discussion Paper No. 05/4 (March 2005).. See also Peter S. Heller, Menachem Katz, Xavier Debrun, The0 Thomas, Taline Koranchelian, and Isabel Adenauer, “Making Fiscal Space Happen!,” World Economics, Vol. 7 (July-Sept. 2006), 89-132. 26 The degree o f financial development, together with other institutional factors (such as the nature o f the country’s exchange rate arrangement) also limits the ability to raise revenues through seigniorage. 27 See for instance John Weeks and Terry McKinley, “Does Debt Relief Increase Fiscal Space in Zambia? The MDG Implications,” Country Study No. 5, International Poverty Centre (September 2006), for a case study o f the impact (or lack thereof) o f debt relief o n fiscal space. 28See Ales Bulir and A. Javier Hamann, “Volatility o f Development Aid: F r o m the Frying Pan into the Fire?,” Working Paper No. 06/65, International Monetary Fund (March 2006), and UNCTAD, Doubling Aid: Making the Big Push Work, Geneva, 2006. 29 In fact, as argued for instance by Pierre-Richar Agenor and Joshua Aizenman (“Aid Volatility and Poverty traps”, unpublished, University o f Manchester, M a y 2007) a high degree o f aid volatility m a y deter 23 Haiti: Public Expenditure Manaqement and Financial Accountability Review foreign aid (in addition to short-term Dutch disease effects) i s an adverse moral hazard problem - the fact that aid might weaken revenue performance and lead to higher (unproductive) spending. I f so, then the fiscal space initially created by higher aid flows might vanish fairly quickly. Lessons learned. There are two main lessons that can be drawn from the foregoing discussion. The first i s that considerations related to fiscal space are best made in the context o f an explicit medium-term budget framework, with a clear outline o f the government’s expenditure priorities - particularly in the area o f public investment - and available financing options. In that sense, the current debate is “old wine in new bottles.” The second important lesson that can be drawn i s that assessing ways to create fiscal space, while ensuring sustainability o f the fiscal stance, i s inherently country specific and requires the use o f a quantitative model that accounts for the country’s key structural economic features, i t s revenue and expenditure structure, the composition o f i t s debt and i t s cash flow implications, and i t s prospects for greater access to foreign resources. Such models are essential to discuss h o w a reallocation o f spending from “unproductive” outlays to investment in health and education, for instance, might affect growth and the budget, and whether “true” fiscal space can be achieved. As discussed elsewhere in this report, issues o f efficiency and strengthening o f fiscal institutions can also be framed in the context o f these models. 1.46. Beyond the issue of expanding the fiscal space, quantifying the potential growth and poverty impact of increased fiscal space i s more relevant from a policy perspective for a country such as Haiti. U s i n g Haiti’s macro-framework, p o l i c y simulations o f increased fiscal space are run and their g r o w t h and poverty impact are assessed. Chapter 8 reports on the results o f these simulations. 1.3 Policy Recommendations on the findings o f this chapter, the following p o l i c y recommendations are made: rate the implementation of policies for macroeconomic stability and structural reforms. Improved macroeconomic stability that Haiti has achieved since 2004 would need to b e reinforced to create conditions for a stable environment favorable to private sector development. To this end, accelerating the policies for macroeconomic stability i s critical and should b e done in the context o f the ongoing IMF-PRGF program. Areas f o r further improvement include price stability, reducing current account deficit, boosting gross official reserves, and improving fiscal strategy. However, macroeconomic stability policies w o u l d n o t achieve their expected growth and poverty reduction objectives, unless they are complemented by structural reforms. Reforms should b e done to strengthen economic governance. Specific areas o f focus include public financial management, transparency and accountability o f public sector operations, and efficiency and transparency in public infrastructure management. The Bank Economic Governance Reform Operation I1 and subsequent Development Policy Lending (DPLs) and IDB’s P B L s would offer the opportunity to advance the agenda o f economic governance. The challenge facing H a i t i i s to sustain a broad- based reform over the next decade and permanently move away from the policy o f “stop and go” the country experienced over the past decades, which has locked H a i t i in a low-growth-high poverty incidence. Implement revenue-enhancing measures and a more aggressive policy to attract foreign grants. Haiti’s revenue t o GDP ratio i s one o f the lowest in the world. Fortunately, there i s potential t o increase revenue. T h e Government should implement the revenue-enhancing governments from implementing (costly) reforms aimed at raising domestic resources, and may well lead to a poverty trap. 24 Haiti: Public Expenditure Management and Financial Accountability Review measures identified by the IMF-FAD and IDB. While these measures could increase revenue, increasing the revenue/GDP ratio to international levels would require time to implement. In the short term, Haiti would continue to depend on foreign aid to finance i t s development pnorities. The Government should therefore put in place an aggressive policy o f attracting foreign grants. Large flows o f foreign grants might raise several issues, including: (i) Dutch Disease effect through appreciation o f the real exchange rate and loss o f competitiveness; ( ii ) absorption capacity problems and the r i s k o f wastage o f resources, in particular in the context o f limited human and institutional capacity; ( iii) governance problems; and (iv) moral hazard through the reduction in incentives to collect domestic resources. An effective management o f grant flows would thus be needed to avoid their potential destabilizing macroeconomic effects. I t could require setting an institutional mechanism to this end. The trackmg mechanism o f aid flows to be implemented in the context o f donor coordination offers an opportunity to discuss the outlines o f such mechanism. Strengthen the management o f external debt to maintain debt sustainability. Debt management has improved in Haiti over the past years. While the MEF i s rebuilding i t s database, the Central Bank o f Haiti (Banque Centrale de la Republique d’HaTti, BRH) has a relatively complete debt databa~e.~’ The BRH updates i t s database at every payment cycle, ensuring the authorities’ database i s broadly in l i n e with the creditors. Overall, the coverage o f public debt (external and domestic) i s appropriate. The BRH produces monthly, quarterly and annual reports that contain data on external debt. These reports cover the transactions (disbursements and payments) as well as the stock o f debt and the accumulation o f arrears. The reports are disseminated throughout the Central Bank and the MEF. T h i s allows authorities to integrate the relevant information into the macroeconomic framework. The data i s available to the public upon request w i t h one month after the reference period, and it i s subsequently published with some additional delay. Looking ahead, Haiti needs to M e r strengthen i t s debt management capacity by: (i) clarifying by a protocol of understanding between the BRH and the MEF the debt management responsibility; ( i i)improving information sharing, including frequent debt reconciliation exercises, between the BRH and the MEF; ( iii)shortening the procedures for debt service payments; (iv) improving the tracking o f disbursements; (v) acquiring a modern debt reporting system; (vi) training o f staff; and (vii) improving the capacity to produce debt sustainability analyses. I n addition, authorities should implement a prudent external debt policy to keep the levels o f external debt at sustainable levels. T h i s requires that contracting future debt obligations should be scrutinized to avoid jeopardizing the debt situation as emphasized in the HIPC Decision Point Document. Improve the efficiency of public spending to “kick-start” and “sustain” growth to alleviate poverty. The impact o f government spending on growth and improving human development depends on the efficiency o f these outlays and how well they are targeted at the poor, not just on the level o f spending. I n Haiti, improving efficiency o f public spending by better allocating the Government’s scarce resources could contribute to improved growth performance. I n fact, policy priorities should focus on allocating more resources to productive spending (infrastructure, education, health and security) and less to unproductive spending (transfers). Efficiency arguments also suggest that public spending should be directed to areas with the highest social returns and should complement, rather than compete with, the private sector. T h i s involves either financing or supplying directly needed public 30 Currently the Central B a n k o f H a i h and the MEF are j o i n t l y responsible for debt management in Haiti. The archives o f the MEF were devastated by a fire in 2002. 25 Haiti: Public Expenditure Management and Financial Accountability Review goods that the private sector will not supply adequately because o f market failure. In Haiti, on the contrary, Government failure to provide adequate education and health services has resulted in the presence o f a large non-public sector (see Chapter 5). Efficient use o f public resources in this context requires finding the ways to use Government resources so that poor students currently un-enrolled in school could attend non public primary schools free o f charge. Chapter 5 provides policy recommendations to that effect. Several categories o f public expenditure can influence long-term growth in Haiti - especially spending on infrastructure, education, health and security. Efficient use o f public resources requires that the Government focus o n these sectors. Higher growth, in turn, would generate increased fiscal resources to finance productive spending, further bolstering the dynamism o f the Haitian economy. Chapter 7 presents a macroeconomic framework that l i n k s public investment disaggregated into investment in infrastructure, education, health and security, and growth and poverty. Chapter 8 assesses the impact o f higher efficiency o f public spending on growth and poverty reduction. 0 Design and implement a multi-year and integrated public investment program to be supported by foreign aid flows. Haiti has prepared a three-year PIP, using a back-of-the- envelope methodology and a top-down approach. T h i s i s a good start. The challenge i s n o w to move to an integrated multi-year PIP, which accounts for the linkages between various sectors, most notably infrastructure and education and health. The new PIP should adopt a bottom-up approach based o n sectoral strateges. I nfact, the new approach should allow line ministries to prepare their sector PIP, which should identify required investment, anticipated recurrent expenditure and needed aid flows. The sectoral PIPs should then be put together to produce a Government integrated and multi-year PIP. It i s critical that the integrated PIP account for the linkages between the sector and not be a simple adding-up o f sectoral PIPs. Thus, there is a need to coordinate among various ministries and spending agencies for establishing a coherent, core PIP. Aid requirements might thus be lower and should be identified at the outset o f the designing process o f the integrated PIP. The integrated PIP should also be a rolling exercise allowing some flexibility to account for resource constraints. The design o f the integrated PIP should be done in the context o f the revised PRSP (three years from now) gwen the current timetable to complete the ongoing PRSP. The financing o f the integrated PIP would require that all the resources possible be added. In other words, resources freed from debt relief should be complemented by other sources o f financing to ensure additionality o f resources. I t also involves that the resources are channeled through the Government’s budget to ensure that Government has full control o f all resources allocated to execute the PIP. This, in turn, implies improved effectiveness, transparency and accountability in budgetary operations: the core objective o f this PEMFAR. 26 Haiti: Public Expenditure Management and Financial Accountability Review CHAPTER 2 : PUBLIC EXPENDITURE REVIEW: BROAD TRENDS AND PATTERNS 2.1. Chapter 1 o f this PEMFAR report has reviewed the macroeconomic and fiscal context of Haiti. I t has analyzed the broad structure o f Haiti’s public finance by focusing mainly o n the economic composition o f spending: trends o f expenditure categories (wages and salaries, goods and services, interest payments, and transfers and subsidies). Chapter 2 reviews the broad trends and patterns o f public spending by functional classification. 2.2. The analysis o f public expenditures by functional classification centers o n whether sufficient resources are being allocated t o sectors and line ministries. However, the extent to which resources allocated to sectors are considered sufficient or n o t i s limited by the lack o f sectoral strategies with clearly defined spending targets. In this context, the current analysis limits itself to broad trends toward the Government’s development priorities spelled out in the extended ICF and the I-PRSP. The analysis also highlights the broad trends in actual spending and discusses the drivers o f the execution o f spending. The structure o f the Government public investment program (PIP) i s discussed, focusing particularly o n the influence o f donors’ financing o f the PIP. However, because o f the lack o f a complete series o f data, the analysis o f the execution o f the PIP i s limited in time (FY2004-05 PIP), and scope and source o f funding (national resources). 2.3. The analysis o f pubIic expenditure structure and trends i s complemented by an analysis o f spending in the priority sectors - agriculture, education, health, infrastructure, and justice and security - in Chapters 4, 5, and 6 to determine whether the allocation and the use o f public resources has been consistent with the ICF and I-PRSP and sector strategies (where they exist). 2.4. The main findings o f this review are the following. First, the functional classification o f Haiti’s budget i s dominated by the predominance o f the “executive branch”, which accounts for 95 percent o f total allocation o n average over FY2002-07. Second, over the past five years, there has been a shift in expenditures away f r o m the “political sector” in favor o f the “economic” and “social” sectors, reflecting the authorities’ recent efforts to align expenditures to economic and social objectives as w e l l as the return o f donors’ financing t o support investment programs since 2004. The share of “economic sector” in total budget allocation increased sharply f r o m 10.8 percent o f the total FY2004/05 budget to more than 48 percent in FY2006/07. The budget share o f the “social sector” also increased from 14.1 percent t o 18.4 percent in that period. Third, the recent trends in recurrent expenditures reveal the Government’s efforts to translate its development priorities set forth in the ICF and the recently approved I-PRSP into budget priorities. The share o f recurrent expenditures allocated t o “economic sector” increased to an average o f about 15.1 percent o f total recurrent expenditures (excluding interest payments) over the FY2005-07 period, up f r o m an average o f 11.8 percent dunng the FY2002-04 period. Allocations to the “social sector” followed a similar increasing trend up to 24.8 percent on average over FY2004-06 f r o m 23.8 percent. Fourth, the bulk o f the increase in allocations t o “economic” and “social” sectors was directed to growth-enhancing and poverty reducing sectors, including agriculture, transports and communications, education, health, and justice and security. Accounting altogether f o r about 16 percent o f the FY2006/07 GDP, the share o f these five sectors in the total budget increased by about 10 percentage points over the period under review: up f r o m 36.3 percent in the FY2002-04 period to about 50 percent o n average over the three fiscal year-period FY2005-07. Fifth, geographic allocation o f expenditures reflects the highly centralized government’s expenditures, 27 Haiti: Public Expenditure Manaqement and Financial Accountability Review which are allocated mainly t o the central government. About 72 percent o f the FY2006/07 budget was allocated to the central government while only 28 percent was granted to the ten departments. Sixth, total spending (including recurrent and all capital expenditures) declined slightly in real terns by about 2 percent on average between the FY2002-04 and the FY2005/06 periods. Seventh, there has been a gradual decline in the use o f comptes courants as a source o f execution o f spending in priority sectors since 2004. This decline reflects the shift toward the use o f normal expenditure procedures, namely “direct payments”, and the Government’s efforts to improve transparency and accountability in expenditure management in the context o f adhesion to governance reforms. Eighth, the structure o f Haiti’s public spending, in particular the investment budget, i s mainly determined by foreign aid flows. Volatility o f aid flows, limited absorptive capacity and poor budget planning have resulted in l o w execution rates. Ninth, the strong influence o f donor financing o n expenditure choices has not been offset by a “jiungibility effect”, i.e. by a reallocation o f the Government’s own funding to non-priority sectors, as donor financing for priority sectors became available. 2.5. Policy recommendations emphasize the need for H a i t i to: (i)improve budget classification by enabling the distinction between sector directly productive and sector indirectly productive; ( ii)enhance the geographic coverage o f the budget by better targeting the poorest departments; ( iii) accelerate the policy o f reducing the comptes courants; (iv) reallocate resources in the budget to account for “fungibility effect”; and (v) improve the execution o f spending by improving budget planning, increasing absorptive capacity, and efficiency o f public resources. 2.1 Analysis of Public Expenditure by FunctionalClassification 2.1.1 * Shift in Budget Allocation toward Economic and Social Sectors 2.6. A key feature o f Figure 3. Broad Structure of Budget Allocations in Percentage of budget allocation in Haiti i s Total Budget (average over FY2005-07) the predominance o f the “executive branch” (economic, political, social, and cultural sectors), which accounts for about 95 percent 61 h e c u t i i e of total allocation on average Legislative over FY2002-07. Legislative BI Judiciary and judiciary branches, and hdependent Organizations other independent organizations Firblii Debt Anartization (CSCCA, University o f Haiti, etc.) together make up on average barely 5 percent o f total allocation over the same I I period (see Figure 3). Source: Le Moniteur, Journal Officiel de la RCpublique d’Ha’iti and Staff Calculations. 28 Haiti: Public Expenditure Management and Financial Accountability Review 2.7. Over the past five years, Figure 4. Shares o f Executive Branch Components in there has been a shift in Percentage of Total Budget FY2004-07 expenditures away f r o m the I “political sector” in favor of the “economic” and “social” sectors, reflecting the authorities’ recent efforts to align expenditures to economic and social objectives as well as the return o f donors’ financing to support investment FY2004-05 FY200506 FY2006-07 programs since 2004. Budget allocations to “economic sectorsy731 increased sharply f r o m 10.8 percent Cultural Sector Others Administrations I o f total FY2004/05 budget to more Source: L e Moniteur, Journal Officiel de la RCpublique d’HaYti and than 48 percent in FY2006-07. This Staff Calculations. reflects mainly the large increases in real terms o f allocations to the LLeconomic sectors”, which Table 12. General Trends in Budget Allocations, FY2002- increased six-fold in FY2005/06 and further by 55 percent in FY2006/07. The allocations to the “social sector” doubled, leading to an increase in their share in total budget allocation f r o m 14.1 percent Economic Sector Political Sector Social Sector 4236.7 4060.0 45 14.5 13027.5 4908.5 55 12.5 I in FY2005-06 to 18.4 percent in Cultural Sector 319.4 374.9 FY2006/07. In the meantime, the Other Administrations 6929.1 6369.0 share o f “political sector” declined 2. LEGISLATIVE 489.8 355.7 by five percentage points from 19.3 3. JUDICIARY 280.2 241.2 percent o f total allocations o n 4. OTHER 435.0 508.5 ORGANIZATIONS average over FY2004/05 to 14.3 TOTAL 21264.7 32491.3 percent in FY2006/07, reflecting Source: L e Moniteur, Journal Officiel de la RCpublique d’Haiti and the decline in shares o f the Ministry Staff Calculations. o f External Affairs, the President’s Office, the Prime Minister’s Cabinet and the Ministry o f Interior. Most notably, allocations to the President’s Office decreased by about 20 percent between FY2005/06 and FY2006/07. Allocations to the Ministry o f External Affairs also declined by the same magnitude over the same period. Allocations to the Prime Minister’s Cabinet were also cut - though by less (about 3 percent) -between FY2004/05 and FY2006/07. 31 In Haiti budget classification, economic sectors are represented by Ministxies o f Planning, Economy and Finance, Agriculture, and Public Works, Transport and Communications, Commerce and Industry, Environment, and Tourism. Social sectors include Ministries o f Education, Social Affairs, Health and Population, Women Rights, and Youth and Sports while “political sectors are represented by Presidency, the Prime Minister’s Office, the Interior Ministry, Ministry o f Foreign Affairs, and Ministry o f Haitian Diaspora. 29 Haiti: Public Expenditure Manaqement and Financial Accountability Review 2.8. The recent trends in recurrent expenditures reveal the Government’s efforts to translate i t s development priorities set forth in the 13. Broad Trends in Budget Allocations to Priority ICF and the recently approved I-PRSP into budget Sectors, FY2002-07 (real priorities. The share o f recurrent expenditures allocated t o the “economic sector” increased t o an average o f about 15.1 percent o f total recurrent expenditures (excluding interest payments) over the FY2005-07 period up f r o m an average o f 11.8 percent during the FY2002-04 period. Allocations t o the “social sector” followed a similar increasing trend up to 24.8 percent o n average over FY2004-06 from 23.8 percent. T h i s increase was mainly driven by the allocations to education sector, whose share in total Source: L e Moniteur, Journal Offciel de la recurrent spending (excluding interest payments) W u b l i q u e d’Hai% and Staff Cakulations. grew from 15.6 percent over the FY2002-04 period to 17.9 percent during the FY2005-07 period. In real terms, the allocations o f recurrent budget to the education sector increased by more than 33 percent over the past five fiscal years.32 The share o f the “political sector” shrank to 26.8 percent o n average over the FY2004-07 period from more than 27.5 percent during the FY2001-03 period, following the reduction in resources allocated to the President’s Office combined with a flat trend in allocations to the Ministry o f External Affairs and the Prime Minister’s Cabinet. 32 The FY2006107 combines the allocations t o Ministere Education Nationale et Formation Professionnelle and Ministbre de l a Jeunesse des Sports e t 1’Action Civique. Adding “interventions publiques” (subventions-fonctionnaires rentrire des classes) would further increase the share of education sector. 30 Haiti: Public Expenditure Management and Financial Accountability Review 2.9. T h e bulk of the increase in allocations Table 14. Broad Trends in Budget Allocations to Priority Sectors, kY2002-07 (in to “economic” and “social” sectors was real HTG Million) directed to growth-enhancing and poverty reducing sectors, including agriculture, transports and communications, education, health, and justice and security. Accounting MIN AGRICULTURE RES 297.8 1320.8 NAT & DEV RURAL altogether for about 16 percent o f FY2006/07 MIN TRAV. PUBS, 2,157.2 6,449.7 GDP, the share o f these five sectors in total TRANSP & COMM budget increased by about 10 percentage points MIN DE L A JUSTICE 1,547.4 2,675.7 over the period under review: up from 36.3 MIN EDUCATION NAT, J 2,606.4 3,297.4 percent in the FY2002-04 period to about 50 & SPORTS MIN JEUNESSE SPORTS percent o n average over the three fiscal year ET ACTION SOCIALE period FY2005-07 (see Table 14). T h i s reflects UNIVERSITE DETAT 280.8 260.3 the combined effect o f increased resources to DHAITI the investment budget and a reallocation o f INTERVENTIONSPUBS EDUCATION recurrent expenditure over the past three fiscal EDUCATION ENSEMBLE 2,887.2 3,635.2 years. In real terms, total allocations to these (INCL. UNIV HAITI AND sectors more than tripled over the period INT PUBS-RC) FY2005/06 fiscal years compared to the MIN SANTE PUB & POP 1,233.6 1,902.7 FY2002-04 fiscal years. A s indicated above, the TOTAL PRIORITY 8,123.3 15,984.2 GROWTH AND POVERTY reallocations o f recurrent expenditures in the SECT budget were done by reducing or keeping TOTAL 21,264.7 32,460:3 almost unchanged the allocations t o the SHARES i m TOT& 36*3---- 49.9 “political” sector, most notably the Presidency, ource: L e Moniteur. Journal Officiel de la theprime ~ i ~ Office, and the~ i ~ ~ ~ ~of 7 ~ d’Hditi i Rbpubllqne ~ i and Staff ~ Calculations. ~ External Affairs. The shift in budget allocations was particularly pronounced during the past two fiscal years. Total allocations t o the growth-enhancing and poverty-reducing sectors more than tripled in real terms in FY2005-06 and nearly quintupled in FY2006/07 compared to FY2003/04. 2.10. Geographic allocation o f expenditures reflects the highly centralized government’s expenditures, which are allocated mainly to the central government. About 72 percent o f the FY2006/07 budget (46.4 billion gourdes or US$l.1 billion) was allocated to the central government while only 28 percent (18.2 billion gourdes i.e. US$433 million) was granted t o the 10 departments. In terms o f recurrent budget, the central government received more than 80 percent o f the FY2006107 budget. The distribution o f resources between departments seems not to reflect the poverty incidence. For instance, the Department o f West (which has the lowest poverty incidence) was granted about 8 percent o f total FY2006/07 allocations while the North-East department (with the highest poverty incidence) received less than 1 percent. The discrepancies are more strilung when accounting only for recurrent expenditures. The Department o f West received about 37 percent o f the recurrent budget while the North-East Department was allocated only 4.5 percent. M o r e worrying i s the fact that the poorest department (the North-East Department) received only 2.5 percent o f total investment expenditures allocated to the departments. Catching-up o f poor departments might therefore be difficult as the budget does not appear to be an instrument to address differences between the departments. Part o f the explanation o f this problem i s related t o the lack o f a spatial approach in the definition o f the Government’s budget. In fact, budget conferences have not factored the geographical and spatial dimension o f the budget into discussions about budget allocations. This also reflects the fact that decentralization i s at an early stage. However, i t i s worth mentioning that the Government recently has made an effort to account for this spatial dimension o f the budget. The FY2006/07 budget includes an annex o n the budget 31 Haiti: Public Expenditure Manaqement and Financial Accountability Review allocations to the departments. The real test would be for the Government to move toward a more decentralized budget at the time o f the completion o f the ongoing decentralization process. Table 15. Geographical Distribution of Government Budget by Department (In nominal HTG Department o f West 1,765.90 3,191.30 4,957.20 7.7 Department o f South-East 260.1 670.9 93 1 1.4 Department o f North 511 1,030.80 1,541.80 2.4 Center Department 206.7 412.6 619.3 0.9 Department o f Artibonite 560 3,467.50 4,027.60 6.2 Department o f South 363 3,112.60 3,475.60 5.4 Department o f North-East 209.1 341.6 550.7 0.8 Department o f Grand-Anse 301.3 203.1 504.4 0.8 Department o f North-West 196 900.3 1,096.30 1.7 Department o f Nippes 285.3 200.2 485.5 0.7 I Total 25,183.80 39,382.90 64,566.70 100 Source: L e Moniteur, Journal Offciel de la Rkpublique d’Haili and Staff Calculations. 2.1.2 Analysis o f Actual Expenditures 2.1.2.1 Broad Trends in Actual Spending 2.11. Accounting for about 13.8 percent of Table 16. Total Spending by Broad GDP on average over the FY2005/06 period, Categories (real HTG Million), total spending (including recurrent and all FY2002-06 capital expenditures) declined slightly in real ‘iverage terms by about 2 percent on average between FY FY FY2002-04 and FY2005-06 (see Table 16). This 2002104 2005i06 declining trend o f spending i s driven mainly by TOTAL 18,282.21 17,920.81 the “legislative and judiciary branches”, where 1. EXECUTIVE spending declined in real terms by more than 60 POWER 17,337.86 17,179.46 percent and 4 percent, respectively (see Table 16). ECONOMIC SECTOR 2,923.31 3,277.08 However, because o f the l o w share o f these POLITICAL entities in total spending (both account for less SECTOR 5,434.05 3,891.69 than 4 percent o f total spending), the impact o n SOCIAL SECTOR 3,811.70 3,353.85 total spending was quite limited. CULTURAL SECTOR 327.40 348.23 OTHER PUBS ADMINISTRATIONS 4,841.40 6,308.61 2- LEGISLATIVE POWER 393.73 142.65 3- JUDICIARY POWER 213.54 205.38 4- INDEPENDENT ORGANISMS 337.08 393.32 Sources: L e Moniteur, Journal Officiel de la RCpublique d’Ha1ti and Staff Calculations. 32 Haiti: Public Expenditure Manaqement and Financial Accountability Review 2.12. At the sub-sector level, spending o n economic sectors has increased while Table 17. Functional Distribution of Actual political and social sectors recorded a Total Expenditures-Payment basis decline over the period under review. Total spending in the economic sectors increased in real terms by 13 percent on average between the period FY2002-04 and the FY2005/06 period, mainly driven by spending o f the Ministry o f Economy and Finance (recurrent 1326.4 1202.1 Communications spending) and Ministry o f Planning and Justice and Security 1705.0 1843.7 External Cooperation (investment spending), Education (Incl. University o f 2,547.0 2,298 5 which both account for about 50 percent o f total spending o f economic sectors o n average over the past two fiscal years (FY2004/05 and FY2005/06). The sharp decline in spending o f Source: MEF, W o r l d B a n k and IDB staff calculations. the Presidency largely determined the decline in total spending o f t h e political sector by more than 28 percent. I t i s worth noting that the share o f the political sectors in total spending has been declining since FY2002/03 and much further in FY2004/05: 20.5 percent o f total spending compared t o 34.5 in FY2001/02. Despite significant increases in allocations, total spending in social sectors declined by about 12 percent on akerage over the period under review. T h i s reflects mainly the decline in investment spending in health sector, education and the Ministry o f Social Affairs, while recurrent spending has been increasing in these sectors (see Table 18). 2.13. Total investment on priority sectors declined in real terms by about 40 perceah between FY2002-04 and FY2005/06, stemming from a decline in investment spending in the social sectors (as mentioned above) and in investment spending in transports and communications sector. 33 Haiti: Public Expenditure Management and Financial Accountability Review 2.14. Accounting for 4.2 percent of FY2005/06 GDP, recurrent spending for priority sectors increased by Table 18. Functional Distribution of Actual Public Recurrent Expenditures-Payment basis nearly 5 percent in real terms between FY2002-04 and FY2005/06 (see Table 18). As the result, the share o f priority sectors in total recurrent spending (excluding interest payments o n the debt) increased by more than 300.3 303.4 seven percentage points, from an average o f 37.8 percent in FY2002-04 1443.8 1759.0 to 44.3 percent in FY2005/06. This 2296.0 2483.7 positive trend was dnven by the 954.7 732.1 increases in the share o f social sectors 5269.6 5522.6 (education and health) and justice and 8937.0 78933 security sectors, which altogether 14206.6 13416.0 account for more than 90 percent o f total recurrent spending o n priority sectors spending (the largest share). Meanwhile, recurrent spending o n amculture and transports were marginal, reflecting the very nature of Source: MEF, World Bank and IDB staff calculations. these sectors, which are mainly “investment sectors”. 2.15. Authorities intend to further increase the are o f spending on priority sectors, particularly health and education in the coming years. This signals the Government’s commitment t o implementing i t s development priorities. However, the success o f this approach will depend o n the Government’s ability t o increase absorptive capacities and enhance resource management in the sectors. T h i s also requires that a l l priority sectors develop comprehensive sectoral strategies with clear and realistic spending targets and outcomes t o monitor the progress achieved. While designing the spending targets, the authorities should n o t only take into account their priorities but also the absorptive capacities o f the sectors. The sectoral targets could b e presented in the ongoing full PRSP to b e completed by the end o f 2007. 2.16. Yearly fluctuations o f spending on priority sectors during the period under review, reveal the fragility o f a spending policy highly dependent on volatile donor aid flows (investment budget) and subject to abrupt changes in political context (recurrent spending). The sudden decline in spending o n priority sectors by 44 percent in FY2003/04 compared to FY2002/03, following political i n ~ t a b i l i t y ~ ~ donors’ “wait and see” attitude illustrates the lack and o f capacity to conduct a sound and predictable spending policy in such an environment. Reducing volatility o f spending would require broadening the domestic resource base t o strengthen the ability o f the economy to compensate for unpredictable changes in foreign aid by more stable domestic resources. However, this is a medium-to-long-term goal g v e n the current status o f the revenue basis in Haiti. In the short-term, developing policies t o protect spending in the priority sectors should receive particular attention. They could take the form o f execution o f priority spending 33 President Aristide’s departure from the country on February 29, 2004 l e d to a “wait and see” attitude by m a j o r donors. 34 Haiti: Public Expenditure Management and Financial Accountability Review n the items, the l i s t o f which could be drawn from the recently developed poverty reducing list. I medium-term however, an integrated PIP agreed upon with the donor community could help to prevent abrupt changes in allocations and their destabilizing effects on spending execution. 2.1.2.2 Sources of Spending and Use of Comptes Courants 2.17. I n Haiti, the execution of spending has used both normal procedures and discretionary accounts, so-called the comptes courants. The pervasive use o f comptes courants which evolved as a result o f an attempt to circumvent the normal expenditure execution process (itself long plagued with complicated procedures between the requisition and disbursement stages), has resulted in lack o f transparency in public resource management and has l e d to corruption. The use of comptes courants has been highly discretionary and circumvents the normally complex ex- post control procedures o f the Supreme Audit Institution (Cours SupCrieure des Comptes et du Contentieux Administratifs or CSCCA). The authorities thus committed to reduce the recourse o f comptes courants as part o f their efforts to return to normal budget execution procedures and improve transparency and accountability in the use o f public resources. 2.18. Table 19 shows the gradual Table 19. Shares of Compte3 Courants in Total Spending for Selected decline in the use o f comptes courants in the execution of spending in priority Rural Development sectors. The share o f Public Works, Transport and 87.6 comptes courants in Communications total spending of Justice and Secunty 19.6 Education (Incl. University o f Haiti) 31.7 9.8 0.5 selected priority sectors Health 30.3 23.7 14.9 0.9 dropped to less than 3 24.4 31.4 19.7 2.7 percent in FY2004/05 Source: MEF, World B a n k and IDB staff calculations. compared to about 20 percent in FY2003104 and 31 percent in FY2002/03. The decline was across all priority sectors with the Ministry o f Public Works, Transports and Communications recording the highest decline. 2.19. The decline reflects the shift toward the use o f normal expenditure procedures, namely “direct payments”. I t also signals the Government’s resolve to improve transparency and accountability in the expenditure management in the context o f adhesion to governance reforms. The test facing the authorities n o w i s to ensure a sustained decline in the comptes courants in particular in priority sectors. Phasing out completely the use o f comptes courants i s a challenging objective given the fact that the recourse to this expenditure mode has been a deep-seated practice since the early 2000s. In addition, the flexibility and rapidity o f execution o f spending that the comptes courants have offered t o line ministries require that their phasing out be replaced by an expenditure mechanism that encompasses those aspects o f the comptes courants (flexibility and rapidity) and ensures transparency in spending execution.34 34 A l o n g the same lines, it i s worth mentioning that the IMF considers that phasing out completely the use o f current account might b e difficult, at least until there i s a clear strategy for a) development o f a TSA, in w h i c h ministries’ revenue and spending accounts would eventually b e zero-balanced (but not necessarily 35 Haiti: Public Expenditure Manaqement and Financial Accountability Review 2.1.3 I n t e r n a t i o n a l Comparison o f F u n c t i o n a l D i s t r i b u t i o n o f P u b l i c E x p e n d i t u r e 2.20. I n terms of resources allocated to the Table 20. Comparison of Spending on Priority Sectors. I n percent education, health and rural development sectors, H a i t i fares far below in comparison to selected ComDarable Health 5.6 9.3 11.3 15.7 10.8 countries in sub-Saharan Africa (see Table 20). k p ; and 3.7 6.7 NA 11.0 4.4 Allocations for transports Development and communications have 19.1 Transport and 10.5 NA 7.3 13.7 been relatively high over Communications the past 2-3 years. Justiceand 9.2 NA NA NA NA Security 2.2 Factors D e t e r m i n i n g the Structure of P u b l i c E x p e n d i t u r e 2.21. Two factors stand out as major determinants of the structure of public expenditure in Haiti: (i)the level o f domestic resources; and (ii) the influence of external financing. 2.2.1 L e v e l o f Domestic Resources 2.22. Because of Haiti’s current low domestic resources, the structure o f public expenditure shows a dual aspect. The Government’s scarce resources are mainly allocated t o recurrent expenditure while investment expenditures are mainly financed by external assistance. “Fixed and quasi-fixed” expenditures (wages and salaries, debt service, and transfers, and subsidies to some public enterprises) account for a large part o f the Government’s resources, leaving little for the other category o f recurrent expenditures - goods and services. Expenditure o n goods and services appears as a “residual” item and operates as an “adjustment” item in the recurrent budget. F o r the same reasons, the Government’s limited resources are allocated t o few line ministries or priority sectors (education, health, economy and finance, etc.) with little funds directed t o other sectors. 2.23. National resources to finance the investment budget are limited and their allocation i s pre-determined. Investment expenditures are not identified and planned in a coherent framework. In fact, there has not been a clearly integrated multi-year investment program. Only recently has the Government prepared a rough multi-year public investment program, the quality o f which needs to be improved. There i s a lack o f strategic planning o f investment expenditure after about 20 years o f neglect o f capital budgeting and weakening o f institutions involved in preparing and monitoring the public sector investment program (especially the Ministry o f Planning and External Cooperation). The situation has caused a disconnect between public sector capital expenditures on the one hand and poverty reduction and socio-economic development needs on the other. The lack closed) at the end o f each working day; b) procedures for supplying petty cash needed by spending ministries, including for advance accounts. 36 Haiti: Public Expenditure Manaqement and Financial Accountability Review o f public investment planning and monitoring has meant that public sector investment has been dnven by donor-funded and NGO projects, further causing a disconnect between the Government’s current expenditure budget and the overall investment made in the public sector (including through donor-funded projects). Government has thus little influence o n the investment budget. As a result o f weak investment planning, changes t o the investment budget occur in an ad-hoc manner during budget execution. In fact, sectoral ministies have had little control over the execution o f capital expenditure in their respective sectors. 2.2.2 Influence o f External Financing 2.24. External financing has a strong impact on the structure o f expenditure since more than 20 percent of Haiti’s total public expenditure and about 70 percent o f i t s capital expenditures in 2004-06 were financed by external assistance. The donors’preference for financing projects and programs in priority sectors has a strong influence o n the share o f priority sectors in total public expenditure (as highlighted in Chapter 1). 2.25. Haiti’s heavy reliance o n external budget support reinforces the outside influence o n expenditure choices. B o t h tied and, t o a lesser extent, untied budget support requires the Government to earmark budget resources for specific sectors and/or types o f expenditures. In the case o f tied financing, this link i s direct. Yet even untied budget support operations are usually based o n an agreement between the authorities and the donor to earmark a certain share o f this support t o specific sectors. However, t o what extent this earmarlung o f expenditure has translated into a reallocation o f expenditure toward non-priority sectors (jiungibilify effect)? 2.26. There are clear indications that the strong influence o f donor financing on expenditure choices has not been offset b y a “fungibility effect” i.e. by a reallocation o f the Government’s own funding to non-priority sectors, as donor financing for priority sec&m became available.35An analysis o f the fungbility o f external financing and domestic resources . reveals that the possibility o f external funding expenditures crowding out expenditures that are traditionally financed from domestic resources i s limited. External project financing accounts for nearly 70 percent o f Haiti’s investment budget. In addition, counterpart funding t o external projects constitutes a large share o f the domestic contribution to the investment budget. 2.27. T h e large share of donor financing o f Haiti’s budget has resulted in a lack of autonomy in decision making by the authorities and a strong vulnerability o f sector spending to fluctuations in external financing (as mentioned above). The Government has recognized this and has launched a process o f enhanced donor coordination with and among donors within the framework o f the Extended ICF and current PRSP implementation. A medium-term program i s under preparation and its successful execution will require that the Government and donors jointly adopt medium-term spending priorities and integrate their funding. However, such positive development will require that the Government make an effort toward improving the budget preparation process, in particular t o eliminate the dichotomy in the formulation o f the recurrent and investment budgets. Experiences in other countries (e.g. Uganda) has shown that it takes a long time and concerted efforts to fully integrate all funds from domestic and external sources in the budgeting process. 35 See World Bank (1998) for a discussion o f fungibility o f foreign aid. 37 Haiti: Public Expenditure Management and Financial Accountability Review 2.3 A n a l y s i s o f t h e P u b l i c I n v e s t m e n t Program 2.3.1 S t r u c t u r e o f t h e Public I n v e s t m e n t P r o g r a m 2.28. T h e structure o f the PIP i s skewed t o w a r d economic sectors, o n average consisting o f m o r e t h a n 80 percent o f total resources allocated t o the PIP over FY2005-07, w h i l e social and cultural, a n d political sectors account f o r 12 percent and 5 percent, respectively (see Figure 5). The analysis o f the intra-sectoral allocations o f spending o f the FY2005-07 PIP reveals the dominant position o f transport sector, which accounts for more than 40 percent o f total allocations while agriculture, education and health made up about 8 percent, 5 percent and 6 percent, respectively. PIP2004105 PIP2005106 PIP2006107 cioCultura1 Sector Sources: National Authorities Data, and B a n k and IDB staff calculations. 2.29. Economic sector. T h e PIP in the economic sub-sector i s largely dominated by the allocations t o the MTPTC ( M i n i s t h e des Travaux Publics, Transports et Communications) a n d the M i n i s t r y o f planning, w h i c h together account f o r about 75 percent o f allocations t o the economic sector, over FY2005-07. Allocations to investment programs reflect the large program and projects being executed (or in the process of) in the transport sector, following the return o f foreign aid flows as well as developing planning programs under the control o f the Ministry o f Planning. 2.30. Social-cultural sector. Investment programs and projects in t h e social sector are mainly directed t o t h e education and health sectors. Altogether, these sectors account for 90 percent o f the social-cultural sector PIP on average over FY2005-07. This structure o f the social- cultural PIP reflects the Government’s policy choice o f priority given to improvement o f education and health outcomes while social protection appears to be a secondary objective. 2.31. Political sector. At about 70 percent o f the total PIP, justice and p u b l i c security dominate the structure o f the political sector PIP. Mainly as the result o f renewed attention to 38 Haiti: Public Expenditure Manaqement and Financial Accountability Review public security, foreign resources to justice and public security PIP are projected to increase in the FY2006/07 budget (2.1 billion gourdes or US$49.9 million). The extent to which these resources could be absorbed by the sector i s not clear as they are not backed by a full-fledged strategy with spending targets (see Justice and Security PER for further discussion of these issues). 2.32. M o r e generally, the lack of specific spending targets in the PIP does not provide a baseline for assessing the trends of resources allocated to the sectors. Thus, the increase in allocations (or in the shares) becomes meaningless. Part o f the problem i s the absence o f a clear link between the PIP and sectoral strateges. I n fact the PIP i s built with little (or no) reference to n addition, most o f the sector strateges (when they exist) do not have specific sectoral strategies. I spending targets. 2.33. The volatility o f both domestic and external resources complicates the assessment and translates into variability of shares and relative importance of the sub-sectors. Yearly changes in allocations appear drastic, uncontrollable, and do not display any specific pattern. For instance, allocations t o planning, health and justice/security dropped significantly in FY2005/06 PIP. In the meantime, resources allocated to transport and to a lesser extent agriculture sectors increased sharply (see Table 21). Part o f the volatility o f the allocations i s explained by the cycle o f externally-financed programs/projects which might start or end with little Government’s control. I t i s also due to a lack o f rigorous resource planning on the part o f the Government. Yet, planning resource allocation i s a daunting challenge in a context of a PIP largely dominated by unpredictable donor financing. 2.3.2 Influence o f Donor Financing on the Public Investment Program 2.34. Haiti’s public investment program (PIP) i s mainly financed by external assistance. The arialysis of available data reveals that 80-90 percent o f Haiti’s PIP in 2004-06 was financed by donor assistance (see Table 21). Investment in economic and social sectors i s predominantly donors’ financed. Ninety percent o f the public investment program in agnculture and transports was financed by donors over the period. I n addition, more than 70 percent o f education and health PIP was financed by donors. Meanwhile foreign resources to the justice and security sector has been limited as only 30 percent o f the PIP o f the sector was financed by donors in 2004/05 and n o external resources was allocated t o the sector in 2005/06. 2.35. The increase in resources allocated to the PIP over the past 2-3 years reflects donors’ increased allocations. Allocations to the PIP increased from 9,014.4 million gourdes in FY2004/05 to 16,774.9 m i l l i o n gourdes in FY2005/06: an increase o f more than 60 percent in real terms. T h i s reflects the doubling o f donor financing over the period, from 7,226.9 m i l l i o n gourdes to 15,235.9 million gourdes. 39 -p?- r-0- - m - $Nly 8"- --??- --mo-0000 ? - m c i 2 % - N 0 9 ? o m w m 2 s -t0o!-tp?c\-, o m w o - o o o c " 7 N d o m m N W I " 2 0 300 2 , - - & - P A i d N o o ~ o d o o m - ooooooooc 0 d- 2 - .m2 3 W N G ? O mOO 0 0 0 0 0 0 ~ 0 0 0 ~ o w v N 2 % w m Haiti: Public Expenditure Manaqement and Financial Accountability Review 2.36. T h e high dependence of the PIP on donor financing raises the issue o f predictability of aid flows. Indeed, an effective implementation o f the PIP over the medium term requires a sustained inflow o f resources. T h i s implies that an unexpected shortfall could derail the program and hamper its execution. Predictability o f aid flows i s much needed if Haiti’s PIP has to b e a tool for shaping the basis for i t s investment and development priorities. 2.4 Budget Execution 2.4.1 Broad Trends o f Budget Execution 2.37. T h e subsequent chapters provide detailed analysis o f the execution o f spending in each sector covered by this PEMFAR report, including agriculture, education, health, infrastructure, and justice and security. T h i s section attempts to summarize the broad features o f execution o f spending by covering the three broad budgetary sectors, namely political, economic and social sectors. The section focus only o n the overall budget while subsequent chapters discuss details o f the execution o f both the recurrent and the investment budgets. Because o f the poor quality o f data, caution should b e taken when analyzing the execution rates and interpreting their evolution. 2.38. A key feature of the execution rates i s their volatility during the FY2002-06 period. On average, T a b l e 22. Execution Rates i.n execution rates were relatively high during FY2001-04 sub-period, reflecting poor budgetary planning o f resources required by the sub-sectors rather than high absorptive capacity, and the use o f non transparent execution procedures: the comptes courants. The execution rates have dropped since FY2004-05: 68.9 percent o n average over FY2005106 compared to 91.3 Sources: Government Statistics and Staff percent in FY2002-04. This reflects mainly the limited absorptive capacity o f the sectors in the face o f sudden increases o f allocation o f resources since FY2004105, following large increases in foreign aid. 2.39. This general trend tends to hide high variability among sub-sectors. Broadly speakmg, the political sector tended t o fully spend (or even overspend) the resources allocated during the first sub-period under review (FY2002-04). The political instability that characterizes this period l e d to recourse to increased spending by the executive branch, most notably the Presidency, the Prime Minister’s Office, and the Interior Ministry. The execution o f spending was facilitated by the increased recourse to the comptes courants. Execution rates in the political branch declined (but s t i l l remain at high levels), reflecting the adjustment o f spending more and more in line with resources allocated, in the context o f improved transparency and accountability in public finances. 2.40. While social sectors recorded execution rates relatively high36, resources allocated to economic sectors were poorly executed. Part o f the problem i s the weak capacity o f ministries composing this sector, including agnculture, public works, and commerce and industry, to absorb the large resources provided over the past two fiscal years, mainly foreign a i d flows. Another 36 The execution rates in social sectors should however been interpretated with cautious as they m a y reflect the large share o f wage expenditures, which more often have execution rates o f close to 100 percent. 41 Haiti: Public Expenditure Management and Financial Accountability Review explanation i s the difficulty o f line ministries to adapt to new budget procedures, including the use o f regular execution procedures in place o f the comptes courants. 2.4.2 Execution o f the PIP 2.41. Data on the execution o f the PIP i s scarce and exists only for the part o f the PIP executed on national resources in FY2004/05. Caution should be taken when interpreting the execution data, which might not reflect the whole picture o f the execution o f the PIP in H a i t i as the data does not capture the part of the PIP executed on foreign resource^.^' F o u r main features o f the execution of the PIP are worth emphasizing. First, poor budget planning o f the PIP has resulted in disbursement o f resources higher than resources projected under the revised PIP (less optimistic than the original PIP): 2.3 billion gourdes (US$ 59 million) compared t o 1.8 billion (US$ 45.9 million) projected in the PIP. Second, overall execution o f the PIP i s quite low: only 25 percent o f domestic resources allocated t o the PIP are actually absorbed. T h i s reflects the combination o f l o w absorptive capacity o f the overall Haitian economy, slow start o f the execution o f the PIP, and delays on the part o f line ministries in processing the required documentation for the disbursement o f funds, in executing their projects and programs. In addition, the revision o f initial allocations to the PIP resulted in further delays by line ministries, most o f which adopted a “wait and see attitude”. Third, large disparities exist between and within sub-sectors. Economic and political sectors recorded relatively l o w execution rates: 20 percent and 15.6 percent, respectively. Within the economic sector, most o f line ministries in charge o f the execution o f programs and projects recorded’low execution rates, with the exception o f commerce and industry. Along the same lines, some ministries composing the political sector, including the Ministry o f Interior and Ministry o f Haitian Diasgora did not spend the resources allocated t o their PIP. Part o f the problem i s related to the political instability that the country experknced in the year 2004, which disrupted the investment process and the spending procedures. kth, most o f the line ministries focused more o n executing the recurrent budget most notably gh the use o f comptes courants, while little I ’ attention was paid to the investment budget. 2.42. What Drives the Execution Rates? Exogenous factors could have influenced the execution o f spending in Haiti, including for instance political instability and donors’ withdrawal from the country in FY2002/03. These factors should not have been underestimated when analyzing the trends in execution rates. However, more generally, three factors can influence the execution o f spending in a typical LIC, such as Haiti: (i) a cash rationing decision to limit an eventual fiscal deficit (a macroeconomic issue); (ii) poor budget planning (a macroeconomic issue); and ( iii)limited absorptive capacity at the sectors level (micro-sector issue). The first factor i s well established as a key determinant o f execution rates in many LIC in sub-Saharan Africa.38Over the past four years, Haiti did not experience a sudden cut in expenditures during the fiscal years, which might have prevented the country from fully executing i t s planned spending. In fact, the l o w execution rates are essentially due to the combination o f poor budget planning and the limited absorption capacity o f the Haitian economy. Poor budget planning i s reflected by the lack o f a sound budget preparation process. The numbers included in the Government budget are more often meaningless as they do not reflect the real needs o f the sectors and the country as whole. 37 There are allegations o n the side o f the Government which indicate that in July 2005, some US$400 m i l l i o n were invested by the donors t o execute the PIP. However, the details o f the execution (break-down between axis, programs and projects) are n o t available. 38 The typical example i s Niger, w h i c h uses a cash rationing system t o comply with budget deficit Criteria agreed upon in the context o f IMF-PRGF supported programs and the West A f r i c a Economic and Monetary U n i o n Fiscal Convergence Criteria. 42 Haiti: Public Expenditure Management and Financial Accountability Review Chapter 3 o f the report documents the weaknesses in budget preparation and planning and their implications for the performance in budget execution. I t also proposes some remedial measures. Absorptive capacity typically refers t o limits o n a country’s ability to use aid effectively owing to the quality o f a country policies and institutions and lack o f administrative capacity in the f o r m o f specific skills or, more generally, o f insufficient human resources and physical conditions (infrastructure and equipment) for policy and program implementation. Chapters 4,5, and 6 discuss the issue o f absorptive capacity o f the sectors in the presence o f huge foreign aid flows. 2.5 Policy Recommendations 2.43. Based on the findings o f this chapter, the following policy recommendations are made: 0 Improve budget classification b y enabling the distinction between directly productive sectors and indirectly productive sectors. The current functional classification o f the budget - broken down between “executive power”, “legislative power”, “judiciary power” and “other independent entities” - reflects more the need to match budget allocation and spending with the existing separation o f functional power between the three entities o f the state. Also, the rationale for including the sectors and sub-sectors under each category i s not well explained. From a policy perspective, it would be better to complement the current classification by a classification that allows the distinction between directly productive (agriculture, infrastructures, health and education) and indirectly productive sectors (judiciary and legislative branches). T h i s would allow distinguishing between productive and unproductive spending.39 Moreover, as Haiti i s preparing i t s PRSP, which will become its strategic framework for i t s development agenda over the medium term; it i s critical that the budget classification reflects the Government’s declared objectives to grant priority to key sectors. To I . this ,efid. distinguishing between priority and not priority sectors could be a classification to consider. The advantage o f doing so is two-fold: First, it would allow the Government to assess whether the allocations and actual spending are in line with i t s deveiopment priorities. This requires that the sectoral strategies are in place and include specific spending targets. Second, such classification would enable the Government to track the execution o f specific spending in the priority sectors and provide a policy response to correct deviations from the spending targets. 0 Improve the geographic coverage o f the budget by better targeting the poorest departments. H a i t i has made great strides to improve i t s budget presentation. The FY2006/07 budget includes an annex o n geographic allocation o f expenditures. This i s a good start. Yet, geographic allocation o f spending seems n o t t o reflect the country’s spatial poverty incidence. There i s a need t o make the budget reflect the Government’s poverty reducing priorities by better targeting the poorest departments. Starting in FY2008/09, budget conferences should include sessions to discuss levels o f allocations to the various departments. This would require involving representatives o f the departments into the budget discussions. However, the issue i s the ability o f the department representatives to b e familiar with budget procedures and influence the budget decisions. Loolung ahead, the Government w o u l d need t o strengthen local capacities in budgetary processes. This could be done in the context o f the ongoing decentralization program. 39 For this purpose, the authorities could explore how to use a GFS-based functional classification system in the coding system for spending. 43 Haiti: Public Expenditure Management and Financial Accountability Review 0 Accelerate the policy o f reducing the coniptes courants. A major achievement of budget execution in H a i t i since 2004 has been the reduction in the discretionary accounts: the comptes courants. However, there are concerns about the existence o f comptes courants, which some line ministries s t i l l run. A first step could b e for the Government to clearly define the l i s t o f existing conzptes courants, the institution which still are using these accounts and their volumes. A second step should be t o agree with the line ministries concerned o n a calendar t o phase out the comptes courants. Finally, the Government w o u l d need to strictly implement the condition agreed upon with the IMF and the World B a n k to limit the ratio o f conzptes courants to total expenditure (excluding salaries) t o 10 percent. This would signal i t s resolve t o improve transparency and accountability in the use o f public resource. 0 Reallocate resources in the budget to account for the “fungibility effect”. Since 2004, H a i t i has benefited f r o m large a i d flows, mainly directed to priority sectors. This has provided some leeway t o the Government’s budget. A close l o o k at the financing picture reveals that the ‘tfungibility effect” o f external financing and domestic resources did not occur. Government did not reallocate its own resources to non-priority sectors as the foreign aid became available. F r o m a policy perspective, the Government w o u l d need to assess its maneuver r o o m to reallocate resources t o non-priority sectors. This does not however mean t o change i t s development strategy, which rightly focuses o n priority sectors. This means that the resources should be allocated as priorities t o sectors where the needs are, in particular when foreign aid flows, abundant. The marginal returns o f increasing the allocations o f public resources to priority sectors facing absorptive capacity issues might be declining while the resources are needed in other sectors. The challenge for the Government i s to find the right balance between allocating more resources t o non- sectors and keeping the Government’s development oHjectives intact. 0 Improve the execution o f spending b y improving budget planning, increasing absorptive capacity, and efficiency of public resources. The trend o f execution o f spending over the past five years was not satisfactory. T o improve the execution o f spending, policy actions should focus on: (i) improving budget planning; and ( ii ) increasing sectors’ absorptive capacity. In the short-term, the Government would need to undertake b o l d actions in the area o f budget preparation by strengthening the capacity to prepare the budget (revenue and expenditure projections, sectors needs, etc.). I n the medium-term, policy actions should focus o n increasing the capacity o f sectors to absorb the flows o f foreign resources. T h i s would include increasing their budget management capacity, human resource, increasing the scope o f programs and projects to b e executed, etc. But beyond the objective o f increasing the levels o f execution rates, the m a i n issue i s the issue o f efficiency of spending. The quality o f spending matters more than i t s levels. Improving the efficiency o f public spending w o u l d require better targeting the allocations o f public resources. Increasing resources t o the infrastructures sector might have a stronger growth and human development impact (through i t s direct and indirect effects o n education and health outcomes) than transfers to public utilities o r subsidies t o ill-functioning public schools. 44 Haiti: Public Expenditure Manasement and Financial Accountability Review CHAPTER 3 : COUNTRY FINANCIAL ACCOUTABILITY ASSESSMENT 3.1. Effective, transparent budget management and accounting instruments are essential to allow Government officials to make the right spending decisions, monitor budget implementation, evaluate the cash position o f the Government, and adjust spending levels to available resources as needed. This chapter o f the PEMFAR report summarizes the detailed assessment and recommendations provided in the full Country Financial Accountability Assessment (CFAA) and the Country Procurement Assessment Report (CPAR), attached to this PEMFAR report as volumes I and 11, re~pectively.~’ 3.2. While reliable data i s not available to assess some o f the indicators, by using the OECD/DAC and the PEFA methodology, the PEMFAR lays the foundation for future work. 3.3. The key findings o f the chapter are the following. First, the budget i s not forward looking. The link between sectoral policies and priorities and the budget i s very weak. The budget i s only based o n projections o f activity f r o m the previous year and i t i s not rooted in a medium-term investment plan. Second, there are many off-budget operations. An important part o f public spending i s still channeled through comptes courants, and “own resources” collected by the Ministry Departments and Agencies (MDA), and special accounts o f the Treasury. -It i s difficult t o quanti@ the extent o f these unreported operations. Third, budget execution i s affected mainly by the lack o f cash f l o w planning and monitoring. Budget releases to the M D A s are made based o n the , ceilings fixed by the Constitution (one-twelfth o f the annual budget appropriations) and not on the effective cash f l o w needs. Fourth, a sound accounting system, including clear stamlards and a related automated information system, s t i l l needs to be developed. The link between the different automated financial management software and that used by the MEF to report on Revenue i s not automatic. Fifth, despite a relatively comprehensive institutional and legslative framework, the internal and external oversight o f the budget i s s t i l l not effective given the lack o f capacity o f the institutions. Sixth, the adoption o f the new procurement l a w i s a necessary f i r s t step but not sufficient, by itself, to institutionalize the recent procurement reforms. Seventh, the successful implementation o f the PFM reforms requires that institutions be adequately staffed with appropriately qualified and motivated personnel. 3.4. In order to improve financial accountability in public finances in Haiti, the CFAA recommends that the Government should: (i) strengthen the legal and institutional framework for public finance management; ( ii)reinforce budget preparation and execution; ( iii)enhance accountability and financial reporting; (iv) strengthen debt and cash flow management; (v) reinforce information systems; (vi) strengthen internal and external controls and legislative scrutiny; (vii) enhance institutional and procurement management capacity; (viii) reinforce procurement operations and market prices; and (ix) enhance the integrity o f the public procurement system. 40 T h e full CFAA and CPAR reports are available u p o n request. 45 Haiti: Public Expenditure Manaqement and Financial Accountability Review 3.1 Background 3.5. (o T h e Government of H a i t i G€) € has made significant progress in strengthening fiscal discipline and improving the efficiency of i t s PFM and procurement systems during the past three years, in a relatively difficult context41. Along with achieving macroeconomic stabilization, the Government has adopted several PFM laws and regulations4’, strengthened the budget preparation and execution process, increased the transparency o f budget information, and strengthened budgetary oversight. The Government has adequately reflected in the last fiscal year’s budget, the policy priorities defined in the Interim Cooperation Framework (ICF) and the Interim Poverty Reduction Strategy Paper (I-PRSP) prepared in 2006. This was done notably through increased allocations for critical sectors related to the implementation o f the “Programme d’ Appaisement Social” The budget preparation process was strengthened by the introduction o f improved coordination and consultation with the Ministries Departments and Agencies (MDAs), including the Ministry o f Plan and External Cooperation [Minist&e du Plan et de la Coopkration Externe - MPCE]. The budget classification system has been improved both o n the expenditure and the revenue sides. The new budget classification i s based o n the administrative and economic nature o f expenditures, and comes relatively close t o meeting international standards. The automated financial management system, SYSDEP and i t s progressive implementation in MDAs have improved the budget execution process. Financial Comptrollers as well as public accountants are being recruited to strengthen the MDAs and accelerate the budget execution process. The audit reports for fiscal years 2001/0243and 2002/03 were also completed and transmitted by the Auditor General [Cour Supkrieure des Comptes et du Cgntentieux Administratf- CSCCA] to parliament. I ( 3.6. I n addition, in response to the recommendations of the ICF, a number of important measures were taken between 2004 and 2006 to strengthen the public procurement system. Most notably, a new Procurement Decree was published on February 14, 200544. While this 2005 Decree fails to address many o f the flaws in the old Decree (1989), i t enabled the creation o f the Commission Nationale des Marchks Publics (CNMP), with five full-time commissioners named with input from the public and private sectors. The Decree also strengthened the commission’s mandate and incorporated many elements o f internationally accepted procurement practices including, most importantly, the stipulation that competitive methods are the norm and not the exception. The Commission i s n o w operational and as an important f i r s t step has produced preliminary and workable versions o f standard bidding documents. Advances in the transparency o f the procurement process have also been achieved through the establishment o f the CNMP’s Web site, which publishes Government contract awards as well as a supplier database, and through the creation o f the Anti-Corruption Unit (Unitk de Lutte Contre l a Corruption - ULCC), as an autonomous entity under the Ministry o f Economy and Finance. 41 Most o f the reforms were implemented during the transition period (2004-2006) after a long period o f political instability. 42 The most important P F M laws and decree adopted were: the decree o f February 16, 2005 related to the formulation and execution o f appropriation laws, The Order o f M a y 19,2005 establishing general regulations for public accounting, the decree o f November 23, 2005 establishing the organization and operation o f the Auditor General o f Haiti, the decree o f M a y 25, 2006 creating the General Finance Inspectorate. 43 The fiscal year in Haiti runs from October 1 to September 30. 44 Dtcret fixant la rkglementation des marches publics de services, de fournitures et de travaux du 3 decembre, 2004 (publit le 14 fevrier, 2005). 46 Haiti: Public Expenditure Management and Financial Accountability Review 3.7. I n order to improve the procurement system further, the Government has also prepared a draft law45to replace the 2005 Decree. This law i s expected t o be submitted to Parliament for ratification in 2007. Based on a review o f the most recent draft o f the law by the World Bank’s Legal Department, i t appears that the law’s adoption would address most o f the failings in the current Decree and that the draft text constitutes a solid legal basis for the establishment o f a modem and transparent procurement system. However, despite these improvements in both the public financial management and procurement systems, there remain significant challenges, as summarized below. 3.8. The budget i s not forward-looking. The link between sectoral policies and priorities and the budget i s very weak. The budget i s only based o n projections o f activity during the previous year and i s not rooted in a medium-term investment plan. As a result, the impact o f recent investments i s not properly taken into account in projecting current expenditures or in measuring the effect that current investments might have on future government operations. In addition, budgetary classifications differ fkom accounting classifications, so that i t i s very difficult to maintain a proper accounting and reporting system. 3.9. There are many off-budget operations. The profusion o f off-budget accounts and activities severely constrains MEF’s ability to control resource allocation and public spending. Despite i t s declining trend, an important part o f public spending i s s t i l l channeled through comptes courants and, “own resources” collected by the MDAs, and special accounts o f the Treasury. I t i s difficult to quantify the extent o f these unreported operations. However, it seriously impairs the Treasury’s efforts to manage the public resources and reconcile the Government’s cash position. Budget execution i s affected mainly b y the lack o f cash flow planning and monitoring andl $Be I weak capacity of the line ministries. Budget releases t o the M D A s are made based on the d i n g s fixed by the Constitution (one-twelfth o f the annual budget appropriations) and not the effective cash Row needs. As a result MDAs have difficulties in executing their activities in the absence o f a budget release based on their real cash flow needs. In particular, this arbitrary constraint on the availability o f funds hampers the implementation o f multi year contracts and discourages procurement planning by the MDAs. The budget execution process i s also laclung a manual. The existing manual i s outdated and does not reflect the recent changes in the Public Financial Management (PFM) system. The deployment o f financial comptrollers to support the MDAs, in accelerating the budget execution process, i s moving slowly. Budget allocations approved for regional departments are made available with significant delays resulting in slow implementation o f activities at that level. 3.10. Formal coordination mechanism to link aid policies, project and programs to the Country’s priorities and budget needs to be established. A large part o f externally financed expenditures i s executed outside the budget, with donors using their own implementation arrangements. T h i s results in poor information flow between the spending ministries and the MEF and the MPCE, poor coordination between the development partners and the Government, the lack o f a clear fkamework for the execution o f those expenditures and the lack o f database related to project based assistance. 45 Avant Projet de L o i fixant les Rbgles Generales relatives aux MarchCs Publics et aux Conventions de Concession d‘Ouvrage de Service Public. 47 Haiti: Public Expenditure Management and Financial Accountability Review 3.11. A sound accounting system, including clear standards and a related automated information system, s t i l l needs t o b e developed. A new accounting framework was adopted recently. It i s expected that a sound accounting system will soon be in place in the Treasury department, remedying one o f the major weaknesses o f the P F M system in Haiti. T o support the implementation o f this framework, the accounting module o f the computerized Government expenditures management system (Systeme d’Informatisation des Depenses - SYSDEP) needs to be developed. Moreover, the link between the different automated financial management software and that used by the MEF to report on Revenue i s not automatic. In addition to the risk o f errors from data transfer, the current system cannot generate a clear and comprehensive report o f the financial situation o f the Country. 3.12. Despite a relatively comprehensive institutional and legislative framework, t h e internal a n d external oversight o f the budget i s still n o t effective given t h e l a c k o f institutional capacity. Internal controls are very weak or not yet functioning. External controls are limited by the capacity o f the CSCCA. The role o f the parliament has been limited to the approval o f draft appropriation laws during the last years. Budget Review Acts as well as annual audit reports o f the CSCCA, when available, were not reviewed. 3.13. T h e adoption o f t h e new procurement law i s a necessary f i r s t step but n o t sufficient, by itself, t o institutionalize t h e recent procurement reforms. M u c h s t i l l needs t o be done in terms o f fleshing out the procedural details o f the reforms and the CNMP’s ability t o enforce them in order to ensure that all procuring agencies adhere to the new requirements o f the legal framework. 3.14. T h e successful implementation o f the P F M reforms requires that institutions b e adequately staffed with appropriately qualified a n d motivated personnel. This i s one o f the main challenges o f the Government as the inadequate quality and quantity o f human resources have been a pnmary impediment to public sector efficiency in Haiti. The capacity o f the Human Resources Unit, recently created in the Pnme Minister’s Office, should be strengthened. T h i s will enable the Unit to achieve i t s mandate notably by establishing and implementing a n e w fiamework for human resource management, including qualifications, slulls assessment and capacity building program, developing a transparent and merit-based procedure for new recruitments and promotions. 3.15. As a result o f t h e weaknesses discussed above, the overall scoring o f t h e Country, under the PEFA and procurement indicators i s relatively l o w in most o f these areas. Table A.3.1 in Annex 3 provides t h e overall PEFA assessment results. A detailed explanation o f the rationale for the ratings for the Performance Indicators benchmarks and the Procurement Baseline Indicators, as well as the underlying information, i s provided in Volumes I1 and I11 o f the PEMFAR. The assessments cover the fiscal years 2003/2004, 2004/2005, 2005/2006. The paragraphs below elaborate o n the main areas in which the Government will need to focus its efforts. 3.2 Legal and Institutional framework 3.16. T h e legal and regulatory f r a m e w o r k f o r t h e public financial management systems in Haiti i s set f o r t h in t h e 1987 Constitution as w e l l as several decrees and orders46prepared a n d 46 A complete l i s t o f a l l different Decree and Orders enacted i s provided in Volume I1 o f the PEMFAR. 48 Haiti: Public Expenditure Management and Financial Accountability Review enacted by the Executive. Although these texts provide an adequate corpus, they have some limitations. The first limitation relates to the coherence o f the different text and laws. This i s due mainly to the absence o f an organic law, enacted by the parliament, which outlines the preparation and execution o f the budget, the public accounting rules and principles, the decentralization of commitment o f expenditures in M D A s as stipulated in the Constitution, and the role and functions o f CSCCA. Certain broad budgetary principles - e.g., budget unity and single treasury account - are also not reinforced by the legal fi-amework. The principle o f performance and results-based management in order to improve the budget execution process and make the MDAs accountable for the results o f their respective departments does not exist in the existing legal and institutional public financial management framework. 3.17. The 2005 procurement decree i s limited in i t s formulation and has a number of gaps. These include an inadequate articulation o f the relationship between Haitian procedures and those applicable t o contracts financed by external assistance and a limited field o f application, as the Decree does not cover concessions or other public-private partnerships, contracts concluded between public bodies, defense contracting, the contracts o f “modernized” public enterprises, or contracts below the thresholds established by the Decree. The 2005 Decree also concentrates a multiplicity o f functions in the mandate o f the CNMP, to include ex-ante reviews, ex-post oversight, audits, protest review and work as a regulatory body, creating duplicate oversight requirements which do not meet international standards. Further exacerbating the situation, the mandatory preliminary opinion o f the CSCCA o n a l l government Contracts duplicates the control exercised by the CNMP and the CSCCA’s dual oversight function, ex-ante and ex-post, also creates an inherent conflict-of-interest. 3.18. I n order to address SQDW of these weaknesses, a draft law was prepared b y the Government and i s expected to be presented to the Parliament for ratification E n 2007. This new L a w will be partially completed by a number o f standard bidding documents, including standard administrative clayses (CCAG) and general technical specifications (CCTG) prepared by the CNMP and t o be published in the form o f an Order by the Prime Minister. 3.3 Budget Preparation 3.19. Despite recent improvements, the budget preparation process i s still hampered by the lack of comprehensiveness; moreover, the budget i s not based on forward-looking plans. The links between sectoral policies and priorities and the budget i s very weak. The Investment Program (Programme d’hvestissement Public - PIP) i s formulated for one year and it i s s t i l l prepared separately from the recurrent budget. The lack o f budgetary envelope ceiling also has undermined the quality o f M D A ’ s proposals and the budget preparation process overall. 3.20. Procurement has not yet been mainstreamed into public financial management in Haiti. Procurement planning i s not part o f the budget process and no organized and consistent data o n procurement i s available. In addition to the lack o f interface between financial management and procurement systems, the absence o f procurement planning can be credited in large part t o the fact that the Budget L a w and financial procedures do not permit multi-year contract execution and, as described in section 3.6 below, budget releases are based on monthly allocations o f one-twelfth o f the annual budget appropriations, rather than any estimate o f cash flow requirements for the period. 49 Haiti: Public Expenditure Manaqement and Financial Accountability Review 3.21. The existing budget classification i s n o t compatible with the accounting classification. Moreover, there are several types o f classifications for preparing the budget and for recording resources and expenditures o n the books; one in particular i s that used by the ministries for their comptes courunts transactions and to take into account the specificity o f some o f their expenditure^.^^. Thus, i t i s impossible to ensure proper consolidation o f all government budgetary operations emanating from the various budget authorization officers and budget managers. 3.22. With r e g a r d t o budget comprehensiveness, there are m a n y off-budget operations; w h i c h might b e b r o k e n d o w n i n t o the following categories: “own r e s o ~ r c e s ” ~expenditures ~, financed by Grants and Loans, and expenditures financed through the special accounts o f the Treasury. Revenue and Expenditures forecasts related to those expenditures are not presented in the Annual Financial Act. The amounts involved might be substantial, particularly for “own resources”. 3.4 Budget Execution 3.23. The most i m p o r t a n t r i s k associated with the budget execution i s the l a c k o f preparation and approval o f Budget Review Acts during the last fiscal years. This led to weak expenditure controls and legislative oversight o f budget execution. . . 3.24. I n o r d e r t o i m p r o v e t h e budget execution process and t h e commitment controls o f expenditures, t h e Government ‘has started t h e deployment o f financial comptrollers in MDAs but the process i s moving slowly. The budget execution process i s also lacking a manual. The existing manual i t outdated and i t does not reflect tbe recent changes in the Public Financial Management (PFM) system. 3.25. P a y r o l l managemen also an area o f concern. T h e l a c k o f discipline in t h e % recruitment o f c i v i l servants has generated saIary arrears. The Government i s conducting a survey to determine exactly the level o f salary arrears but there should be effective sanctions for recruitment made beyond the budget appropriations. 3.26. On externally financed expenditures, t h e Government i s facing a big challenge. A large part o f these expenditures i s executed outside t h e budget. This results in poor information flow between the spending ministries and the MEF and the MPCE, poor coordination between the development partners and the Government, the lack o f a clear framework for the execution o f those expenditures and the lack o f database related t o project based assistance. 3.5 Accounting and Financial Reporting 3.27. The Government has initiated the implementation o f a new accounting and financial f r a m e w o r k (Rkglement GCnCral de la ComptabilitC Publique - RGCP) based o n accrual accounting. Within this new framework, the Treasury Department (Direction du TrCsor - DT) i s 41 Fore example the purchase o f drugs for the Ministry o f Health. The Ministry o f Agriculture has already, for its part, established a n e w accounting classification specific t o its own operations. 48 The “own resources” are resources collected by some ministries for different services rendered : e.g. university enrolment fees, Sales o f drugs or services provided t o patients at public hospitals, Fines (Haitian National Police), Soil analysis (Ministry o f Public Works, Transportation and Communications-MTPTC), express passport fees, etc. 50 Haiti: Public Expenditure Manaqement and Financial Accountability Review responsible for maintaining the general accounts and all public financial management transactions. In order to achieve this objective, a three-year plan to recruit, train and deploy public accountants in the MDAs i s being implemented. 3.28. According t o the RGCP, DT should establish the annual financial statements n prepared f o r the State General Accounts, supported by a General Account Balance. I practice, i t has never been possible to produce the General Account Balance. 3.29. Disbursements occurring w i t h o u t p r i o r requisition appear in t h e General accounts but these operations are n o t subject t o oversight by t h e Treasury, w h i c h i s n o t responsible f o r payments. For the last t w o fiscal years they totaled 24.3 percent49o f the 2005/06 budget, and 23.7 percent5’ o f the 2004105 budget. 3.6 Debt and cash flow Management 3.30. D e b t i s monitored by t h e C e n t r a l Bank o f Haiti (Banque Centrale d’Haiti - BRH) and t h e debt directorate o f t h e MEF but w i t h o u t the direct involvement o f the Treasury Department. As a result, debt flows do not appear on the Government’s book. 3.31. There i s n o r e a l cash f l o w p l a n n i n g and monitoring. According to the Constitution, ceilings for expenditure commitments are limited to one-twelfth o f the annual budget appropriations. This ceiling i s not in line with the effective cash flow needs o f the MDAs, especially given the nature o f the MDAs spending. 3.7 Information Systems 3.32. The Government has made significant efforts t o improve budget execution with t h e installation and deployment within M D A s o f the computerized expenditures management system, SYSDEP. However there are still some weaknesses in the overall system. Although SYSDEP has been designed to manage Budget Expenditures, there are many exceptions that allow the execution o f a large portion o f the budget by other procedures i.e. the investment budget, debt payments, public interventions comptes courants and other budgetary expenditures. 3.33. Moreover the link between SYSDEP and t h e other computerized financial management systems (SYSPAY, used t o process the payroll o f c i v i l servants, SYSPENS, used to process payment o f c i v i l and military pensioners, C H E K used by Treasury t o issue all checks and SYDONIA used by the Customs Department) i s not automatic and there i s n o interface with them. The Government i s s t i l l workmg o n adding three modules: Accounting, Investment operations and Fixed Assets. However, the overall financial management information system lacks a master plan that will provide a medium term vision to organize and coordinate the required financial management system improvements. HTG 5,2 milIions out o f a total o f HTG 2 1,4 millions (payment basis). 49 50 HTG 5,2 millions out o f HTG 22,l millions. 51 Haiti: Public Expenditure Manaqement and Financial Accountability Review 3.8 Internal Controls 3.34. I n t e r n a l controls are undertaken by t h e financial comptrollers and t h e newly created General Finance Inspectorate (IGF). Some MDAs have internal audit units whose controls often duplicate those o f the DCB. Overall, the internal audit systems are very weak in most cases. In others, they are not yet functioning. 3.9 E x t e r n a l C o n t r o l s and L e g i s l a t i v e S c r u t i n y 3.35. E x t e r n a l controls are exercised by the CSCCA. Although it has recently made a determined effort t o clear t h e backlog o f audit reports, external controls are still l i m i t e d by i t s capacity. C S C C A does n o t apply t h e I n t e r n a l Standards on Auditing and has no manuals o f procedures and code o f ethics. A capacity building program for CSCCA’s staff was recently developed. 3.36. With r e g a r d t o legislative oversight, t h e role o f the parliament has been l i m i t e d t o approval o f d r a f t appropriation laws during the last several years. Budget Review Acts as well as annual audit reports o f CSCCA were not reviewed and approved. B o t h chambers o f the parliament (Chamber o f Deputies and Senate) also are limited in their economic and financial analysis capabilities. 3.10 I n s t i t u t i o n a l F r a m e w o r k and P r o c u r e m e n t M a n a g e m e n t Capacity e i t s establishment by t h e 2005 Decree as t h e functional and normative b o d y f o r public procurement in Haiti, the CNMP has been successful in gradually including various elements o f best practice in t h e Haitian public procurement system. To this end the CNMP has strengthened the oversight function to capture all public procurement through the introduction o f a prior review process above a certain threshold, created Ministerial and Specialized Commissions in all Ministries and State Enterprises, increased the transparency o f procurement through mandatory advertising, promoted a higher level o f efficiency and reliability by introducing standard bidding documents, established consistent bidding procedures throughout the system and trained personnel handling procurement within the public service and the private sector. 3.38. However, as mentioned above, some o f the CNMP’s functions and responsibilities create an inherent conflict-of-interest. The conflict emanates primarily f r o m the ex-ante review function o f the Commission and the CNMP’s role as a dispute resolution forum for aggneved bidders. I t would be preferable for the C N M P t o remain involved as a functional and normative body in public procurement through the ex-ante clearance o f contracts, while confining the dispute resolution function to an independent body. The almost complete absence o f procurement planning, which results in large part from the lack o f integration o f such planning with budget preparation and approval, also constitutes a major obstacle to full implementation o f the 2005 procurement reforms. Planning for procurement i s fiuther impeded by the shortage o f trained personnel handling government contracting. 3.39. I n 2006, the CNMP was able t o provide introductory procurement training t o m o r e than 300 public officials. However, Haiti’s institutional capacity to develop procurement professionals remains limited. With the formation o f Ministerial and Specialized Commissions, 52 Haiti: Public Expenditure Manaqement and Financial Accountability Review which are in fact procurement units within ministries and public agencies, the initial steps have been taken, but ad-hoc one-time training will not be sufficient to develop these personnel into procurement officers and professionals. 3.1 1 Procurement Operations and M a r k e t Practices 3.40. Since the adoption o f the 2005 Decree and the establishment of the CNMP there i s evidence that procuring entities are making some effort to follow the new rules and are using more competitive procurement methods than in the past. However, analysis o f available data o n contract awards financed by Treasury funds (vs. donor financing) indicates that: (a) the use o f the single source and/or direct contracting (grC-a-grC) method i s s t i l l frequent and (b) restricted tendering through direct invitation from pre-determined lists (appel d’offi-es restreint) remains the predominant procurement method used by contracting entities. 3.41. W h e n more competitive procurement methods, such as national and international tendering (appel d’offres ouvert), are used bid opportunities are advertised in local newspapers and at the W e b sites o f some procuring entities. However, bids are not yet being advertised at the CNMP Web page created for this purpose. Transparency in the procurement process has been further limited by the discontinuation o f the publication o f government contract awards at the CNMP W e b site - a practice that was in place from December 2004 t o August 2005, when contract awards by multiple procuring entities were published. The splitting o f bids into multiple packages below the threshold for prior review by the CNMP, which are then procured using the informal quotation procedure known as “three p r o forma quotations”, i s another example o f lingering lack o f transparency in the current system. 3.42. Since i t s creation, the CNMP has done a commendable job of standardizing tender documentation in Haiti. The standard documents produced by the ChWP are based on models I . : consistent with W o r l d B a n k and IDB models and reflect the requirements o f the n e w draft Procurement L a w . However, there are n o systems in place t o ensure that procuring entities are using the standard documents properly, without significant modifications. There i s also substantial r o o m for improvement in the preparation o f the technical specifications which are a critical component o f each bidding document, as such specifications are often based on summary or outdated designs w h i c h either fail to provide sufficient details o r provide information that i s incorrect. At the stage o f bid evaluation there i s evidence that some procuring entities do n o t fully understand the evaluation criteria contained in the standard documents and that bids might be awarded on the basis o f the “lowest price”, rather than the “lowest evaluated price”. 3.43. There are no formal provisions applicable to contract administration across the public service in Haiti. Procedures followed for contract administration vary depending o n the procuring entity, with n o central coordination or data collection. W h i l e provisions for dispute resolution are included in a l l government contracts, for Treasury-funded contracts where cases are referred to the CSCCA, disputes are rarely submitted f o r resolution. Contractors and suppliers are apparently reluctant to submit their disputes to the CSCCA, because o f concerns about the timeliness o f decisions rendered by the Court and also about the independence o f the Court, whose members are all former c i v i l servants. A s a result, the local dispute resolution process i s only marginally functional. 53 Haiti: Public Expenditure Manaqement and Financial Accountability Review 3.12 Integrity o f the Public Procurement System 3.44. The system for control and audit of procurement in Haiti i s weak. There are no provisions for internal audit within the procuring entities themselves. Although the CNMP, as the normative body for procurement, has the responsibility for organizing or conducting independent procurement audits, it has not yet taken on this additional task. The external audit function i s the responsibility o f the CSCCA. However, as noted above, the CSCCA also provides an official opinion (avis motivk) on all draft contracts and approves and registers contracts once they are signed, which creates an inherent conflict o f interest in that the CSCCA must audit the same contracts that i t has approved. The appeals mechanisms available to government contractors in Haiti are limited, as most remedies in place are based o n amicable resolution o f the appeal. 3.45. A n important requisite for maintaining the integrity o f public procurement i s transparency and a key measure of transparency i s the ease of availability of information to the public. I n this regard, CNMP’s initial establishment o f a Web site to broaden public access to procurement information i s commendable. However, for reasons unclear, much o f the information on the Web site has not been updated regularly since late 2005. The early reactivation o f the Web site, with an expanded range o f information, i s o f utmost importance. 3.46. The GoH has taken several steps to curtail procurement related corruption, the most important being: (a) creation o f an Anti-Corruption Unit (ULCC) that has been active in the domain o f public procurement and (b) inclusion o f provisions addressing the issue o f corruption, fraud, conflict o f interest and unethical behavior in tender and contract documents. The draft procurement law also envisages inclusion o f fraud- and corruption-related clauses along with defined levels o f responsibilitie e CNMP i s in the process o f drafting a Code o f Ethics for .. c i v i l servants handling procur owever, these measures are only the beginning and the effectiveness o f the ULCC, t h e 8 G W and the CSCCA in combating corruption in government contracting will depend in large pdrt on their.ability to w o r k together to leverage their limited resources for the most impact. 3.13 Policy Recommendations 3.47. Based on the findings o f this chapter,we issue the following policy recommendations: 3.48. I n order to complete the legal framework for financial management, the Government should: 0 Supplement the Decree o n the preparation and execution o f budget laws with a clear and exhaustive reiteration, in a specific section, o f all the broad budgetary principles; and 0 Incorporate the concept o f performance and results-based management in the framework. 3.49. In regard to the legal framework for public procurement, the Government should: 0 Adopt the new Procurement Law in essentially the same form as the current draft; 0 Enact all application texts needed by the Procurement Law; e Enact CCAG and CCTG in the form o f decrees with detailed provisions t o cover the execution phase o f the contracts as well as the award phase; and 0 Coordinate CNMP and CSCCA reviews/controls and prevent any conflict-of-interest. 54 Haiti: Public Expenditure Management and Financial Accountability Review 3.50. To improve budget preparation, the Government should: 0 Review the sectoral strategies and priorities, link them with the budget, and move toward a M e d i u m Term Expenditure Framework (MTEF); 0 Strengthen the procurement and financial management capacity o f the M D A s to provide support for the preparation o f sectoral strategies, their corresponding budgets and procurement plans; 0 Communicate expenditure ceilings based o n the macro economic framework t o the ministries at the beginning o f the budget preparation process in order to allow them to prioritize realistic proposals; 0 Improve existing budget classifications t o make them compatible with accounting classifications, refine the definition o f budget lines containing poverty reduction expenditures, and establish three main sections (Functional classification broken down into sub-hnctions corresponding to the broad strategic intervention themes defined by the sectoral ministries in various technical areas; Administrative and territorial classification; Economic classification or classification by type o f expenditure); and Identify a complete l i s t o f all off-budget operations and incorporate i t into the budget. For autonomous funds, revenue and expenditures, forecasts should be appended as an annex to the budget. 3.5 1. n order to improve budget execution, the Government should: I 0 Clear the backlog o f Budget Review Acts for the last three fiscal years; 0 Accelerate recruitment and deployment o f financial comptrollers in MDAs. This should be done together with the elaboration and dissemination o f a budget execution manual; 0 Complete the survey on salary arrears, and effective sanctions should be taken for recruitment made beyond the budget appropriations; and 0 On externally financed expenditures, elaborate and implement a new accounting and financial fkamework for expenditures. This framework would notably cover: (i) a detailed budget classification; ( ii)a connection with the SYSDEP; ( iii)a specific statute for project accountants who should be accountable t o the MEF; (iv) opening o f a l l needed project accounts at BRH; and (v) establishment o f a data base for physical progress reports o n all investment projects. 3.52. nthe area o f accounting and financial auditing, priority measures include: I 0 Continue implementation o f the action plan for deployment o f public accountants in M D A s and reinforce the role o f the Treasury Department in the management o f the State Accounts; 0 Continue the process o f reducing the coinptes courants and have the public accountants manage the existing comptes courants once they are deployed in M D A s . The details of non-requisition expenditures and comptes courants expenditures should be included as a note to the annual financial statements; and Clear the backlog o f State General Accounts and General Account Balance. 55 Haiti: Public Expenditure Management and Financial Accountability Review 3.53. n order t o improve debt and cash flow management, the Government should: I Appoint a public accountant at the Treasury department t o manage debt and establish an information-sharing circuit between DT and the debt directorate; and 0 Establish a cash planning and monitoring committee, which would project and monitor cash-flow plans in coordination with MDAs o n a monthly or quarterly basis. 3.54. T o enhance information systems, w e recommend the following policies: 0 Elaborate and implement a Financial Management Development Master Plan that will organize and coordinate the required financial management system improvements with a medium-term vision. The plan will identify necessary changes, prepare institutions for those changes, and establish the sequence o f the technical improvements and the reengineering o f the administrative and control procedures. It also will coordinate the technical and financial assistance needs; and 0 In the short term, improve SYSDEP by: (a) reducing the exceptions that allow skipping SYSDEP when processing expenditures, such as agencies’ comptes courants; (b) installing SYSDEP in the other MDAs; and (c) elaborating and implementing the other modules o f S Y SDEP, particularly the accounting application. 3.55. T o reinforce internal controls, the Government should: Deploy financial comptrollers in MDAs, develap, and make available to oversight entities tools and manuals, enabling them t o perform their functions; 0 Define a harmonized framework for the conduct o f internal controls within the technical ministries; and H i r e and train General Finance Inspectors. In addition, the modalities for using the results o f inspection missions and for their publication, as necessary, should be set forth in a regulatory text. 3.56. In the area o f external control and legislative scrutiny, the Government should: Implement the capacity building program o f CSCCA together with a streamlining o f CSCCA staff, so as to ensure that only qualified officers are responsible for operational activities; 0 Develop an action plan to clear the backlog o f audit reports (fiscal years 2003104, 2004105 and 2005106); Strengthen economic and financial analysis capabilities o f the Finance Committees o f b o t h chambers, which should be strengthened for effective legislative oversight; and 0 Submit to the parliament the draft Budget Review A c t withm the prescribed timefi-ame. 3.57. T o enhance the institutional framework for procurement, w e recommend the following actions: 0 The CNMP should relinquish i t s dispute resolution h n c t i o n t o an independent body such as the CSCCA and expand its activities to include comprehensive data collection and procurement audits for use as inputs to formulation o f procurement policy; and 56 Haiti: Public Expenditure Management and Financial Accountability Review In order t o better integrate procurement into the PFM system, steps should be taken to strengthen the procurement planning function. 3.58. In the area o f institutional capacity development, the Government should: Establish an expanded training program needs o n a more sustainable basis to support the development o f a public sector career path in procurement; Mobilize the resources o f both the CNMP and the c i v i l service training center (Centre de Formation et de Perfectionnement des Agents de l a Fonction publique - CEFOPAFOP) in order to facilitate an increase in the proportion o f staff with the procurement s h l l s appropriate to their level o f responsibility; and Develop a program o f introductory procurement training for other public sector stakeholders (parliamentarians, judges, ministers, staff o f CSCCA and ULCC, etc.) as w e l l as private sector operators and c i v i l society organizations. 3.59. To enhance procurement operations and market prices, the Government o f H a i t i and C N M P should redouble their efforts to: Limit contracts awarded through single source and/or direct contracting to a maximum o f 5 percent o f the total value o f contract awards; Discourage the use o f restricted tendering in favor o f national and international tendering (appel d’offies ouvert); Use the existing CNMP Web page to advertise all government bidding opportunities and increase international advertising to broaden the government’s access to foreign suppliers; Ensure that contract awards are made public through the CNMP Web site and/or publications; Require procuring entities to submit t o CNMP, at regular intervals, data o n contract awards below the CNMP review threshold; Discontinue the practice o f awarding contracts based o n “trois pro formas” and replace i t with either an open national tendering process or the use o f competitively awarded “indefinite delivery contracts” or “standing offer arrangements”; Ensure the quality o f bidding documents, including technical specifications, prepared by procuring entities by instituting periodic prior review o f bid documentation by the CNMP; Enhance the training available t o procuring entities on bid evaluation procedures; and Develop guidelines for contract administration and build capacity in CSCCA for dispute resolution. 3.60. To enhance the integrity o f the public procurement system, the Government needs to: Institute internal quality control in procuring agencies to promote internal audits o f procurement, develop preventive measures and increase capacity and efficiency; 57 Haiti: Public Expenditure Management and Financial Accountability Review Build capacity in the CNMP and/or the CSCCA to organize and conduct external (“independent”) procurement audits; Eliminate the conflict-generating activities (issuing an opinion and approving contracts ex-ante) from the CSCCA’s mandate in order t o allow the CSCCA t o focus o n its audit activities; Empower the CSCCA t o handle appeals as part o f a more robust system for dealing with protests and complaints from suppliers; Reactivate the CNMP Web site with more comprehensive information and advertisement o f bidding opportunities; Reinforce collaboration between the CNMP and the ULCC t o develop measures to prevent corruption in government contracting; and Strengthen the working relationship between the CSCCA and the ULCC t o establish a secure mechanism to allow the public at large to report fraudulent, corrupt or unethical behavior without fear o f reprisal. 58 Haiti: Public Expenditure Management and Financial Accountability Review 59 Haiti: Public Expenditure Manaqement and Financial Accountability Review 60 Haiti: Public Expenditure Management and Financial Accountability Review CHAPTER 4 : ECONOMIC SECTOR EXPENDITURE REVIEW 4.1. This chapter presents a review o f expenditure in two priority sectors, agriculture and infrastructure. I t first analyzes the recent trends o f allocations o f public resources t o these sectors over the past five fiscal years, and the extent to which they are consistent with the development priorities o f the Government o f Haiti. It then analyzes the structure o f spending in these sectors and highlights the relative weight and trends o f recurrent versus investment expenditure in these sectors. The chapter also discusses the execution o f spending and i t s determinants over the past fiscal years. The main characteristics o f the PIP in these sectors are discussed. The weaknesses, issues and challenges o f public expenditure management in these sectors are also discussed. The chapter concludes by providing some policy recommendations to improve effectiveness o f public spending in order t o achieve higher economic growth and reduce poverty. 4.2. The key findings o f this chapter are the following: First, the allocation t o the agriculture sector almost doubled in real terms between FY2002-04 and FY2005-07, mainly driven by investment expenditure. Second the budget structure i s dominated by investment spending, on average accounting for more than 68 percent o f the sector’s budget while recurrent spending makes up about 32 percent over the period FY2005-07. Third, the execution rates fell between FY2001-03 and FY2004-06, reflecting the drop in execution o f programs and projects. However, the execution rates were relatively high for wages and salaries and goods and services over the period under review. Fourth, resources allocated to the PIP are projected to triple in FY2006/07 as compared to FY200 1/02, reflecting higher than expected external financing. Fifth, despite improvements during the past three years, several weaknesses remain most notably in the areas o f budget preparation, effective execution o f the investment budget, transparency in public expenditure as w e l l as public procurement issues. T h e chapter recommends: ( i) improving the process o f elaborating the budget by defining a clear calendar o f preparation and separating the policy monitoring structure f i o m the technical structure; ( ii) improving transparency and financial accountability by accelerating the deployment o f an MEF financial controller t o the Ministkre de 1’Agriculture, des Ressources Naturelles et du DCveloppement Rural (MARNDR), (for the recurrent budget). Regarding the investment budget, the improvement in transparency requires closing the existing “current account” used for executing the investment budget and the “suspense account”; ( iiidescribing in detail the priorities in the agriculture PIP and establishing a three year ) rolling investment budget by objectives; (iii) establishing a Procedures Manual for a transparent management o f the sector’s “own funds”; (iv) introducing procurement procedures for tenders at the beginning o f the fiscal year for the management o f purchases o f the MARNDR; and (v) reinforcing human resource management including through technical assistance to be provided t o staff o f the MARNDR in the areas o f budgetary management and public procurement. 4.3. With regard t o the infrastructure sector, the key findings o f the PER are: First, the allocation to the infrastructure sector more than tripled in real terms on average between FY2002- 04 and FY2005-07, mainly driven by increased external resources allocated to the investment budget since the re-engagement o f donors in 2004. Second, the structure o f the infrastructure budget i s dominated by investment spending, reflecting the nature o f the infrastructure sector, mainly composed with economic activities, which involve investment spending. Third, while the investment budget tripled on average over the FY2002-04 and FY2005-07, the recurrent budget increased slightly by 5 percent in real terms over the same period. This suggests that investment spending i s not backed by sufficient associated recurrent allocations in the budget, reflecting the limited Government resources, poor planning o f recurrent resources associated with investment programs, as well as the disconnect between the preparation o f the investment budget and the 61 Haiti: Public Expenditure Manaqement and Financial Accountability Review recurrent budget. Fourth, because o f limited public resources and little involvement o f the private sector, it i s critical that resources be allocated to the best uses and be used as efficiently as possible in the infrastructure sector. The management o f resources allocated t o MTPTC, in particular the recurrent budget, could be improved and therefore used to help save some resources to be reallocated for other purposed. The staffing structure and transfers o f public resources to public utilities are two areas where savings could be achieved. Fifth, the analysis o f intra-sectoral allocation o f expenditure shows that over the period FY2005-07, o n average, 60 percent o f the total recurrent budget o f MTPTC was allocated to finance the Ministry’s general administration services. T h i s left few resources t o other services and entities under the MTPTC supervision, and might have affected their capacity to perfonn activities. Sixth, the execution o f spending in the infrastructure sector was extremely volatile and fell t o relatively l o w levels over the period FY2002-06, reflecting the volatility o f the execution o f the investment budget. T w o major factors explain the l o w execution o f the investment budget: (i) the l o w absorption capacity o f the investment budget in the MTPTC due mainly t o slow execution o f programs and projects, human resource constraints, and limited technical capacity and ( ii) the poor capacity planning due partly to a lack o f qualified staff in MTPTC. Seventh, the infrastructure PIP i s mainly financed by donors. More than 90 percent o f the spending o n infrastructure in the FY2004-06 PIP was financed by foreign aid. Aid-funded allocations to the infrastructure PIP almost quintupled in two years, between FY2004-05 and FY2005-07. Eighth, despite some recent improvements, weaknesses and challenges s t i l l exist in the areas o f budget planning, efficiency, transparency and accountability o f the execution o f the investment budget, resource management and procurement. A specific issue concerns management problems o f autonomous agencies such as C O N A T E L (telecommunication sector), and FER (road transport sector), which prevent them fi-om delivering high quality infrastructure regulation. Policy recommendations to improve efficiency include: (i) prepare a full-fledged costed and integrated infrastructure strategy with sub-sectoral action plans, which are linked t o spending targets; ( ii)prepare an integrated multi-year PIP, which reflects the translation o f the integrated infrastructure strategy into resources requirements; ( iii) phase out by 2009 the use o f cornptes courants; (iv) connecting new investments with maintenance spending and support to community-based organizations managing rural infrastructure; (v) strengthen the functioning o f the Road Maintenance Fund (FER); (vi) strengthen the functioning o f the Road Maintenance Fund (FER); and (vii) improve donor involvement and better coordinate their intervention. 62 Haiti: Public Expenditure Management and Financial Accountability Review 4.1 AGRICULTURAL SECTOR 4.1.1 Background and sector objectives 4.4. Over the last two decades, Haiti's rural economy has Figure 6. Agriculture Value Added in 2004" (YOof total) declined dramatically. Agriculture, which generated about 40 percent of GDP in the early 1990s, contributes today to about 25 percent of GDP. The contribution of agriculture to Haiti's value-added compares poorly to comparable LICs. I t stood below the performance of Burundi, Togo, Rwanda, Niger, Benin and M a l i (see Figure 6). Agriculture exports fell from 28 percent of total exports in 1981 to * in 2003 for Niger and Chad. only 6 percent in 2004. Chronic Source:WDI,2006. underinvestment in the rural public Figure 7. Economically Active Population in Agriculture goods (e.g. infrastructure, public in 2004 (YOof total) services and programs), low public Burkina Faso Rwanda resources, poor security, natural Burund, Nioer disaskrs and ineffective natural CkiiiTea yiay I Ethmpla resource management have steadily MzaMique mi depleted the rural productive base. Tanzanba j Trade liberalization in the 1990s % ;: l M d ascar %negal included n o transitional or pro-active Chad CarrbOdB 1 I Zarrbia export measures, suddenly making Viet Nam Zimbabwe Haiti one o f the most open markets Haiti Togo Ghana j in a l l o f L a t i n America, yet leaving Camoon I NBuritania the rural economy without the means to adjust or to facilitate labor market shifts. L a c k o f rural Benin Ngeria 1 0 20 40 7i 60 ~ 80 I 100 infrastructure, limited access to Economically Active Population in Agriculture (YOtotal) 4.5. Yet, agriculture remains by far the most important economic and social activity, and Haiti i s still predominantly a rural country. Almost two-thirds o f Haitian households lived in rural areas (4.7 million people residing o n around 800,000 farms)52. Agnculture is the main economic activity and a source o f income for two-thirds and one o f t w o workers, respectively, in rural areas and in the country.53 According to FA0 data, agriculture employs up to 60 percent o f the economically active population in Haiti, higher than the figures for Togo, Benin, and 51 See Haiti. Agriculture and Rural Development. Diagnostic and Proposals for Agriculture and Rural Development Policies and Strategies. World Bank. Report No. 36785. October 5, 2005. 52 / Source: I H S I ; Recensement 2003 53 I Ibid. 63 Haiti: Public Expenditure Manaqement and Financial Accountability Review Mauritania; but far lower than those o f Rwanda, Burundi and Niger, countries with the same income level (see Figure 7). 4.6. Given i t s large economic and social importance, the ICF and the I-PRSP considered agriculture as one priority sector for public intervention. The I-PRSP indicates the need t o ensure the availability o f a p c u l t u r a l products that ensure food security and increased incomes. The Government action will give priority t o improving farmers’ access t o credit and basic farming inputs, rehabilitation o f agricultural infrastructures, and security o f land ownership; this security i s necessary t o make the best o f policies aimed at promoting investment in irrigation, public transportation, communications and storage and marketing infrastructures. This priority will also be given to products promising the greatest possibilities o f profits. M o r e specifically, MARNDR will design and implement a strategy designed t o increase the production o f basic foodstuffs so as to enlarge the supply o f food as quickly as possible. 4.7. Meeting the objectives set forth to the agriculture sector would require both the provision of adequate resources to the sector and an efficient use of these resources to ensure that they translate into improved economic and social outcomes. In the absence o f a comprehensive amculture strategy including spending targets, t h s chapter seeks to: (i) define the current structure o f expenditures in the amcultural sector and i t s consistency in regard t o stated priorities; (ii)measure progress accomplished and remaining weaknesses concerning financial transparency and the execution o f the budget; ( iii)examine the level o f transparency in the current process o f awarding public contracts and its conformity with the new regulations adopted in late 2005; and (iv) propose some policy recommendations in order to reinforce the current expenditure efforts. 4.1.2 Structure and trends in agricultural expenditures 4.1.2.I Total budgetaly allocation 4.8. Total annual allocations to the Ministry of Agriculture increased by more Table 23. Budgetary allocations (in real HTG millions) than 80 percent in real terms o n average I Averages between FY2002-04 and Fy2005-07 (see Table 23), in line with the large increase in the 2002/04 200907 Government overall budget, reflecting the large Agriculture total budEet 757.4 1374.6 increase in allocations t o programs and projects Agriculture Share in - 3.3% 4.4% total Government the Government’s resolve to provide more budget resources to the agriculture sector (a priority Source : L e Moniteur, Journal Officiel de la sector) as stated in the I-PRSP. Rtpublique d’HaYti and Staff Calculations. 64 Haiti: Public Expenditure Manaqement and Financial Accountability Review Figure 8. Share of Agriculture in Total Budget in 4.9. However, the increase in selected countries, 2000-2005 allocations also reflects the l o w levels o f allocations to begin with. In addition, the increase reflects more the large resources devoted to programs and projects mainly funded by the donors that the Government’s resolve to provide more resources to MARNDR. In fact, the level o f allocations i s not the outcome o f a thorough assessment o f the Niger Benin Haiti MARNDR needs. The allocations are pre-determined as a large share o f the Source: MARNDR budget i s dominated by the investment budget, which i s funded by the donors (see below). Also, MARNDR has no influence o n the resources provided to the sector as i t barely participates in budget discussions. Moreover, because o f the lack o f a comprehensive strategy with clearly defined spending targets, it i s difficult to assess the extent t o which the allocations are sufficient to perform the objectives set forth in the I-PRSP. But at less than 5 percent o f total Government budget o n average over FY2005-07, the agriculture sector share seems below the figure o f A h c a n countries at the same level o f income per capita and comparable economic structure: for instance, in Niger it stood above I O percent on average over 2000-02; and in Benin at 7.3 percent o n average over 2002-03 (see Figure 8). 4.1.2.2 Budget structure 4.10. T h e structure of the agriculture Table 24. Budgetary Allocations Structure ‘(in sector budget i s dominated by t h e investment budget (programs and projects), which accounted for more than 67 percent o f total budget o n average between FY2005-07 while the operating budget accounted for 226.0 214.0 slightly above 32 percent (Table 24). 43.7% 25.8% Other current exp. 4.11. T h e large concentration of the MARNDR budget o n programs and 313.9 263.6 57.5% 32.2% projects reflects the impact o f donor Programs and projects 443.4 1111.0 allocations over FY2005-07. During that 42.5% 67.8% period, the budget allocated to Programs and Sources: L e Moniteur, Journal Officiel de la Rkpublique projects more than doubled in real terms d’Hafti and Staff Calculation while total recurrent budget declined. This raises three critical issues. First, the increases in the investment budget do not seem to be backed by sufficient allocations in the recurrent budget. This reflects the lack o f Government resources to support the programs and projects financed by donors. I t also mirrors the lack o f an integrated budget process marked by a parallel process o f preparation o f the investment budget and the recurrent budget. Second, the programs are only very partially implemented, reflecting the l o w capacity o f the MARNDR to execute large resources flows. Third, the pre-determined nature o f the investment budget (programs and projects) limits the ability o f MARNDR to have full control o n the implementation o f the agnculture policy and prevent it t o influence economic outcomes. The lack o f coordination o f donors’ activities in the agriculture sector and the absence o f full 65 Haiti: Public Expenditure Manaqernent and Financial Accountability Review control by MARNDR o f donors’ operations make i t harder for MARNDR to have a clear picture o f the agriculture policy and i t s implementation. As a result, MARNDR has little impact on economic outcomes despite the large share o f agriculture in Haiti’s economic activity. 4.12. C u r r e n t expenditure allocations concentrate o n wages and salaries (82 percent o f t o t a l recurrent), in payment o f the employees w h o w o r k in MARNDR. The share o f goods and services in total expenditure i s l o w compared to normal standards, despite low employment and wages (Table 25). This signals the lack o f room for MARNDR to maneuver and use its budget as a policy tool. Recurrent resources provided are so l o w in view o f the wages and salaries payment obligations that MARNDR cannot arbitrate between alternative use o f i t s resources (wages versus goods and services). Constrained by i t s obligations to meet its emergency and short-term needs (payments o f wages and salaries), MARNDR thus operates o n a routine basis without a medium-term objective o f influencing the economic development o f Haiti. 4.13. T h e l o w level and wage concentration o f current expenditure reflect, as in other sectors, the very binding constraints imposed by the Government’s o w n revenues and the difficulty o f f u r t h e r curtailing wage expenditures. Limited fiscal space results in l o w resources allocated t o the sectors, including MARNDR. Table 25. Structure o f MARNDR’s Operating Budget (allocations in real terms HTG millions), FY2005-2007 GENERAL HARDWARE 22..3 7Yo 31.8 9Yo TANGIBLE ASSETS 1.4 0% 10 2 3% INTANGIBLE ASSETS 0.1 0% 0.1 0% 304.9 100% 371..1 100% Sources : MARNDR data ;L e Moniteur, Journal Officiel de la RCpublique d’Ha1ti and Staff Calculations 66 Haiti: Public Expenditure Management and Financial Accountability Review 4.1.2.3 Budget e.xecution trends 4.14. I n contrast to the priorities stated in Table 26. Trends in MARNDR’s budget allocations, effective expenditures Expenditures, (in real HTG show a very different performance. %le real recurrent expenditures roughly mirror their expenditure allocations, expenditure on programs and projects declined sharply in real terms, in contrast to the intended very large ALLOCATED increase in the budget allocations (Table 26). Salaries 226.0 21 1.9 Current expend. 87.9 50.8 4.15. As in most o f the sectors, wages and Projects 443.4 447.6 salaries are fully executed while programs TOTAL 403.7 312.3 and projects recorded a decline in their EXPENDITURE execution. Three factors explain the l o w Salaries 209.5 207.2 Current expend. 65.3 37.8 execution o f programs and projects in Projects 128.9 67.4 MARNDR over FY2004-06 period: (i) poor TOTAL 81% 52% budget planning; ( ii) limited absorptive capacity EXECUTION RATE o f the Ministry o f Agriculture; and ( iii)lack o f Salaries 94% 98% Current expend. 90% 76% cash flow planning and monitoring. First, on the Projects 122% 26% one hand, MARNDR lacks the capacity to Sources: MARNDR data; L e Moniteur, Journal effectively and accurately prepare and plan i t s Officiel de la Rbpublique d’HaYti and Staff budget needs (see beiow). On the other hand, i t Calculations barely influences the programs and projects o f the sector, which are mainly implemented by 1 ie donors. Put together, the aIlocations to the investment budget (programs snd projects) are thus beyond MARNDR’s control. The execution o f spending are thus disconnected from the allocations and might vary significantly from one period (FY200 1-03) to another period (FY2004-36). Second, MARNDR’s limited absorptive capacity faced the huge flows o f donor resources provided to the agnculture sector over the period FY2004-86. The inability o f MARNDR to execute large flows o f resources resulted in l o w execution rates over that period. Third, the ceilings fixed by the Constitution (one-twelfth o f the annual budget appropriations), which determine budget allocations does not account for seasonality o f some agnculture programs. As a result, some agnculture programs are not fully implemented as MARNDR’s lacks the financial resources to execute them. 4.16. An intra-sectoral analysis of expenditure reveals that spending i s concentrated on few items, with fuel and lubricants representing almost 30 percent o f MARNDR’s non salary operating expenses (Table 27). The extent to which this structure o f spending reflects MARNDR’s strategic priority o f ensuring food security and increased incomes o f the poor i s unknown as the sector does not have specific spending targets. 67 Haiti: Public Expenditure Manaqement and Financial Accountability Review Source: MARNDR data 2002-2003 BtJDGETBY AMOUNT SPENT Allocated Spent Rate Part 3 12 Fuel and lubricants 9,690.3 7,578.6 78% 20% 304 Parts and accessorics for vehiclcs 8,048.8 4,624.6 57YO 12% 3 10 Fertilizer, plant material and secds 5,094.6 4,323.6 85Yo I190 47 1 Hydraulic works 15,270.0 4,295.7 28% 11Sb ” 251 Vehicle maintenance 3,432.@ 1,923.8 56% 5% 303 Veterinary tools and supplies 2,023.7 1,858.5 92% 5% 1 240 Payments to training institutions abroad 2,555.4 1,534.2 60% 4% 702 Subsidies for autonomous organizations 3.000.0 1,412.8 47% 4% 31 1 Insecticides, disinfectants, and other chemical inputs 1,400.8 1,235.8 88% 3% TOTAL 28787.5 76% 4.1.3 Public Investment Program (PIP) Table 29. Shares in Budget Execution (In 4.17. The scenario regarding investment expenditures i s somewhat different in the PIP. Figures pertaining to investment programs in the agncultural sector are incomplete, at most preliminary, and can be confusing at times. For % o f total expenditure 52% 67% consistency, one single source o f data has been Current Expenditure 65.3 37 8 % o f total expenditure 16% 12% reviewed in this report, the PIP prepared by the 128.9 67.4 Ministry o f Planning and External Cooperation (see above). Sources : MARNDR data ;Le Moniteur, Journ: I 4.18. Our analysis focused mainly on the Officiel de la RCpublique d’Haili and Staff 2004-2005 and 2005-2006 fiscal years, for which Calculations 68 Haiti: Public Expenditure Manaqement and Financial Accountability Review partial data was gathered. 4.I.3. I Allocations 4.19. The agriculture PIP reveals three major Table 30. Investment Budget (PIP), yearly allocations (In characteristics. First, yearly PIP allocations more than tripled in real terms in FY2006-07 compared to FY2005-06, mainly driven by donor resources. However, as in other sectors, this also reflects the l o w levels to begin with. Second, the PIP i s largely driven by donor resources, which accounted for more than 90 Source: MARNDR and MPCE data percent o f total allocations o n average over FY2004-06 (Table 30). Third, the provision o f the Government’s own resources (national treasury) to finance the agriculture PIP has been volatile, reflecting the lack o f predictability o f Government’s resources. While the allocation from the treasury declined significantly by more than 7 0 percent in real terms between FY2004/05 and FY2005-06, i t was multiplied by more than six in FY2006-07, a year marked by a large increase in foreign aid. T h i s signals the fact that the counterpart o f government’s resources allocated to finance the programs and projects evolve mainly in response to donors’ financing. MARNDR seems not t o possess an independent agriculture strategy. 4.1.3.2 Execution of the PIP 4.20. There i s no reliable data on effective PIP expenditures in Agriculture, a signal o f shortcomings in the budgetary and monitoring process. However, a recently published Government report, which assesses the implementation o f the ICF indicates that as o f March 2006, only about US$30.2 m i l l i o n was disbursed out o f US$85.8 m i l l i o n committed mainly for investment projects and programs in the agriculture sector (i.e. a disbursement rate o f 35 percent). The report points out that despite this l o w disbursement rate, significant investment programs have been implemented to reinforce agriculture production and productivity as w e l l as food security. These include among others: (i) investment in rehabilitation and maintenance o f irrigated areas; (iirehabilitation o f rural roads; and ( ) ii i) projects and programs in the areas o n agriculture production, irrigation, and marketing o f local products. 4.1.4 Weaknesses, Issues, and Challenges in Public Expenditure Management in Agriculture Sector 4.21. Budget preparation at the Ministry o f Agriculture i s practically disconnected from the existing sector policy framework. The investment budget i s more a compilation o f the priorities o f each donor supporting the sector, and the operating budget i s conceived through a projection o f the budget executed the year before. When a rational programming effort i s undertaken, i t i s generally undermined by public finances and absorptive capacity constraints. For instance, during FY2005-06, an innovation was introduced linking the budget to objectives: each o f the Ministry’s entities was asked to detail the additional resources needed to meet the objectives defined in comparison to the initial situation. T h i s exercise was conducted in a similar manner in FY2006-2007. As shown in Table 3 1, both exercises resulted in budgets that widely exceeded the allocation approved by the Ministry o f Economy and Finance (MEF). 69 Haiti: Public Expenditure Management and Financial Accountability Review 4.22. I n 2006-2007, less than 10 percent of t h e amount requested Table 31. Comparison of Budgets Requested and by MARNDR was allocatedby the Allocated (in nominal HTG million) MEF and approved by Parliament (with only about 4 percent o f the investment budget). Such a Investment 418.6 60.7 15 procedure o f budget elaboration 0 c.l (Treasury) constitutes a waste o f resources and Total 1,209.4 302.7 25 discourages Ministry staff, which Operating 454.4 265.5 58 considers i t o f n o use. Considerable Investment 4,480.1 189.6 4 work will have t o be executed to 0 (Treasury) integrate agricultural policy I s Total 4,934.5 455.1 9 orientations into a realistic and Sources: MARNDR and MPCE data and Staff Calculations. operational budget. The Worlung Group o n Agriculture created at the MARNDR i s a positive initiative that goes along these lines. 4.23. On t h e one hand, t h e execution o f the operating budget has m a r k e d l y i m p r o v e d in recent years with t h e computerization o f expenditure m o n i t o r i n g a n d the elimination o f p r i o r control f r o m t h e C o u r des Comptes. Further progress must be made in accelerating the spending process. 4.24. On the other hand, much remains t o b e done in t h e area o f managing t h e investment budget, t h e t o t a l amount o f w h i c h i s superior t o that o f t h e operating budget w h e n salaries are excluded. There are n o written procedures and the chain o f personnel involved shows weaknesses in their knowledge o f the procedures. There i s n o systematic accountancy tracking o f investment expenditures at the Ministry o f Agriculture’s level. Project evaluation and monitoring are practically non-existent and are carried out solely on the basis o f reports submitted by project directors. Projects financed by foreign aid are for the most part managed by autonomous entities (IDB, IFkD), and there i s n o formal relationship between these project units and the Ministry’s administrative structure, whether for monitonng and evaluahon or for their management. Therefore, the Ministry o f Agnculture cannot develop a global vision o f actions within the sector. 4.25. F u n d s that are generated by the M i n i s t r y (so- Table 32. Generated Funds in Ministry of Agriculture (In called “own resources”) are generally managed directly by staff responsible f o r t h e unit that generates t h e m and t h i s allows the unit t o partially o r totally renew i t s stock o f goods and services sold (vaccines, fry fish, as examples). Only the provision o f well-digging services generates funds that are deposited in the Ministry’s current account for G operations D bearing the Source: MARNDR signature o f the Minister and the administrator. Four main sources o f generated funds are available in the Ministry o f Agnculture, as shown in Table 32. They represent less than 1.5 percent o f total current expenditure o f Ministry o f agriculture (including salaries). 70 Haiti: Public Expenditure Manaqement and Financial Accountability Review 4.26. Despite efforts t o improve transparency in budget execution, t h e execution o f spending in MARNDR i s s t i l l affected the use o f coinptes courants There are actually two main current accounts in t h e MARNDR o n w h i c h there i s a j o i n t signature o f t h e M i n i s t e r and t h e administrator: an account f o r deposits accruing f r o m financial proceeds o f grants provided by Japanese aid (KR2 grant) and a main current account called “Institutional support”, that received a l l generated funds, consolidated advances for all investment projects financed within the framework o f specific programs, and special advances f o r current expenditure through “interventions publiques” (Table 33). Table 33. M a i n currents account in Ministry o f agriculture institutional Suppon Account Total KK2 TOTAL expeiiditur 7 ‘ ” account <;encrateti <‘urrent Trallsitoire Protection 51ilit. e funds exp. civile IMmobil. 2003-2004 Opening statement 372 645 2 939 924 Inflow o f funds 906 945 7 698 142 Resources 7 943 383 1 279 590 10 638 067 31 778 696 Uses 2495756 1004051 9470399 23444657 303958658 8% Statement on 30’09’04 ___ -- 5 441 626 __ . 539 275 _.. ._1 - 167 668 8 334 039 5 447 626 275 539 1 167 6G8 0 0 8 334 039 Inflow offunds 0 3770796 3947322 8965216 1500000 25 901 450 5447626 4946334 5 114989 8965216 1500000 34 235 489 4 821 635 3 703 269 3 998 544 8 896 236 0 29 121 496 372 695 908 8?4e 625991 343066 1 116445 68980 1500000 0 5 113 993 Inflow of funds 403 586 1 244 314 630 749 23 508 487 458 206 3 430 000 31 550 405 1029577 1587380 1747195 23577467 1958206 3430000 36664398 841930 1543442 1742373 22840352 1928683 3 187630 35370227 360330000 lo’% Source: MARNDR data. 4.27. Public expenditure through current accounts has increased in t h e last three years, f r o m 8 percent expenditure t o one o f 10 percent o f the t o t a l budget in t h e agricultural sector. I t i s specifically due to the special integration program o f demobilized militaries financed through public intervention. 4.28. I n addition, about 80 other current accounts service investment projects, w h i c h are financed by donors o r resources f r o m the Treasury. At the level o f the MARNDR’s Administrative Directorate, there i s n o updating inventory o f these accounts and neither i s there any regular and systematic monitoring o f these accounts. 4.29. Public procurement issues also hamper budget management in t h e agriculture sector. T h e Ministerial Commission f o r Public Procurement (Commission MinistCrielle des M a r c h b Publics) has been in operation f o r f o u r months (October 2006) and has examined 71 Haiti: Public Expenditure Management and Financial Accountability Review approximately ten projects for public procurement. T w o have been returned t o the entities having prepared them; the others have been approved by the ministerial commission for transmission t o the CNMP. In accordance with the terms o f existing regulations, only proposals for public procurement with amounts o f more than 800,000 gourdes are examined by the commission. Proposals for smaller amounts can be used for purchase, o n the basis o f a comparison o f pro-forma invoices. 4.30. I n practice, all procurement proposals examined b y the ministerial commission for agriculture have been prepared by project teams supported b y external donors (PIA, PPI, PICV). All procurement contracts attributed on MARNDR’s operating budget funds have been purchases of less than 800 000 HTG. 4.3 1. T h e analysis o f purchases (less than 800 000 HTG) on MARNDR’s operating budget for fiscal year 2005-2006 shows that on 226 purchases amounting to a total of nine million gourdes, which represent 26 percent o f MARMDR’s non salary current expenditures, 83 have been carried out with a single company: Elf Oil d’Haiti, for purchases o f fuel and lubricants. T h i s company by itself accounts for 38 percent o f purchases and 70 percent o f payments (more than 6.6 m i l l i o n gourdes). Three companies account for more than 50 percent o f purchases and nearly 80 percent o f payments. But in terms o f cumulative amounts, only purchases with the Elf company are over the authorized limit for purchases without tender (800,000 HTG). This particular situation concerning Elf Oil results f r o m the fact that the fuel market in Hayti i s a controlled market (no price competition, prices determined by the government). In all, 43 different providers have benefited from purchases by the MARNDR, with an average o f 217,950 gourdes by provider. 4.32. The ministerial commission for public procurement seems insufficiently articulated at the present time with the National Commission. It has not been officially consulted for example*regarding the project for the establishment of a public procurement code. The General Directorate o f the Ministry o f Agriculture seems however to have provided its own comments on the project for establishing a code, independently o f those made by the ministenal commission. 4.1.5 Policy Recommendations 4.33. Based on the findings o f the PER, w e issue the following policy recommendations: 0 Increase allocation of resources to the agriculture sector. While the budget allocations to MARNDR increased over the past two years, they are s t i l l l o w in comparison with countries at the same level o f income. For the agriculture sector to drive economic growth and play a key role in the poverty reduction strategy, resources allocated to MARNDR would need t o increase so as t o enable i t to fully execute the sector programs and projects. However, efficient use o f public resources requires that the sector’s absorptive capacity be increased. This could be done by reinforcing administrative and technical capacity to execute the programs and projects. A focus o n strengthening MARNDR staffs capacity in the areas o f budget processes and procurement procedures would significantly improve its ability to execute agriculture programs and projects. Prepare a full-fledged costed and integrated agriculture strategy with sectoral actions plans, which are linked to spending targets. This requires that the sector should adopt an integrated approach, which involves a l l actors at the Ministry concerned by budget operations issues. A rigorous process o f spending planning should be implemented. T h i s requires that MARNDR improve its budget preparation process by elaborating a clear timetable and allocating more time to budget planning (both resources 72 Haiti: Public Expenditure Management and Financial Accountability Review expected and expenditure planned t o be executed). This will enable the sector to request realistic budget allocations and avoid the wide difference observed between budget proposals and budget allocated. Prepare a sectoral multi-year PIP, which will translate the integrated agriculture strategy into resource requirements. The sectoral PIP should be the product o f a thorough costing o f programs and projects and prepared jointly with the Ministry o f planning and the MEF. A starting point for the preparation o f the PIP i s the existing aggregate PIP, prepared by the Ministry o f Planning. Thus, MARNDR should use the information provided in the aggregate PIP to design a detailed sectoral PIP that will ultimately serve as a basis for reinforcing the overall Government’s PIP. The financing o f the agriculture PIP should be secured. This means that the PIP should identi@ donor current and future resources available for its financing and the Government’s recurrent resources to support the investment programs. M o v e gradually toward an integrated budget preparation process. Ensuring that the integrated sectoral strategy (to be prepared) i s effectively implemented and produce outcomes requires that MARNDR put in place a medium-term expenditure framework (MTEF). However, given the current status o f budget preparation in MARNDR, this i s a medium-term objective. I n the short-term, focus should be on reinforcing the preparation o f the annual budget and introducing gradually a medium-term approach. Improve data collection. Budget execution i s hampered by the lack o f data or the poor quality o f existing data. A first step t o improve data collection i s for MARNDR t o work closely with the MEF (for the recurrent budget) and the Ministry o f Planning (for the investment budget) so as to regularly collect budget data available in those ministries and reconcile i t s data with those o f the MEF and M i n i s t r y o f Planning. A second step involves strengthening the statistical department o f MARNDR. This could be done quickly through a specific capacity building program focusing on data collection and analysis. This second step will be the starting point o f a full-fledged capacity building program in budget management, which MARNDR should install. Improve transparency and accountability in the use of public resources. This implies three actions o n the part o f MARNDR: (i) close the existing current account used for executing the investment budget; ( ii)establish manual procedures for a transparent management o f the sector’s “own funds”; and ( iii) introducing procurement procedures for tenders at the beginning o f the fiscal year for the management o f purchases o f the MARNDR. Promote private investments. Agriculture cannot be an engine o f dynamic growth nfact, Government’s interventions unless there i s a strong participation o f private sector. I should facilitate private investments t o ensure private-sector led growth. This could be done through public sector support to “partnerships” between small producers and buyers o f their products.54 Reinforce human resource management. I n the short-term, technical assistance would need to be provided to staff o f MARNDR involved in budgetary management and public procurement issues. In the medium-term, the Ministry o f Agriculture would need to design a full-fledged human resource management plan that identifies the needs in particular in the area o f budget management. 54 T h i s policy recommendation i s drawn from the World Bank report o n Diagnostic and Proposals for Agriculture and Rural Development Policies and Strategies. Op. Cit. 73 Haiti: Public Expenditure Manaqement and Financial Accountability Review 4.2 INFRASTRUCTURE SECTOR55 4.2.1 Background and sector objectives 4.34. Haiti’s infrastructure coverage and quality are poor and constrains economic growth and poverty reduction. Table 34 presents information on different infrastructure sectors: electricity, roads, water and sanitation, and information and communications technologies (ICT). Despite improvements in some access rates, between the early 1990s and 2004, access t o all types o f infrastructure in Haiti remains below the average levels for Latin America and the Caribbean, (LAC), sub-Saharan Afnca (SSA) and low-income countries in general. In 2004, electricity consumption in Haiti was 31 KWh per capita - well below levels o f an average sub-Saharan Afi-ican country. When Haiti’s figures are compared with those o f successful Afncan countries - such as Mauritius, Ethiopia, Ghana, and Senegal - the gaps are even more apparent. For instance, access to electricity in H a i t i i s much lower than that o f Ghana. Despite improvement in access t o water, Haiti i s well behind other countries, with an average access rate o f 67 percent compared t o 89 percent for LAC countries in 2002. Similarly, access to improved sanitation facilities i s only 34 percent in H a i t i compared to 36 percent in sub-Saharan Afnca. The same applies t o access to telecommunications and road networks. At 64 per 1,000 in 200456, telephone density in Haiti i s well below average for sub-Saharan Africa: 84 per 1,000. H a i t i i s poor in terms o f other information and communications technologies as well. Only 59 o f every 1,000 individuals use the internet, which i s below the average values for sub-Saharan African countries and low-income countries. Only 18 percent o f all Hditian roads are paved, compared to 35 percent and 100 percent in Botswana and Mauritius, respectively. In addition, it i s worth noting that in Haiti, only 20 percent o f the paved roads are considered acceptable condition, 10 percents in go ition, and 10 percent in fair ~ o n d i t i o n . ’ ~ Improved Water Source Access 53 49 82 67 58 89 Improved Sanitation Facilities Access 15 32 68 34 36 74 Road network ( !a per 1000 people) 0,54 2,15 6,27 0,40 0,50 1,09 Telephone Density (per 1000 people) 7 10 61 64 84 499 Internet users (per 1000 people) 0 0 0 59 19 115 Source: WDI 2006. 4.35. The Government has made infrastructure a key sector o f i t s growth and poverty reduction strategy?8 However, it lacks a comprehensive and integrated infrastructure 55 Infrastructure i s here defined as a sector including transport, water and sanitation, electricity and telecommunication. But, for transport sub-sector, only road issues are addressed here, as it i s the main transport mean in Haiti. 56 It i s the telephone density for fixed lines and mobile phones. ""Strategic pour l e Secteur des Transports 2006-2011” (November 2006). 58 T h e I-PRSP states that “Sustained economic growth i s necessary for the success o f any poverty reduction program. But economic growth cannot be sustained if basic physical infrastructures are lacking or are o f poor quality.. .”. 74 Haiti: Public Expenditure Manaqement and Financial Accountability Review strategy with clearly defined spending targets.59 Efforts are oriented in that direction though they are still limited. During the donor conference in Madrid, Spain in November 2006, authorities presented the Government's infrastructure program and financing needs to implement i t over the next five years. Significant levels o f investment would be required t o increase the infrastructure network coverage and the quality o f services. Maintenance i s also a major issue, especially in the road sub-sector, g v e n that road rehabilitation in H a i t i has almost never been followed by proper maintenance, which has resulted in very l o w quality o f services relating t o infrastructure. Moreover, infrastructure spending in H a i t i has been heavily focused o n service provision through utilities, which provide services in parts o f the urban areas. Infrastructure needs o f the poor living mainly in rural areas and in slums have n o t received the necessary attention. 4.36. U n l i k e in the education and health sectors, the private sector i s hardly present in infrastructure sector. I n this perspective, the burden o f infrastructure expenditure o n the government budget i s quite large. 4.37. Increasing the population's access to improved quality of basic infrastructure services will require that sufficient public resources are allocated to the Ministry of Public Works, Transport and Communication (MTPTC), the key institution within the infrastructure sector. However, it i s also critical to improve the quality o f public spending, to ensure effective use of public resources in infrastructure sectors. In particular, weaknesses need t o be addressed in budget preparation and execution, accountability and transparency in budget process and management, public procurement issues as w e l l as human resources management constramts. Transparency and accountability o f resource use and general quality of administration within utilities are also key issues - as w e l l as overall legal and regalatory frameworks and the degree o f enforcement o f these resources. They are pre-requisites for public spending o n infrastructure to translate into increases in access and quality o f infrastructure services and t o make this sector a cornerstone o f the growth and poverty reduction strategy o f Haiti. 4.38. T h e infrastructure sector i s also dominated by the critical role that public entities play in the delivery o f infrastructure services, including: (i) the autonomous public companies: EDH (Electricitk D 'Haiti) in the electricity sector; TELECO in the sector o f telecommunications and CAMEP (Centrale Autonome Mdtropolitaine d 'Eau Potable) and SNEP (Service National pour I 'Eau Potable) in the water sector; iii)the autonomous organizations in charge o f regulation and maintenance, such as the CONATEL (Conseil National des Tdlkcommunications) for telecommunication sector, the LNBTP, for the construction sector and the FER for the road sector; and iv) the Conseil de Modernisation des Entreprises Publiques (CMEP), a technical entity that oversees the reform o f public enterprises. While the issue o f improving their capacity to deliver high-quality infrastructure services t o the poor Haitian might not necessarily require increasing the level o f public resources allocated to them, it i s crucial that their financial situation be sound. 59 Although strategies exist in some sectors, such as roads and electricity, they do n o t include spending targets. 75 Haiti: Public Expenditure Management and Financial Accountability Review Box 4. K e y Characteristics o f the Infrastructure Sector The road network: The Haitian transport network i s not only insufficiently developed but it deteriorated significantly in the decade before 2004. That decade has seen the transport network in constant degradation. In 2004, according to the MTPTC6’, the road network consisted o f 3400 km o f roads, o f which only 18 percent were paved. T h i s network consists o f 20.3 percent o f primary roads (routes nationales), 44.4 percent o f secondary roads (routes dbpartementales) and 35.4 percent o f tertiary roads (routes communales) that link the rural communities to the secondary road network. Furthermore, only 20 percent of these roads are considered to be in an acceptable condition. T o address the maintenance issue, an autonomous entity - the Fonds d’Entretien Routier (FER) - has been progressively put i nplace from 2003 to ensure maintenance in the sector. The FER benefits from earmarked hnds (coming from taxes o n gasoline, car licenses, etc.) transferred directly by the MEF and act as a donor o n issues regarding road maintenance (MTPTC plans and contracts while the FER finances). But, this executing system has experienced some delays mainly due to a lack o f planning and executing capacity o f the MTPTC. Electricity: In Haiti, the production, transport and distribution o f electricity i s the responsibility o f EDH (Electricitb D’Haiti), a public commercial company that i s under the supervision o f the MTPTC. Currently, the company provides electricity to 184,000 customers - who represent about 10 percent o f the population. And there has long been almost no maintenance. Thus, from an installed productive capacity o f 220 MWh, the company now produces around 60 MWh. EDH also faces a high number o f illegal connections. The losses due to these illegal connections are estimated to be about 30 to 40 percent o f the output. Water and sanitation: The Haitian water and sanitation sector i s dominated by community-based water comnaittees and two state-owned companies: the CAMEP (Centrale Autonome Mbtropolitaine d ‘Eau Potable), which i s in charge o f water supply in Port-au-Prince and the SNEP (Service National p o w I’Eau Potable), which i s in charge o f water supply in secondary towns and rural areas. According to Haiti’s C E M 2006, i n 2002, 88 percent o f population had access to an improved water access 1 11areas that are urban and 55 percent areas that are rural. Service quality i s poor with intermittent supply and dubious water quality, forcing those who can afford it to r e l y o n expensive bottled water and water from tanker trucks. N o real sewer network exists i n Port-au-Prince or in secondary towns. Existing sanitation facilities are individual and consist o f latrines, septic systems and clandestine connections to the storm water drainage system. According to Haiti’s C E M 2006, in 2002 52 percent o f population had access to sanitation facilities in urban areas and only 23 percent in rural areas in Haiti. The sector i s currently under a reform process through a Water and Sanitation Framework L a w that i s currently being discussed in Parliament. According to the Law, regional water and sanitation companies would b e created to replace CAMEP and SNEP. As part o f the reform process, a General Directorate for Drinking Water and Sanitation (DGEPA) i s expected to be created under M T P T C with the task o f developing policies and to regulate the sector. Telecommunication: There are four major operators in the telecommunications sector: TELECO - the national telecommunication company - DIGICEL, COMCEL and HAITEL, the last three being cellular operators. TELECO retains a monopoly over fixed lines but the cellular market has been open to competition since 1997. In Haiti, there are about two million mobile phones for about 1.8 million subscribers - o f which about 1.2 million are DIGICEL subscribers. The telecommunications sector i s regulated by the CONATEL (Conseil National des Tdbcommunications) which i s an autonomous entity and depends on MTPTC. The national company, TELECO, has been experiencing a very difficult situation since opening to competition. I t i s scarcely represented in the cellular market, whereas this market experiences a rapid growth. 60 But, according to MTPTC, these data are only estimations as the last inventory was made more than 10 years ago. A new one i s being performed. 76 Haiti: Public Expenditure Management and Financial Accountability Review 4.2.2 T r e n d s and Structure o f I n f r a s t r u c t u r e Expenditures 4.2.2.1 Trend in Total Allocations 4.39. Allocation to the infrastructure sector more than tripled in real terms o n average between FY2002-04 and FY2005-07, mainly driven by increased external resources allocated to the investment budget since the re-engagement o f donors in 2004. 4.40. T h e investment budget tripled Over the Same Period. As a Table 35. Allocations to the Ministry of Public Works, result, the share o f MTPTC budget - in Transports and Communication (MTPTC) the nation’s total budget more than doubled from an average o f 10.1 percent over the period FY2002-04 to an average o f 22.0 percent during FY2005-07 (see Table 35). Total Nation Budget 21,264.5 32,491.3 4.41* Accounting for about Total Recurrent Budget (Excl. 14,073.6 11,792.5 percent of FY2006/07 GDP, Interest Pavments) - infrastructure budget compares I MTPTC iota1 [ n Total Nation 10.1 22.0 I favorably with international MTPTC Recurrent in % of total 2.0 2.4 standards - the infrastructure expenditure peaked at 6.5 percent o f GDp in Dominican Republic Over the Sources: L e Moniteur. Journal Offciel de l a RC-oublidhe I p e l k d 1970 to 2003. However, some d’Haytj and Staff Calculations. concerns remain for three main reasons: First, t h s represents only allocation, and H a i t i experiences some difficulties $0 hlly execute resources allocated to that sector; Second, the increase in allocations i s mainly based 03 donors’ contributions, which are known to be volatile. Thus, securing financing for implementing a multi-year integrated PIP i s not guaranteed; Third, infrastructure spending in H a i t i goes f r o m such a low base that it would be necessary to maintain this level o f expenditure for many years t o achieve significant results. But past experience o f volatility o f resources allocated t o the sector, in particular during the period FY2001-04, raises the issue o f the ability to maintain consistently high levels o f resources. f Allocation 4.2.2.2 Structure o 4.42. T h e formation of the infrastructure budget i s dominated b y investment spending, reflecting the nature o f the infrastructure sector. T h e infrastructure sector i s mainly composed of economic activities, which involve investment spending. M o r e than 90 percent o f the M T P T C budget consists o f investment expenditure while recurrent budget accounts o n average for about six percent o f resources allocated t o the MTPTC over the FY2004-06 period. 4.43. T h e recurrent budget increased slightly by 5 percent in real terms o n average between FY2002-04 and FY2005-07. As a result, the average share o f the recurrent budget allocated to the MTF’TC in percent o f total Government recurrent expenditure (excluding interest payments) remains relatively flat at less than 3 percent over the period under review. T h i s figure raises three major concerns. First, i t suggests that investment spending i s not supported by sufficient associated recurrent allocations (most notably for spending o n maintenance) in the budget. One explanation o f this situation is related to the disconnect between the investment budget and the recurrent budget, which follow a parallel budget preparation process - that is, a budget not fully integrated. Another explanation i s related to the lack of resources that Haiti 77 Haiti: Public Expenditure Management and Financial Accountability Review faces, which constrains the Government’s capacity to adequately finance the recurrent costs associated with the investment projects and programs. Second, efficiency o f investment programs i s closely linked to maintenance spending to ensure durability o f infrastructure investment. Unfortunately, Haiti’s recurrent budget does not account for these aspects o f infrastructure investment. The budget does not clearly indicate how much recurrent spending i s associated with the execution of infrastructure investment program and projects. Congestion costs associated with limited infrastructure network should also be accounted for as they affect the quality o f infrastructure. Third, a legitimate question (and more fundamental problem) that arises i s the opportunity to expand the investment in infrastructure. T h i s PEMFAR takes the view that gwen the huge infrastructure gaps, congestion costs, and the need to increase the accumulation o f physical and human capital (a pre-requisite for higher growth and poverty reduction), there i s a need to expand the infrastructure base with new investments while ensuring that adequate levels o f financing for recurrent costs are provided. Chapter 7 o f this report presents a macro-model in which the composition o f public expenditure, externalities associated with public capital, issues o f maintenance spending, and durability and quality o f public capital are explicitly accounted for. Policy simulations are carried out in Chapter 8, with one assessing the impact o f increased spending on infrastructure and higher efficiency o f spending on growth and poverty reduction. 4.44. Given Haiti’s limited budget, the private sector should play a key role in the infrastructure sector to attract resources to the sector and alleviate the burden on the public sector’s scarce resources. Unfortunately, private sector i s hardly involved in the infrastructure sector, partly because o f political instability over the past years and security issues. H o w t o increase private participation in infrastructure should be thought through. There are windows o f opportunities for private activities in the infrastructure sector that would need t o be identified. Carrying out an investment climate assessment (ICA) should shed some light o n the avenues t o explore land the appropriate forms that the private sector involvement could take t o have maximum impact on the infrastructure sector.61 4.45. Because o f limited public resources and little involvement of the private sector, it i s critical that resources b e allocated to the best uses and b e used as efficiently as possible in the infrastructure sector. The management o f resources allocated t o MTPTC, in particular the recurrent budget, could be improved and therefore help save some resources t o be reallocated to other use. The staffing structure mainly composed o f too many unqualified staffs (poorly remunerated and motivated) could be restructured to allow for a smaller number o f better paid qualified personnel. Indeed, as pointed out in Haiti Country Economic Memorandum (CEM), the staffing structure i s inefficient with too many unqualified people in ineffective roles and shortage o f qualified, technical staff, particularly in the regional department”62. Another avenue o f efficient use o f public resources i s public transfers to public utilities involved in the infrastructure sector. Quantifying the possible savings associated with more efficient use o f public resources in the infrastructure sector could be performed in the context o f a complementary study t o this PEMFAR. 61 There are indications that an appropriate form for private sector involvement i s twinning or arrangements for comprehensive technical assistance w i t h international utilities, as in the Jacmel Power System Rehabilitation Project. This program which i s financed by the CIDA and the Government o f QuCbec with close involvement o f Hydro-QuCbec, provides a comprehensive package o f assistance that has allowed the EDH office in Jacmel to produce and distribute electricity in the area. 62 “Haiti: Options and Opportunities for Inclusive Growth, CEM, World Bank (2006) 78 Haiti: Public Expenditure Manaqement and Financial Accountability Review f Intra-sectoral allocation o 4.2.2.3 Analysis o f spending 4.46. The analysis o f intra-sectoral allocation o f expenditure shows that over the period FY2005-07, on average, 60 percent o f the total recurrent budget o f MTPTC was allocated to finance the Ministry’s general administration services. This left few resources to other services and entities under MTPTC supervision, and might have affected their capacity to perform activities. For instance, resources allocated to CONATEL in the FY2006/07 budget (about five million gourdes) were not sufficient to allow i t to perform i t s regulatory role o f the telecommunications sector. At the same time, allocations to the FER have declined over the past three fiscal years: from 21 million gourdes in the FY2004/05 budget to 5.0 million in FY2006107. I t i s worth mentioning that this subsidy to the FER should disappear with the complete implementation o f earmarked taxes which are supposed to be transferred directly by the MEF to finance the FER. f Spending 4.2.2.4 Economic Composition o 4.47. On average, allocation to investment account for more than 90 percent of total allocation over the FY2004-06 period. This reflects the nature o f activities o f the MTPTC, an “investment ministry”. Wages and salaries accounted for less than 4 percent o f total expenditures on average over FY2005-07. T h i s l o w level o f the wage bill i s partly explained by the l o w level o f salaries paid in the MTPTC. As a result, the sector cannot attract qualified workers and many employees are poorly motivated. 4.48. Goods and services and other expenditures accounted for about 3 percent of total expenditures on average. over the FY2004-06 period, down f r o m an average o f 7 percent in FY2001-83. T h i s reflects the reduction in allocations to goods and services, which fell in real terms by more than 50 percent between FY2002-04 and FY2005-07. T h i s raises the issue of the capacity o f the M T P T C to perform i t s routine operations. 4.2.2.5 Analysis of Budget Execution 4.49. T h e main feature o f budget execution in the infrastructure sector i s the extreme volatility of the execution of resources allocated to this sector. Yearly execution o f expenditures varied significantly over FY2002-06. While in FY2001/02, only 34 percent o f total allocations to M T P T C were spent, the execution rate increased to 86 percent in FY2002/03, but f e l l to 45 percent in FY2004/05. This mainly reflects the volatility o f the execution o f the investment budget. Recurrent spending was executed at about 90 percent over the period FY2004-06. However, the execution o f the investment budget dropped from 78 percent in FY2002103 to less than 40 percent in FY2004105, and further to 12 percent in FY2005/06. 4.50. Because of problems of accurately recording execution data, caution should be taken when interpreting the execution rates. Three factors explain the low execution o f the investment budget: i) the l o w absorption capacity o f the investment budget in the M T P T C due mainly to slow execution o f programs and projects, human resource constraints, and limited technical capacity, ii)the poor capacity planning due partly to a lack o f qualified staff in MTPTC, and ( iii) poor execution capacity o f the private sector to execute in specific areas, such as the building and maintenance o f roads. The absorptive capacity issue has been highlighted with huge flows o f resources following the donors’ reengagement. The data over the past few years showed that the MTPTC absorbed a maximum o f about 1.5 billion gourdes o f the investment budget allocated. However, this does not mean that resources have to be cut and tailor to the sector’s current absorptive capacity. It means that resources have to be used to expand the sector’s 79 Haiti: Public Expenditure Manaqement and Financial Accountability Review capacity through, for instance, the provision o f technical assistance in budget management, increasing staffing and improving maintenance spending. 4.51. Table 36 presents the shares and execution rates of selected items for FY2002-06. Although their execution rate has increased from 74.5 percent o f total recurrent budget in 2002- 2004 to 79 percent in FY2004-06, the share o f roads and network maintenance expenditure in the total recurrent budget has been cut almost by half, f r o m 5.6 percent in 2002-04 t o 2.9 percent in 2004-06. T h i s i s particularly worrisome in light o f the country’s poor quality o f infrastructures and the maintenance needs. I t sheds light o n the financing issue o f maintenance spending that H a i t i faces. A l o n g the same lines, the share o f spending on construction supply has also decreased f r o m 2.3 percent o f the total recurrent budget o n average over FY2002-04 t o 1.2 percent in FY2005-07, despite the increase in the execution rate f r o m 64.5 percent to 85.5 percent over the same period. Starting f r o m FY2005/06 a n e w category o f expenditure has appeared: the subsidies to autonomous entities, such as FER or LNBTP. This expenditure type accounts for about one quarter o f the total recurrent budget and has recorded a relatively high execution rate o f about 84 percent. Table 36. Infrastructure Sector. Share of selected items in Recurrent Budget and their execution r a t e (In %), FY2002/06 Kard and Construction Supplies and Subsidies to netMork supplies accessories aulonontous maintenance for technical organism equip ntent - Allocated 8.5o/u 3.8?’0 0.3% 0% P; 8 Executed 6.9% 3 0% 0.2% 0% N’ 8 Execution rate 62.5% 59 5% 58.2% I Allocated 3.8% 1.6% 0.022% 0% r;, 8 Executed 4.3% 1.5% 0.01 6% 0% N E Executionrate 86.5% 69.5% 58.3% I Allocated 3.2% 1.1% 0.1% 0% 4 8 Executed 3.6% 1.3% 0.1% 0% N Executionrate 93.6% 99.0% 100% I Allocated 2.7% 1.1% 0.012% 24.5% ui 8 Executed 2.2% 1.O% 0.010% 23.0% N 2; Executionrate 64.4% 71.9% 66.7% 73.8% ources: L e Moniteur, Journal Officiel de la Republique d’HaYti and Staff Calculations. 80 Haiti: Public Expenditure Management and Financial Accountability Review 4.52. The low execution rate of the capital budget o f the MTPTC might raise the issue of the relevance of Table 37. Infrastructure Sector. Budget Execution allocating large amounts o f resources to the sector (as was the case over the past two fiscal years). However, at the same time, Haiti’s infrastructure needs are huge and require that massive Total recurrent allocated 277.4 302.9 resources be allocated to investment programs. The issue i s therefore h o w to improve the execution capacity o f the MTPTC to ensure that the resources provided are well executed to meet the Sources: Le Moniteur, Journal Officiel de la Rkpublique infrastructure needs. The challenge i s d,Hafti and Staff Calculations. that this issue involves many other aspects o f the operation o f the MTPTC, including improving i t s management, providing technical assistance to enhance budget management, and increasing human resources and technical capacity. A policy response to expand the sector’s capacity could consider for instance the provision o f technical assistance in budget management, increasing staffing and improving maintenance spending. 4.53. As compared to the investment budget, the pace of Table 38. Infrastructure Sector. Recurrent Budget I execution o f the recurrent Execution (In Million of real gourdes,), budget has been remarkable. FY2001/06 The recurrent budget allocated i s generally fully spent. Personnel spending and expenditure o n Total recurrent allocated 2 77.4 302.9 Of which goods and services reached, more Wages and salaries 172.7 163.7 often, an execution rate o f 100 Goods and services 94.7 43.7 percent during the period under Others 0.4 2.4 review. The MTPTC’s spending Total recurrent executed 300.3 303.4 o n these budget items has Of which sometimes exceeded the Wages and salaries 175.6 157.3 allocations. This mainly reflects Goods and services 124.7 146.1 poor budget planning. Others NA NA Execution rates (in %) 110.2 102.9 4.54. The analvsis o f the Wages and salaries 102 96.3 sources o f execution o f the I Goods and services 136.8 113.4 investment budget reveals the Sources: Le Moniteur, Journal Officiel de la Rkpublique d’Hafti and Staff Calculations. remarkable efforts that the authorities have made since FY2004/05 to eliminate the recourse to the use o f cornptes courants to execute the MTPTC budget. While over the period FY2002-04, more than 95 percent o f investment spending o f the M T P T C was executed through the comptes courunts, in FY2004105, the investment budget was executed through the normal budget procedure without recourse to comptes courunts. Figures pertaining to the recurrent budget also show the elimination o f the comptes courants: less than 1 percent o f recurrent spending was executed through the comptes courunts in FY2004105 compared to 37 percent o n average over the FY200 1-03 period. 81 Haiti: Public Expenditure Management and Financial Accountability Review 4.2.3 Public Investment Programs in the Infrastructure Sector 4.55. The PIP in infrastructure sector displays three main characteristics. First, the infrastructure sector stands out as the major sector o f the Government’s PIP. I t accounts for o n average more than 40 percent o f total resources (including foreign aid) allocated t o the PIP over FY2004-07. Second, the infrastructure PIP i s mainly financed by donors. M o r e than 90 percent o f the infrastructure FY2004-06 PIP was financed by foreign aid. However, this i s not specific to the infrastructure sector. M o s t o f the investment programs in other sectors, including agriculture, education and health are also funded by external aid. Third, aid-funded allocations to the infrastructure PIP literally exploded over the past two fiscal years. Total allocations funded by donors almost quintupled in t w o years, between FY2004/05 and FY2005-07 (see Table 39), reflecting both the authorities’ successful policy to attract foreign aid and donors’ response t o the Government’s quest to finance much needed basic infrastructures. Table 39. Resources for PIP in the Infrastructure Sector. Comparison to other Priority Sectors, Infrastructures 1 934,8 2 426,3 26,9 11 868,4 12 801,4 76,3 le 40 provides details on donors’ interventions in the infrastructure sector. The he European Union stand out as the sector’s two main donors. Together, they account for nearly 80 percent o f the sector’s investment pledges for the 2004-06 period; with the EU representing almost 60 percent o f the sector’s investment expenditures over the period June 2004-December 2005. 82 Haiti: Public Expenditure Manaqement and Financial Accountability Review Table 40. Donors’ financing under I C F (in million $US unless otherwise indicated), 2004-06 pledges and June 2004-December 2005 realizations Road Transport 112.6 6.6 I 0,9 I I I 103.0 223.1 Rehabilitation 112.6 6.6 I I I I I 47.2 166.4 N e w Road Projects I I I I I I I 1.4 1.4 Poverty Reduction I I I 0.9 I I I 54.5 55.4 Electricity I I 6.0 I 23.9 4.0 24.0 I 57.9 Support t o electricity sector I I 6.0 I 19.9 I 24.0 I 49.9 Rehabilitation I I I I 4.0 3.0 I I 7.0 Electrification I I I I I 0.6 I I 0.6 Other I I I I I 0.4 I I 0.4 Water and Sanitation 68.6 1.2 t I I 11.0 I 12.2 93.1 Rehabilitation I I I I I 4.2 I 7.0 11.2 N e w Projects 15.0 0.5 I I I 0.7 I 4.6 20.9 Other 53.6 0.7 1 I I 6.1 I 0.7 61.0 Total 181.2 7.8 6.0 0.9 23.9 15.0 24.0 115.2 374.1 Percent o f total Contributions 40,l 2.3 2.1 0.3 8.5 1.4 8.5 36.7 100.0 Spent Budget 22.09 1.01 0.51 / 0.76 1.52 22.95 67.11 115.95 Roads and Transport 17.21 0.26 / / / , 0.02 045 1422 32.26 Electricity 0.76 0.45 22 0 0.23 23.44 Water and Sanitaion 4.88 0.75 0.51 / / 1.05 0.50 52.56 60.25 - Durce: Bilan des Financements et des RCalisations PBriode :juin 2004-dCcembre 2005 ; mars 2006. 4.57. I n terms of components, support to road rehabilitation accounted for the largest share of donors’ interventions in infrastructure. I t represented more than 50 percent o f total donors’ financing. 4.2.4 Weaknesses, Issues and Challenges in Public Expenditure Management in Infrastructure Sector 4.58. Budget preparation i s dominated b y the investment budget, leaving l i t t l e time to the recurrent budget. As in other line ministries, there i s n o formal planning mechanism or a specific timetable for preparing the recurrent budget. Allocations are therefore mainly determined on the basis o f prior year spending. 4.59. The public expenditure management o f t h e M T P T C does not reflect t h e changes it has to initiate to move from force account activities toward contracted works. The public expenditure management has to evolve in parallel with the change o f MTPTC toward a regulatory body. 4.60. The Haitian public works market i s s t i l l too weak and suffers from great inefficiencies and o f a lack of institutional capacity. In addition, the MTPTC fails to attract 83 Haiti: Public Expenditure Management and Financial Accountability Review foreign companies (especially those based in the Caribbean), in part because o f consequent insurance costs due to the security situation. 4.61. An infrastructure PIP largely financed by foreign aid raises three major issues: (i) absorptive capacity; ( ii)sustainability; and (iii) credibility. First, with its weak implementation capacity, the issue i s raised as to h o w the MTPTC could cope effectively with huge flows o f foreign aid. The experience to date has demonstrated that the limited execution capacity o f the MTPTC does not allow i t to efficiently allocate resource available t o infrastructure projects and programs. This however does not mean that aid-funded infrastructure should be reduced. It calls for both the Government and the donor community to enhance the capacity o f the MTPTC and entities involved in the infrastructure sector (especially the FER). Second, related to the issue o f limited absorption capacity i s the issue o f sustainability o f the infrastructure PIP. Large investments in infrastructure also imply recurrent spending that the Government’s limited budget might not be able to afford. Sustainability problems o f the PIP might thus arise over the medium-term Third, beyond the issues o f absorptive capacity and sustainability, the design o f the infrastructure PIP raises credibility issues. The huge gap between the amount o f allocated resources included in the PIP and the resources executed reflects technical problems t o effectively cost the programs and projects at the sectoral level. The PIP i s not done o n a rolling basis, therefore it does not include backward adjustment calculations. The PIP i s not backed by an MTEF (Medium-Term Expenditure Framework) and hence lacks a forward-looking approach. In fact, the document i s more generally a yearly approximate estimate o f programs and projects in execution or to be executed. It does n o t rest o n sectoral strategies and has n o growth or poverty reduction perspective. ’ 4.62. The infrastructure sector i s certainly ector where the adoption o f a programmatic budgeting approach, within a medium-te budget framework (MTEF and budget- pr6grain tools) i s needed because o f the very ’dynamic nature of investment in infrastructure and its positive dynamic externalities on education, and health outcomes (see Chapter 7). Linkmg public spending with outcomes within and outside the sector would indeed require having programmatic budget tools t o ensure that budget allocation and execution are in line with the multi-year sectoral objectives. However, the pre-requisites o f such an approach are not yet in place. H o w to create the conditions for moving to a medium-term PIP, given the current status o f budget preparation in the infrastructure sector, i s a challenge. I t would require a gradual approach that identifies the different steps (see B o x 5) to follow and technical assistance i s also required. 4.63. Largely dependent on foreign aid inflows, budget execution in the infrastructure sector has been volatile. Increasing the impact o f infrastructure spending o n outcomes requires more predictability in the flow o f resources. But the issue i s that aid flows t o the sector depends (or will rely more and more) o n the country performance in the context o f shift ofthe donor community to a pe$omance based budget allocation. Therefore, there i s n o guarantee that the execution will be stable and have significant impact on outcomes. The volatility o f budget execution o f MTPTC i s also reinforced by the budgetary management rule o f Haiti, which implies one-twelfth execution o f the state’s budget. T h i s implies that the MTPTC often has to delay the implementation o f big infrastructure projects, whose execution requires gathering enough resources over many months. 84 Haiti: Public Expenditure Management and Financial Accountability Review Box 5. Pre-requisite f o r a M u l t i - Y e a r Sectoral MTEF STAGE CHARACTERISTICS I. Development o f MacroeconomiciFiscal Macroeconomic model that projects revenues and Framework expenditure in the medium-term (multi-year) 11. Development o f Sectoral Programs Agreement on sector objectives, outputs, and activities Review and development of programs and sub- programs Program cost estimation 111. Development of Sectoral Expenditure Analysis o f inter- and intra-sectoral trade-offs Frameworks Consensus-building on strategic resource allocation IV. Definition o f Sector Resource Allocation Setting medium term sector budget ceilings (cabinet approval) V. Preparation o f Sectoral Budgets Medium term sectoral programs based o n budget ceilings VI. Final Political Approval Presentation o f budget estimates to cabinet and parliament for approval Source: L e Houerou P. and R. Taliercio (2002): Medium T e r m Expenditure F r a m e w o r k F r o m Concept to Practice, W o r l d Bank Africa Region Working Paper Series. 4.64. Another issue i s related to the lack of clear spending targets, which makes id difficult to assess the effectiveness of budget execution in terms o f achieving the objectives set. The l i s t o f poverty-reducing budget items does not integrate linkages between infrastmcture spending and outcomes o f other sectors, most notably education and health. Moreover, while the practice o f comptes courants has declined, issues o f transparency, accountability, and cumbersome budget procedures s t i l l impact the effective use o f public resources in infrastructure, and ultimately their impact o n human development outcomes. 4.65. A specific issue concerns management problems o f autonomous agencies such as CONATEL (telecommunication sector), and FER (road transport sector), which prevent them from delivering high quality infrastructure regulation. Part o f the problem i s related to irregular transfers o f resources collected by the MEF to these public autonomous entities. As a result, they often experience treasury problems at the expense o f infrastructure services. 85 Haiti: Public Expenditure Manaqement and Financial Accountability Review Box 6. T h e R o a d Maintenance Fund (Fonds d’Entretien Routier, FER) The setting o f a road maintenance fund, largely supported by donors, has been gradual. The FER was created in 2003 (Journal Officiel de l a Republique d’Haiti NO54 July 24th). The Board was appointed in December 2004 and i t held i t s first meeting. The f i s t contract was signed in January 2006. By law, the FER i s an autonomous entity that i s in charge o f financing road maintenance for the eligible network. I t s resources cannot b e used to finance new roads o r rehabilitation that can be considered as investments. The FER i s o n l y in charge o f financing and o n its o w n it cannot process road maintenance o r act as a w o r k master. The FER has n o w achieved a significant operative capacity, particularly in terms o f available resources. But, i t s actual road maintenance activity i s still insignificant. However, the FER’Sfiscal commitments include: (i) a contract for periodic maintenance; ( ii)a dozen for regular maintenance; and ( iii)an emergency maintenance convention. But, a l l these contracts are waiting either for procurement authority approval, o r for the oversight implementation. Therefore, n o request for payment has yet been sent to the FER. I t is thus critical that FER and the MTPTC process the implementation o f these activities. Currently, the FER has a budget o f 135 m i l l i o n gourdes, broken down as 85 percent for maintenance activity financing, 10 percent for coverage o f its operating costs and 5 percent for emergency maintenance. T h i s budget comes f r o m fuel fee revenues and from donor support. The FER i s currently waiting for the transfer o f the drivers’ license and f i r s t registration fees. Therefore, i t i s critical that the Ministry o f Finance implement fiscal resource transfers to the FER o n a consistent and predictable basis. Sources L e Moniteur, Journal qfficiel de la Rtpubhque d’Haiti, N”54, Jeudi 24 Juillet 2003 and Aide-memoire, Transport and Terntonal Development Project, February 2007 4.66. Lastly, like in most other line ministries, human resource management i s also an issue and affects expenditure management. The limited number o f qualified staff well acquainted with budget procedures limits t iveness o f budget preparation and execution in the MTPTC. N o more than four staff s are devoted to preparing the Government infrastructure PIP. This causes delays in preparing the PIP and following-up o n i t s execution. Only recently has the MTPTC produced its first implementation report. The quality o f that report could be substantially improved. 4.2.5 Policy Recommendations 4.67. Based on the findings o f the PER, the following policy recommendations are made: Prepare a full-fledged costed and integrated infrastructure strategy with sub- sectoral actions plans, which are linked to spending targets. T h i s involves having an integrated approach in the sector. The existing sub-sectoral strategies could serve as the basis for elaborating such a strategy. A rigorous process o f spending planning should be implemented. This requires that the MTPTC improve its budget preparation process by elaborating a clear timetable and allocating more time to budget planning (both resources expected and expenditure planned to be executed). Then, a traclung mechanism should be put in place to compare results to initial objectives and readjust the strategy; Prepare an integrated multi-year PIP, which reflects the translation o f the integrated infrastructure strategy into resource requirements. A first step i s to proceed with the elaboration o f a PIP using the classic back-of-envelope calculation methodology. In this case, what i s needed i s to improve the current PIP and make i t reflect the integrated infrastructure strategy. This means that spending targets defined in the strategy have to be reflected in the multi-year PIP. The PIP should also comprise o f annexes that track the yearly execution o f programs and projects covered by the PIP. In 86 Haiti: Public Expenditure Management and Financial Accountability Review the medium-term (3-4 years), the elaboration o f the PIP should be based o n a more advanced methodology, which enables accounting for the linkages between infrastructures and education, health and security. Aid requirements will be calculated and might prove to be lower than the back-of-envelope calculation. This could be done in the context o f the revised PRSP in 2010/11.; e A transparent execution of budget implies the elimination of comptes courarzts. The M T P T C has already made remarkable efforts and has dramatically reduced the recourse t o these accounts. The next step i s for the MTPTC to phase out by FY2009/10 the use o f comptes courants to ensure transparency, credibility and accountability in budget execution; 0 Connect new investments with maintenance spending and support to community- based organizations managing rural infrastructure. The MTPTC should coordinate with FER and other entities responsible o f maintenance, to ensure that its planned investments (e.g. rehabilitation o f existing roads) will be followed by proper maintenance and that sufficient funds are allocated t o these entities t o perform this maintenance role. Also, recurrent budget resources should be allocated t o provide public good functions related to basic rural infrastructure, such as rural water committees that operate and maintain basic water infrastructure, and the promotion o f on-site sanitation. Reaching the poor requires a different approach to public expenditures that acknowledges the fact that much o f the rural infrastructure, especially in water and sanitation, i s operated and maintained by communities. Support to these communities i s a public good function with high pay-offs. Financing and organizing this support, directly or indirectly, i s a k e y state function for which sufficient resources should be made available; e Strengthen the functioning o f the Road Maintenance Fund (FER) by: (i) enhancing coordination o f the programming o f road maintenance activities with MTPTC; ( ii ) ensuring regular and adequate budgetary transfers to the FER account o f proceeds from earmarked taxes for road maintenance; ( iii)developing expertise in FER to promote an efficient model for routine and emergency road maintenance, using either small firms or community-based micro-enterprises; and (iv) improving i t s institutional capacity. T o ensure the FER functioning, i t i s crucial t o secure its financing. Thus, as required under EGRO 1 1, the government should include in the FY2008 budget an annex detailing the planned revenues and expenditures for the FER. The current lack o f a budget line for the FER makes it difficult to track whether sufficient resources have been allocated t o i t for road maintenance; Overhaul state utilities. Funds must be channeled to the rehabilitation o f existing dilapidated assets and improving service to existing customers before coverage can be expanded. Maintenance i s crucial t o ensure the sustainability o f investments made and must be provided for by ensuring adequate funding and training o f staff. Transparency and accountability require better information systems, a more active role for boards of directors and open contracting, and regular public disclosure o f information on levels and quality o f services provided to the public and the performance o f public utilities. And realistic strategies are needed for involving the private sector where possible. In the road sub-sector, particular attention should be paid to strengthening the national public works market and t o attract foreign companies. A restructuring and a decrease in the wage bill i s expected to enable these companies to devote resources t o investments. Moreover, the financial relationship between these companies and the public authority i s not very transparent as government departments do not always pay the bill and utilities do not always pay taxes. It i s crucial for the functioning o f these companies that the budget 87 Haiti: Public Expenditure Manaqement and Financial Accountability Review includes expenditure lines f o r electricity, water and telecoms in the FY2008 budgets to track the payment o f bills to EDH, TELECO and CAMEP; and Improve donor involvement and better coordinate their intervention. This involves b o t h better coordination and securing longer term donor commitments. I n the transport sub-sector, policies are coordinated. Inparticular, the EU, the AFD and the Bank are co- financing a major road segment in the N o r t h region (RN3). In the water and sanitation sub-sector, a protocol i s in progress to coordinate donor actions. I t i s crucial that such agreement be successhlly concluded and be extended t o other infrastructure sub-sectors. 88 Haiti: Public Expenditure Management and Financial Accountability Review CHAPTER 5 : SOCIAL SECTOR EXPENDITURE REVIEW 5.1. This chapter presents a review o f public expenditure in social sectors, education and health. I t first analyzes the recent trend o f allocations o f public resource to these sectors over the past five fiscal years, and the extent to which they are consistent with the social development priorities o f the Government o f Haiti. I t then analyzes the structure o f spending in these sectors and highlights the relative weight and trends o f recurrent versus investment expenditure in these sectors. The chapter also discusses the execution o f spending and sheds some light o n the determinants o f the execution rates. The PIP in these social sectors i s presented and its main characteristics are discussed. The weaknesses, issues, and challenges o f public expenditure management in these sectors are also discussed. The chapter concludes by providing some policy recommendations t o improve effectiveness o f public spending in order t o achieve higher education and health outcomes. 5.2. The key findings o f the chapter are the following: First, the allocation o f public spending t o the education sector increased by more than 30 percent in real terms o n average over the period FY2002-07 while it doubled for the health sector over the same period. Second, the increases reflect the l o w levels o f allocations to begin with and are mainly the result o f donors’ aid flows. Third, despite these increases in allocations, the level o f expenditure in the education (about 2.5 percent o f GDP in FY2006/07) and health (less than 3 percent o f FY2006/07 GDP) sectors are still low by international standards and insufficient to meet the EFA goals (education sector) and the sector priorities and objectives (health sector). Fourth, both sectors are dominated by an important private sector. Eighty percent o f a l l primary level students attend n o n public schools, financed by parents, religious associations, NGBs and other sources, and only 20 percent attend virtual!y fi-ee public schools. The share o f private financing o f the education sector i s esthated at more than 3 percent o f GDP. About 60 percent o f health expenditures are provided by the private sector. Fifth, the execution rates are quite low, reflecting the combination o f weaknesses in budget formulation and planning and low absorptive capacity o f these sectors. Sixth, the PIP in these sectors i s dominated by donors’ resource flows. Seventh, weaknesses and issues o f poor public expenditure management result in inefficiency o f public spending, which impedes Haiti f r o m achieving higher education and health outcomes. 5.3. Policy prescriptions for the education sector center on: (i) increasing allocation o f resources to the education sector, with a focus o n primary education; ( ii ) executing the Medium- Term Expenditure Framework (MTEF); ( iii)improving budget data and information; (iv) Improving public resource management; (v) preparing a l i s t o f poverty reducing expenditures to be protected against a sudden shortfall in the resources and follow-up i t s execution on a quarterly basis; (vi) ensuring sustained investment flows and targeting specific investment programs; (vi) developing a tracking mechanism o f donors’ aid flows provided and executed outside the education sector’s budget; (vii) creating and reinforcing Public/Non Public Sector Partnerships; and (viii) reinforcing technical capacity in public finance management. F o r the health sector, the report recommends: (i) increasing allocations to the health sector, with a focus o n delivering basic health services; ( ii) improving the budget preparation process by elaborating a clear timetable and defining responsibility o f each actor involved in the preparation o f the budget; ( iii) reinforcing the links between the sector’s strategic objectives (and expected outcomes) and spending targets and monitor progress closely; (iv) setting the basis for a programmatic budgeting approach by reinforcing pre-requisites for an MTEF (medium-term projections o f resources and expenditures, prioritization o f spending, linkmg spending t o expected outcomes); (iv) preparing a l i s t o f poverty reducing expenditure to be protected in the case o f sudden shortfall and follow-up i t s execution 89 Haiti: Public Expenditure Manaqement and Financial Accountability Review o n a quarterly basis; (v) strengthening human resource management and reinforcing capacity in budget processes and spending management by the provision o f technical assistance; and (vii) developing a tracking mechanism o f donors’ resources provided outside the sector budget. 5.1 EDUCATION SECTOR 5.1.1 Background and Sector Objectives 5.4. With only 71 percent o f children aged 6-12 enrolled in school, and only 32 percent of primary education students reaching the fifth grade, education levels in Haiti are quite low and among the lowest in the world. The education sector suffers from weaknesses in virtually a l l aspects o f the sector’s performance, including access, equity, quality and institutional capacity. Current primary net enrollment rate in 2006 i s calculated at 7 1 percent, with a gross enrollment rate o f 127 percent and a primary level completion rate o f 66 percent. Approximately 500,000 children aged 6-1 1 do not attend school o f any kmd, and only about h a l f o f all 6-year-olds enroll in first grade.63 Gender bias i s also an issue.64Age o f enrollment rates i s significantly different, with 52 percent o f boys and only 43 percent o f girls starting school at age 6. 65. The education system lacks trained teachers for the system. Currently, public education financing benefits mainly middle and lower middle class households living in urban areas, rather than poor households located primarily in rural areas. Quality o f instruction and learning i s extremely poor. Governance o f the sector i s a major issue, both internally and externally. Internally, there i s a lack o f procedures and transparency in the use o f public funds, and poor coordination across technical departments and o f externally financed investment projects. Moreover, many trained National Ministry o f Education and Vocational Training (Ministere de 1’Education Nationale et de l a Formation Professionnelle, MENFP) civil servants have left the system for opportunities abroad or as externally-financed consultants. Externally, the MENFP i s overshadowed by the ’. nob-public sector (which accounts for 90 perc o f all schools), largely ignoring government regulations and accreditation standards. 5.5. The MENFP’s basic capacities to conceptualize, plan, execute, monitor and evaluate educational programs and its budget in a coordinated manner are weak. Reliable up-to-date education statistics and budget data, essential for planning purposes, are not readily available to MENFP staff. 5.6. The Government has prepared a 10-year National Education and Training Plan, 1998-2008, and elaborated the National Strategy o f Education for All (Stratbgie Nationale d’Education pour Tous ou SNA/EPT) with the support o f the World Bank, UNESCO, UNICEF, and other donors.66 The strategy calls for: (i)expanding and improving comprehensive early childhood care and education, especially for the most vulnerable and disadvantaged children; (ii) ensuring that by 2015 all children, particularly girls and children in difficult circumstances, have access to, and complete, fi-ee and compulsory primary education o f 63 D H S Survey 2005 s While gender bias i s a critical issue, t h ~ report does n o t provide a comprehensive analysis o f gender issues. These issues will be covered in detail in a subsequent report, consistent with the programmatic approach o f the PEMFAR. 6s Similarly, the W o r l d Bank report (2006) notes that up t o age 10, school-participation rates are higher for females than males, but this trend reverses between ages 14 and 24. In other words, households are more likely to continue investing in school if their child is male. 66 This fully costed ten-year p l a n aims at placing H a i t i o n the path toward achieving EFA by 2015. I t i s hoped that this strategy will b e endorsed by the EFA Fast Track Initiative in 2007, which could mobilize additional financing f r o m the EFA FTI Catalyhc Fund. 90 Haiti: Public Expenditure Management and Financial Accountability Review good quality; (iii) ensuring that learning needs o f all young people and adults are met through equitable access t o appropriate learning and life-shlls programs; (iv) achieving a 50 percent improvement in levels o f adult literacy by 2015, especially for women, and equitabIe access to basic and continuing education for all adults: (v) eliminating gender disparities in primary and secondary education by 2015. and achieving gender equity in education by 2015, with a focus o n ensuring girls’ full and equal access to and achievement in basic education o f good quality; and (vi) improving a l l aspects o f the quality o f education and ensuring excellence o f all so that recognized and measurable learning outcomes are achieved by afl, especially in literacy, numeracy and essential life skills. 5.7. Specifically achieving EFA goals would require that about 21 percent o f Government’s recurrent budget (excluding debt service) i s allocated to the education sector, o f which at least 50 percent go to primary education. T h i s would demonstrate the Government’s own commitment to these goals, necessary t o mobilize the additional resources f r o m external sources that will be required to achieve EFA. Developing the education sector and making primary education more accessible to the poor are k e y strategic objectives o f the new Haitian Government. This i s reflected in the I-PRSP and reiterated in statements by President Preval. 5.8. I n line with declared government priorities, public resources devoted to the education sector have recently increased. Yet, at 2.5 percent o f GDP for FY2006/07, public education spending appears relatively l o w compared t o the LAC average o f 4.3 percent and stands out as a major source o f weaknesses in the education sector. Increasing the education outconies toward meeting the EFA goals i s conditional t o ensuring adequate public financing to the education sector, improving management o f government’s resources allocated t o the sector as well as efficiency o f spending. 91 Haiti: Public Expenditure Manaqement and Financial Accountability Review Box 7. Institutional and Legal Framework Education in Haiti i s governed by the Constitution o f 1987 and the Decree o f 1989, which confer on the Ministry o f Education, Youth and Sports (now the Ministry o f Education and Vocational Training) the right to elaborate, implement, evaluate and update the State’s general policy related to Education and Vocational Training. Articles 32.1 to 32.3 o f the Constitution stipulate that the State and local governments are responsible for Education. Additional other official documents, such the L a w o f 1901 on public instruction, the Decrees o f December 1960 o n the creation o f the National University and o f March 1998 o n the creation o f primary teaching, and the National Plan o f Education and Training (PNEF, 1998), define the rest o f the legal framework for education policy in Haiti. They all focus o n the encouraging one’s development in all i t s dimensions, ( three objectives o f school aiming at: (i) ii) training citizens to be economic, social and cultural development agents, and ( iii) promoting national and cultural identity and exposure to universal, regional, and Caribbean values. The March 30 2007 decree known as “Bernard Reform” organizes the education system into five levels: pre-school, basic (grades 1-9), upper secondary, technical and professional, and higher education. Furthermore, education in Haiti i s predominantly non-public-sector provided. The latest education census (2002103) showed that 92 percent o f schools are privately operated and financed. In terms o f student enrolment, the State provides for just 18 percent o f primary school enrolment, while non-public education services make up 82 percent o f primary enrolment.67 Non-public schools are a heterogeneous group, encompassing international-quality schools attended by the country’s elite and extremely l o w quality schools attended by the country’s poorest. The majority o f non-public schooling i s affiliated with religious associations (Catholic, Protestant, Presbyterian and the like), as seen in Figure 5.2 below, followed by independent secular schools. 5.1.2 Trends and Structure o f Education Expenditures 5.1.2.I Rising trends in education expenditure allocations 5.9. There has been remarkable increases in education expenditures between Fy2002-07. The education sector stands out as one that benefited from the increase in the Government budget over the past years. Total yearly allocations to the education sector increased in real terms by more than 30 percent o n average over the period FY2002-07 (from 2.9 b i l l i o n gourdes o n average in the period FY2002-04 to 3.8 b i l l i o n in FY2005-07), following rising trends in the total Government budget. As a result, the share o f the education sector in the total budget has remained relatively high at about 10-11 percent o f total budget since FY2003/04. 67 MENJS, Direction de la Planification et de la CoopCration Externe, Recensement Scolaire 2002-2003 92 Haiti: Public Expenditure Manaqement and Financial Accountability Review 5.10. Yet, at about 2.5 Percent Figure 9. Public Expenditure in Education (YOo f GDP) of GDP in FY2006/07, the level o f Haiti and Selected Comparable Countries, 2006 allocations to the education sector still compares poorly with the LICs average (3.2 percent o f GDP) and b y sub-Saharan Africa standards (average o f 3.4 percent of GDP). Total allocation t o the sector i s 2 percentage points lower than that o f Ethiopia (4.6 percent o f GDP), and more than 3 percentage points lower than that o f Uganda (5.2 percent o f GDP) and Burundi- a post conflict country- (5.2 percent I o f GDP), and far below Kenya (7.0 0 2 4 6 a percent o f GDP) (see Figure 9). Source: World Bank Database and Government Statistics, 2006. 5.11. While the Government declared development priority focus on the education sector has translated into higher budget allocations to the sector, Haiti still devotes relatively insufficient resources to education. The number o f teachers i s quite low: nationwide just 450 new teachers are trained each year to an estimated Figure 10.Shares of Financing of Education Sector 10,000 new teachers needed to achieve EFA.68Access t o public education i s thus limited and quality o f instruction and learning i s extremely poor. Curricula are outdated and in many cases inappropriate for overage students who make up the bulk o f the primary education population. Teaching .~~ practices are i n a p p r ~ p r i a t e Because of a lack o f public resources, n o programs exist to support the hundreds o f multi- grade schools in the system. Approximately 75 percent o f all teachers lack adequate training; many Source: World Bank Database and Government Statistics, have just a 9 ' grade or 12' grade 2o06. education, with n o teacher training whatsoever. As parents lack resources and the public sector does not offer sufficient support, the readiness o f children to learn i s a major issue. Many students are simply hungry and lack the energy they need t o focus o n learning. T h i s problem i s compounded by the weak infi-astructure base. Many schoolchildren have to travel long distances from home to school, often without eating breakfast.'O See W o r l d Bank Phase One for an Education for All Adaptable Program Grant. PAD, April 26,2007. 69 They consist almost exclusively in "chalk and talk", requiring students to recite words and phrases they frequently cannot understand. See W o r l d B a n k Phase One for an Education for All Adaptable Program Grant. PAD, April 26,2007. 70 School feeding i s a k e y element o f the support to primary education. Unfortunately, school feeding programs in H a i t i are declining in coverage, due t o cutbacks f r o m k e y international donors, f r o m 800,000 93 Haiti: Public Expenditure Management and Financial Accountability Review 5.12. Linked to the low level o f public resources i s a large and growing non-public provision o f education services. Eighty percent o f all primary level students attend n o n public schools, financed by parents, religious associations, N G O s and other sources, and j u s t 20 percent attend virtually free public schools. The share o f private financing o f the education sector i s estimated at more than 3 percent o f GDP while public domestically-financed education accounts for less than 2 percent o f GDP (see Figure 10 above). Non-public sector enrollment share in primary education i s quite large, representing more than 80 percent o f total sector enrollment: Twice as high as that o f Togo (40 percent); four times higher than that o f Ghana (20 percent) and more than eight times higher than those o f B e n i n and I v o r y Coast (about 10 percent) (see Figure 11). The large share o f non-public education provision makes i t difficult to assure quality of education in H a i t i as non-public schools largely ignore government regulations, accreditation standards, and are rarely visited by Figure 11.Non-Public Sector Enrollment Share in P r i m a r y M E N F P school inspectors. M o r e Education (%) than 75 percent o f all n o n public schools function illegally (no permit or license f r o m the MENFP). However, the n o n public sector’s ability to respond t o parents’ demand for education services obliges the government to accept the current situation. Indeed, the n o n public sector offers b o t h lower unit costs and faster supply response capacity, with greater accountability 1 to ‘‘climts” (parents) than the public I O 20 40 60 80 100 SeC h comparable results on I Private enrollment share (%), primary nat ams. Source: World Bank Database and Government Statistics, 2006. 5.13. Policy to influence education outcomes go beyond the increases in public resources and cannot ignore the non-public sector. Publichon public partnerships thus become a central element o f an outcome-based education policy. Since the National Education and Training Plan was first issued in 1996, i t has been recognized that no viable EFA strategy in H a i t i can ignore the urgent need to develop institutionalized mechanisms t o promote a partnership between the State and non public education sector. Such a partnership could promote cost-effective expansion o f access, while increasing the MENFP’s capacity to ensure respect for minimum standards o f safety, quality and accountability. Creation o f the National Education Partnership Office (NEPO) provides public and non-public education stakeholders a much-needed institutional structure for policy dialogue, strategic planning and operational c ~ l l a b o r a t i o n . ~ ’ children in 2002 to about 400,000 in 2006. See World Bank Phase One for an Education for All Adaptable Program Grant. PAD, April 26,2007. 7’ Legislation establishing the NEPO was submitted for Parliamentary approval in October 2006, and approved b y the Lower House in March 2007 (now before the Senate). 94 Haiti: Public Expenditure Manasement and Financial Accountability Review 5.14. The current budget structure is skewed toward Table 41. Allocations to Education Sector (In real recurrent expenditure. About 80 percent o f the education budget consists of recurrent expenditures, while the investment budget Total Education Recurrent 2252.0 2995.4 accounts for on average 20 percent Budget of the resources allocated to the Total Education Investment 635.1 829.9 Budget sector over FY2004-06?2 The Total Recurrent in YOof total 81.0 80.0 recurrent budget also has been Education budget trending upward since FY2001/02. Total Investment in % of total 19.0 20.0 Allocations to the recurrent Education budget expenditures increased in real terms Total Nation Recurrent Budget 14073.6 11792.5 (Excl. Interest Payments) by about 33 percent over the period Percentage of recurrent in total 15.2 25.0 under review. As a result, the average nation recurrent budget (excluding share o f recurrent budget allocated to interest payments on debt) the education sector increased up to Total Nation Budget 21264.6 35191.0 Shares of Education in Nation 13.1 10.8 25 percent o f total recurrent expenditure (excluding I Budget (in YO) interest Sources: L e Moniteur, Journal Offciel de la Rhpublique d’I iiti payment) in FY2004-06 f r o m an and Staff Calculations. average o f 15.2 percent in FY2001- Education sector comprises Ministry of National Education, Youth 03. T h i s reflects the n e w authorities’ and Sports and University of Haiti In FY2006/07, it includes Ministry of National Education and Professional Training, Ministry efforts to allocate m u c h needed of Youth and Sports, and Civil Action, University o f Haiti. We alsc resources to the education sector, with include “interventions publiques” in favor of education sector a view to improving education outcomes and achieving the EFA goals (see Table 41). 5.15. A4 the same time, the investment budget increased significantly by about 30 percent . in real terms, reflecting the large increase o f investment budget in FY2006/07 (see Table 41). . 5.16. The analysis of intra-sector allocation o f expenditures shows that less than 50 percent o f the total recurrent expenditures to the education sector go to primary education. The predominance o f allocations to primary education reflects the authorities’ education priorities, which a i m at expanding and improving access and quality o f basic education in public schools. However, recurrent spending to education remains insufficient g w e n Haiti’s abysmal enrollment rates. Indeed, total recurrent public education spending (excluding University o f Haiti) for FY2006/07 o f 4.6 b i l l i o n gourdes (US$ 109.3 million) serves about 2.1 m i l l i o n students enrolled in primary and secondary education (both public and non-public), w h i c h translates into 2,187 gourdes (US$52) for each student each year, extremely l o w even by sub-Saharan African standards (estimated US$110). ’* The education sector refers to “Ministere de 1’Education Nationale et de la Formation Professionnelle”, “Ministere de Jeunesse des Sports et de 1’Action Civique”, and “UniversitC d’Ha’iti”. In FY2006107, allocations for the education sector also account for “interventions publiques: rentree scolaires”. 95 Haiti: Public Expenditure Management and Financial Accountability Review 5.17. Economic composition of education expenditures. The Table 42. Composition o f Spending, (In real HTG economic composition of expenditures reveals that wages and salaries account for about 53 percent o f total recurrent spending Goods and services 780.5 222.0 on average during the period Others 908.7 FY2005-07, while expenditures on goods and services and other items Total Share o f wages and salaries in total ~ (in %) 3 ,”k: 168.8 2407.4 53.0 (mainly subsidies) made up about Share ofgoods and services and 25.7 47.0 47 percent.73 The Government has others in total (in %) made efforts to control the wage bill Sources: Le Moniteur, Journal Officiel de la Rkpublique d’HaTti in the education sector while and Staff Calculations. allocating more resources to ensure proper functioning o f Government entities in charge o f the education sector, most notably the National Ministry o f Education and Vocational Training. As a result, the share o f wages and salaries in total recurrent spending declined over the period under review. However, average wages are s t i l l relatively higher than in the non-public sector. Average public school teacher salaries (estimated at US$2,765 in 2005, 7 times GDP per capita) are more than four times higher than non-public school teacher salaries (estimated at US$589, or 1.6 times GDP per capita).74 Unit costs for public schools are thus estimated t o be at least two times those for non-public schools at the primary level. This major issue nourishes a growing deficient non- public education sector. 5. I .2.2 Analysis of Budget Execution 5.1 8. Because of lack of data, the analysis ob budget execution in the education sec limited to the MENFP and does not cover the University o f Haiti. In addition, due t o detailed data o n execution, the analysis was kept broad and limited to the recurrent budget. Future work will complement this analysis. 5.19. Table 43 shows that recurrent expenditures were executed on average Table43. MENFP Budget Execution (In real at about 78 percent over the past four fiscal years. The execution rates are driven by the large execution o f wages and salaries Budget Allocated 1210.9 2198.8 (nearly 100 percent) while spending on Budget Executed 968.9 1673.3 goods and services faces issues o f Execution Rates (in%) 80.9 76.1 absorptive capacity o f large aid flows most notably during the past two years, as well as Source: Le Moniteur, Journal Officiel de la Rkpublique d’Ha’iti and Staff Calculations. procurement issues. However, this does not mean that allocations to the sector should be reduced t o account for its absorptive capacity. Indeed, the ambitious goals o f the sector require more resources. Ensuring that these resources are absorbed and well managed requires creating capacity within the Ministry to manage public funds in a transparent and fully accountable manner. T o this end, the modernization o f the Department o f Administrative Affairs (DAA) i s crucial. The DAA i s responsible for the overall budget execution (in collaboration with the Regional Education Departments, DDE), and i s the primary interface with the Ministry o f Finance. 73Cautionshould b e taken in interpreting the figures and trends. Indeed, reallocations o f items t o different spending categories were made during the period under review, w h i c h m a y w e l l have affected the trends. 74 Enseignement de Base e n Hayti, W o r l d Bank, 2003, p. 10. 96 Haiti: Public Expenditure Manaqement and Financial Accountability Review 5.1.3 Public Investment P r o g r a m s in the E d u c a t i o n Sector 5.20. Achieving the goals outlined in the EFA strategy for primary education alone will cost an estimated US$1.8 b i l l i o n for the period 2007-2015. The strategy estimates available national resources t o finance approximately US$932 m i l l i o n during that period, leaving an external financing gap o f about US$865 million, or approximately US$lOO m i l l i o n per year. I t i s important t o note that the estimate for local resources i s based o n overall annual GDP growth rates o f approximately 4 percent, as per TMF calculation^^^. Second, i t assumes a growing proportion o f public budget resources to the education sector, f r o m an estimated 21 percent in 2007 t o 28 percent by 2015. Finally, i t plans for dedicating 60 percent o f the Ministry o f Education’s annual budget to the primary level by 2015, a target that i s 5 percent higher than that suggested by the Education for All Fast-Track Initiative. Although these planned budget allocations selected by the Ministry o f Education are noble, they might prove overly ambitious, especially considering that the allocation level f o r primary education in 2006 was an estimated 38 percent76. T h i s would effectively represent more than quadrupling the national budget for primary education f r o m approximately $US 39.2 m i l l i o n in 2006 to US$ 191 m i l l i o n in less than a decade. 5.21. Increasing resources. Table 44 shows that resources devoted to the sector’s investment programs (including foreign aid), as recorded in the PIP, increased over the past two fiscal years, from 474 million oJS$12.2 million) in FY2004/05 to 700 million (US$16.6 million) in FY2005/06. This represents an increase o f about 28.9 percent in real terms. Altogether, the sector was granted 1,174 m i l l i o n gourdes (US$24.3 million), equivalent to the allocation t o agriculture but higher than that t o health and justice and security. 5.22. Resources are expected to more than double in FY2006/07, up to 1,8883 mWon gourdes (US$45 million), following a large increase projected in foreign aid flows to the sector (1,681.3 million gourdes i.e. US$40 million). However, the current increase iE investment budget raises several issues. First, the increase comes on such a l o w base that i t i s difficult to ascertain the extent to which efforts have been made to significantly increase the investment budget. Second, the injection o f large resources in the education sector might certainly help H a i t i achieve better education outcomes and bring education indicators to the levels o f comparable countries. However, large resources will not translate into higher education outcomes unless they are w e l l managed and targeted t o sub-sectors and areas with decisive impact o n outcomes. I n other words, efficiency o f investment allocation and spending stands out as critical issues, w h i c h requires utmost attention if H a i t i has to improve education indicators. Third, related t o the issue o f efficiency i s the issue o f capacity o f the education sector to absorb large foreign resources. Because o f i t s limited administrative, technical and human capacity, the r i s k o f waste o f resources exists. But this does n o t imply that allocation o f resources should b e tailored t o existing capacity. I t calls for a strengthening o f capacity at various levels o f the public finance management in the education sector, including budget preparation, execution, controls, and reporting as w e l l as procurement standards. Fourth, investment flows imply additional recurrent spending, w h i c h will result in an additional strain o n Haiti’s limited budget in the long-term. The extent to which the Government budget could afford these resource requirements i s a major sustainability issue. 7s IMF. 2006. Haiti: Enhanced Initiative for Heavily Indebted Poor Countries. Washington DC: IMF. September. IMF Country Report No.06/338. 76 Ministere de 1’Education Nationale et de l a Formation Professionnelle, 2007, L a Stratkgie Nationale d’Action pour I ’Education Pour Tous, Port-au-Prince, Ani1 2007. 97 Haiti: Public Expenditure Manaqement and Financial Accountability Review 5.23. Investment programs in the education sector are mainly financed by Table44. Resources for PIP in the Education donors. M o r e than 80 percent o f the Sector. Comparison to other Priority sector FY2005/06 PIP was financed by Sectors, FY2004-05-FY2006-07 (In foreign aid. T h i s i s in the same range o f donors’ share o f the PIP in other priority sectors, most notably health, agriculture, and transport. Donors’ interventions focus mainly on five areas: governance, quality, access, efficiency, and youth and sport. Most donors investing in education in Haiti are focusing on supply-side issues such as the constructionhehabilitation o f public schools (access) and teacher in-service Sources: Ministry of Planning and External Cooperation training (quality).77 As a result, intervention and Staff Calculations. in the areas o f quality and access account for three quarters o f donors’ commitments under the ICF and outside the ICF framework. T h e most important donors are the European Union, the Inter-American Development Bank (IDB), USAID, and CIDA.78 Table 45. Status of Disbursement versus Commitments, as of December 31,2005 for all types of Source: Ministtre de I’Education Nationale, de la Jeunesse et des Sports, Cadre de CoopCration IntCrimaire. Bilan des Financements et des RCalisations, PCriode de Juin 2OOCDCcembre 2005. Mars 2006. 5.24. However, donors’ commitments of resources do not necessary translate into disbursements. Table 46 compares donors’ commitments over the period FY2004-07 with 77 Few donors are intervening o n the demand-side, most notably the World Bank 78 The European Union has channeled i t s support through the Euro 25 million Quality Improvement Project (2000-2007), focused on in-service teacher training, classroom rehabilitation, and a pilot program for coverage youth in several regions. I t i s anticipated that the EU will follow this assistance with a Euro 12 million project (2008-201 1) to continue these qualitative investments in four regions. Similarly, IDB’s US$19 million Basic Education Project has supported in-service teacher training, public school classroom construction and rehabilitation, and institutional strengthening. Finally, it i s anticipated that C I D A will commit approximately CAN$200 million between 2008-13, to build new schools (as many as 400), strengthen institutional capacity o f the MENFP, continue its quality improvement programs in the Arbonite region, and improve vocational training. I t i s worth noting that recent figures indicate that as o f April 2007 World Bank and Caribbean Development Bank are major donors, with US$25 million and US$IO million, respectively. 98 Haiti: Public Expenditure Management and Financial Accountability Review disbursements. A s o f December 31, 2005, only US$22 million out o f more than US$ 86 million was disbursed, resulting in a relatively l o w disbursement ratio o f 25 percent. Disbursement under ICF framework appears relatively l o w at less than 20 percent while donors’ aid flows provided outside the ICF framework (linked to the “ rentrCe scolaire” and back-payment o f contractual teachers’ salaries) record higher disbursement rates: about 32 and 35 percent (see Table 46). In terms o f areas o f intervention, governance and to a lesser extent quality record relatively good disbursements o f resources: 41 and 30 percent, respectively. At less than 20 percent o f total disbursement, activities related to increasing access might be limited by poor disbursements o f aid flows. But it i s also due to delays in MENFP execution o f school construction programs, which involve much lengthier procurement proce~ses.’~ If this disbursement trend were to continue, barely 60 percent o f total commitments would be disbursed at the end o f the ICF period. This raises an important concern given the fact that increasing access i s at the center o f the strategy for achieving EFA goals. Table 46. Donors’ Commitments and Disbursements under ICF and Other Interventions (US$ thousands, unless otherwise indicated) ICF Other interventions Axis of ICF Framework Other planned intenentions Interventionsoutside the ICF intervention Commitments Disbursements Commitments Disbursements Commitments Dkbursements (2004-07) (as of Dec 31, (2004-07) (as of Der 31, (2004-07) (as of Dcc 31, 2005) 2005) 2005) Go\ ernace 5.560.9 1.067 9 8.030.8 3.187 1 2.326 0 2.326 9 Quality 10,332.9 1,716.3 17,3 19.6 5,907 2 2,297.4 1,270.6 Access 29,116.2 6,470.1 377.4 89.6 5,898.0 48.0 Efficiency 1,717.3 0 3,169.0 0 Youth, Sport 50.0 0 0 and nvil education Total 46,777.3 9,254.3 28,896.9 9,183.9 10,521.4 3,644.6 Ratio 19.8 31.8 34.6 Disbursement/ Commitments (“w Source: Ministere de I’Education Nationale, de la Jeunesse et des Sports, Cadre de CooperationIntCrimaire. Bilan des Finaneements et des Rkalisations, PCriode de Juin 2004-Deeembre 2005. Mars 2006. 79 While MENFP lacks capacity in school construction, it i s important to highlight F A E S as a positive experience in this respect. Over the past decade, F A E S has gained experience in the execution o f school rehabilitation and construction projects. With good results, the Bank i s implementing a l l its school rehabilitation and construction investments through FAES. 99 Haiti: Public Expenditure Management and Financial Accountability Review 100 8 N 9 G. 0 m 0 vi 2 c? d N o? c '9 0 m 0 9 m 2 r- m 'if io e - 9 m m e vi m 8 d o\ 8 ? m s m 8 0 vi u! m. d G. c 0 9 2 2 5 p? 9 2 d z 0 0 8 vi 9 0 c? 2 - r. p? r- 3' m N d > s t .- - .- . " rD m a U t S 3 0 9 ": u 0 m u e < vi 8 d 9 m N - c? U 8 d Haiti: Public Expenditure Manaaement and Financial Accountability Review 5.25. Donors also provide specific interventions to ensure implementation o f the school year. Table 48 provides details on donors’ interventions to finance school year 2004/05. Unfortunately, existing data i s limited to commitments. Because o f the lack o f disbursement figures, i t i s not possible to provide a clear picture o f the impact of donor financing on the FY2004/05 school year. M u c h in line with donors’ involvement in the education sector as a whole, USAID, CIDA, European Union and IDB stand out as key players in the provision o f assistance during the 2004/05 school year. Altogether, they account for about 90 percent o f the financing commitment o f the school year, with U S A I D representing more than 40 percent o f expenditures over the period June 2004-December 2005. ’able 48. Donors’ Commitments to Financing School Year 2004-05 (in thousand US$) Components Efl IUB ClUA USAID World 0lh.I UNICEF Total Bank School building 1 696.0 405.1 162.1 2 263.2 School rehabilitation 1780.0 1 854.8 405.1 365.4 900.0 4 939.9 School furniture 810.3 5 840.0 333.4 64.4 844.9 Support to school 2 142.9 1458.5 4 900.0 1 089.0 9 126.5 fees School canteens 810.3 7 447.5 Distribution o f 810.3 administrative documentation Payment o f arrears to 1053.4 1053.4 contractual teachers Logistic and 162.1 162.1 management fees Total 3 922.9 3 550.8 5 509.0 10 1089.0 698.7 1 126.4 26647.8 740.0 Percent o f total (14.7) (13.3) (20.6) (40.3) (4.1%) (2.6%) (4.2%) (100%) Source : Bilan des Financements et des RCalisations- Pkriode : juin 2004-dkcembre 2005 ;mars 2006 5.26. I n terms o f components, support to school fees and cafeteria as well as school building and rehabilitation account for the largest shares o f donors’ interventions in the education sector. They represent about 27 percent and more than 60 percent o f the total donors’ financing, respectively. 5.27. Although several donors are active in the education sector, there i s still a significant gap between available resources and needs on the ground. While there has been progress under the ICF, results o n the ground have been slow because conventional modalities for delivering education services t o the poor are limited by the lack o f institutional capacity. Furthermore, most external education funding committed under the ICF aims t o improve educational quality in public schools. M u c h more needs to be done to expand access by addressing both demand-and-supply- side constraints. 103 Haiti: Public Expenditure Management and Financial Accountability Review 5.1.4 Identifying Weaknesses, Issues and Challenges in B u d g e t Process and Public Expenditure Management 5.28. Several weaknesses, issues and challenges would need to be addressed to strengthen public finance management in the education sector. Weaknesses in this area are at the center o f the poor quality o f management o f public resources and ultimately impair efforts to increase education outcomes. An increase in resources will not be enough t o improve education outcomes. Efficient and transparent management o f public resources would be crucial to achieve improved education indicators. 5.29. Budget preparation does not follow any specific timetable, and budget proposals have not been backed up by a sectoral strategy. As with other line ministries, the process starts with the instructions o f the MEF through the "lettre de cadrage'. There i s no formal planning and consultation mechanism established within the Ministry o f Education. Although the sector has a clear strategy, budget proposals have not been based o n that strategy, and allocations do not reflect any strategic achievable objective. M o r e often the sector's budget proposals are overly optimistic beyond its demonstrated absorptive capacity. They are not often substantiated by clear justification and spending targets. The result i s that more often than not the proposals from the Ministry o f Education do not obtain the MEF's agreement. In fact, as for most o f the sectors, budget allocations are just the renewal o f allocations o f the past year with an increase n o t often substantiated by clear justification and spending targets. 5.30. T h e preparation process i s also affected by the centralized approach adopted within the Ministry, which results in little involvement o f the Ministry's directions and entities. The Direction de l a Planification et ce l a Cooperation Exteme (DCPE) i s in charge o f the preparation along with the Department des Affaires Administratives (DAA). Other central and regional departments are not involved in the process. Thus the budget proposals do n o t account for needs expressed by the ministry's entities other than those articulated by the DCPE and DAA. This i s an important issue given the discouraging effect that i t could have o n staff o f the Ministry, who questioned the usefulness o f their contribution to preparation o f the budget. 5.31. Budget execution in the sector suffers f r o m the lack of detailed and accurate data. Aggregate execution might hide difference in budget items. But detailed information o n the execution i s scarce and incomplete. Also, the lack o f clear spending targets makes i t difficult to judge the effectiveness o f budget execution in terms o f achieving the objectives set. Only recently has the GoH prepared a l i s t o f poverty reduction items in the context o f the H I P C tracking mechanism. T o what extent the l i s t reflects the priorities o f the PRSP, currently under preparation i s questionable. Specifically, to what extent the l i s t i s consistent with the sector strategy and the EFA's objectives, i s unclear. However, one might argue that the I-PRSP has been a solid basis t o design the list. The challenge n o w i s to convert this l i s t into specific budget lines to ensure the effectiveness o f the trackmg and monitoring exercise. For the education sector, this would require designing specific spending targets for these items and following-up with their execution. 5.32. Governance issues are also a major constraint to effective expenditure management in the education sector. Lack o f clearly defined procedures, transparency and accountability in the use o f public funds often results in poor expenditure management in the sector. Weak capacity o f the Ministry o f Education to oversee and regulate the dominant non-public education sector has a negative impact o f the sector as a whole. In fact, lack o f resources and weak capacity in the Ministry o f Education (MENFP) have resulted in distortions in the education sector and hampered 104 Haiti: Public Expenditure Management and Financial Accountability Review the MENFP from fulfilling i t s normative functions o f regulation and quality control. As a result, quality in education has sharply deteriorated. A high percentage o f schools do not meet basic sanitary and safety standards (16 percent are houses, 33 percent are in churches, and 9 percent are open-air). 70 percent o f schools are not accredited by the MENFP (10,932 out o f a total o f 15,664 primary schools), while 60 percent o f teachers in the non-public sector are not appropriately qualified. Only 360 school inspectors are responsible for accreditation, pedagogical supervision and administrative support, or an average o f one inspector for every 6,000 students. Also, inspectors are poorly trained and equipped. Because o f the increasing number o f non-public schools and inadequate regulation, the education sector i s increasingly segmented, with little standardization or curricula, pedagogy andor assessment, making systemic qualitative improvements difficult t o achieve, much less measure. Only a minority o f non-public education providers belong to formal school associations (for example FONHEP and CONEP) and aim to provide the conditions needed for quality education. A formal link with the public sector i s missing to harmonize quality standards and coordinate education policy and reforms in the education sector. 5.33. Another issue in education expenditures management concerns the capacity o f the Ministry to better manage i t s own human resources. For instance, there i s a lack o f qualified teachers while the MENFP or the University offers several scholarships to students to study abroad. The future o f these students i s not guaranteed and most o f them do not come back to the country because incentives are not in place. T h e weak human resource capacity becomes a recurrent phenomenon. T h i s shows that improving budget management in the education sector goes beyond the simple policy o f increasing public resources. The design and implementation o f an effective human resource management policy would be critical if the education sector i s to improve public spending effectiveness. 5.1.5 Policy Recommendations and Priorities 5.34. Seven policy recommendations appear as priorities to ensure improved education outcomes in view o f achieving EFA goals. 0 Increase the allocation of resources to the education sector, with a focus on primary education. As documented above, l o w levels o f education indicators in Haiti result in part from the very l o w public education spending, about h a l f o f which goes to primary education. Improving education outcomes require that the level o f resources be increased. More specifically, as mentioned above, putting Haiti on the right track to achieve the EFA targets by 2015 would require that the education sector receives about 21 percent o f total government recurrent spending (net o f debt service), o f which at least 50 percent goes to primary education throughout this period. Rapid expansion o f primary education access (and more gradual improvement o f education quality) would improve basic service provision and would lead to sustainable job creation, through employment o f thousands o f new teachers: the education sector i s the largest source o f employment in the country after agnculture. Recent increases in allocations go in the right direction. But the challenge i s now for the Government to sustain these flows o f resources over a medium to long-term period to ensure that Haiti stays on the right track to achieve the EFA goals. On the one hand, the l o w resource base makes i t difficult for Haiti to provide required recurrent spending to achieve the EFA goals. On the other hand, volatility o f donors’ resources does not provide any guarantee for sustained investment flows in the future. 105 Haiti: Public Expenditure Manaqement and Financial Accountability Review Execute the Medium-Term Expenditure Framework (MTEF). Improving education outcomes requires that the Government has a medium-term perspective. T o this end, the execution o f the MTEF, adopted in the context o f the development o f Haiti's national strategy known as Education for All, should ensure that required level o f resources t o the primary education are protected. However, an effective execution o f an MTEF i s linked to a sound budget preparation and execution process. Currently, weaknesses identified in this area calls for strengthening o f yearly preparation and execution o f an education budget. Establishing a clear timetable, setting responsibilities for the preparation process, and linking budget proposals to the sector Education for All strategy will help reinforce the execution o f the adopted MTEF in order to make i t a sound policy tool. The modernization o f the Department o f Administrative Affairs (DAA) i s crucial. The DAA i s responsible for the overall budget execution (in collaboration with the Regional Education Departments, DDE), and i s the primary interface with the Ministry o f Finance. The W o r l d Bank's ongoing Phase One o f An Education for All Adaptable Program Grant could serve as the main instrument t o reinforce technical capacity o f the DAA. I n the context o f the project, capacity building assistance will be provided to reinforce the ability o f the DAA to manage public resources. Improve budget data and information. T h i s i s perhaps the utmost priority for improving budget management in the education sector. Budget data would need to be improved t o ensure that: (i) sector budget proposals are meaningful; and ( ii) the traclung o f expenditure execution i s done. Ensuring an updated database o f budget proposals and allocations would help the MEWFP prepare more accurate and realistic budget proposals. The MENFP would also need to prepare data on quarterly execution o f government spending in the education sector. T h i s would help facilitate the traclung exercise o f Government spending and their impact on education outcomes. Following up the execution o f spending and assessing whether actual expenditures are in line with the EFA goals cannot be done with data that currently i s weak and inaccurate. Improve public resource management. Increased resources t o the education sector would not achieve higher education outcomes unless public resources are managed effectively. The education sector currently faces issues o f transparency, accountability and effectiveness o f use o f public resources. There i s a need to strengthen transparent, participatory procedures for decision-malung and resource management. Specifically, involving parents in school management decisions will improve transparency, accountability and effectiveness o f the use o f public resources at the school level. T h i s requires strengthening the role o f parent association and school management committee. While currently more than 70 percent o f all schools have some f o r m o f parent association and/or school management committee, many o f these communities are s t i l l weak and dominated by the school director and hisher associates. Ensure sustained investment flows and targeting specific investment programs. Education outcomes will not improve o n a sustained basis unless investment needs are met. This requires that investment resources are provided o n a medium-to-long term basis. M o r e specifically, targeting investment programs will help improve access and quality. An area o f particular focus i s the pre-service teacher training, which has not benefited f r o m any investments in at least a decade. Facilities are in poor condition, teacher trainers emphasize out-dated teaching methods, and trainees lack the material they need to develop both theoretical and practical pedagogical competencies. Sustaining investments in the education sector also requires that disbursements o f aid flows match commitments to 106 Haiti: Public Expenditure Management and Financial Accountability Review ensure predictability o f resources and the execution o f investment programs. Moreover, sustained investment flows need to be accompanied by sufficient recurrent resources. In turn, this implies that the Government’s budget i s able t o support additional recurrent spending and calls for an effort on the Government side t o increase domestic revenue through adequate revenue-enhancing measures. Create and Reinforce Public/Non Public Sector Partnerships. Because o f the large share and critical role that the non-public sector plays in the education sector in Haiti, improving education outcomes can not be achieved by implementing policies focusing solely on public education. In fact, even large increases in resources and improvement in management o f public resources in the public education system would n o t be enough to achieve the EFA goals. Publichon public sector partnerships should be created and reinforced. Access t o non-public schools stands out as an area where this partnership could be explored. A possibility that could be explored i s the provision o f subsidies t o poor students currently not enrolled in school so they could attend n o n public primary schools free o f charge.*’ Reinforce technical capacity in public finance management. Technical capacity in public finance needs t o be reinforced as it has been detrimental t o the quality o f budget preparation, planning and execution. As a first step, the MENFP would need to reinforce capacity o f staff involved in budget preparation and execution to ensure that the production o f detailed budget data i s made o n a regular basis. An area o f particular emphasis is the investment budget, where lack o f capacity has resulted in a lack o f good understanding o f the investment budget. M o s t notably, i t i s critical that the MENFP develop technical capacity t o track and quantify donors’ resources injected into investment progams and projects, and reconcile this information with the investment budget as currentiy recorded. This requires that technical assistance i s provided t o staff concerned with the preparation and execution o f the budget. But this i s not a specific issue to the education sector. I t cuts across sectors and calls for a comprehensive approach, which might require full-fledged capacity building in the specific area o f the investment budget. 5.2 HEALTH SECTOR 5.2.1 Background and sector objectives 5.35. Food deprivation and limited access to health care, because o f poor infrastructure, lack o f public funding and lack o f qualified personnel and drugs, has resulted in dire health conditions for Haiti’s poor. Table 49 provides some background statistics pertaining to the health sector. Only 28 percent o f the population has access t o health care. Health services are predominantly provided by non-public institutions (70 percent) and the quality i s poor. Infant and maternal mortality i s the highest in the Americas, more than four times the regional average. O f children younger than 5 years o f age, 23 percent endure chronic malnutrition, and 5 percent suffer from acute malnutrition (which causes stunting and reduces brain development). The mortality rate 80 In the context o f the Education for All Adaptable Program Grant, Bank staff calculated that the amount o f the per student subsidy to US$90 to cover average tuition fees (US$70) and learning material costs (US$20) in non public schools. T h e total cost o f this program i s estimated at US$19 million. See World Bank Education for All Adaptable Program Grant, April 2007. 107 Haiti: Public Expenditure Management and Financial Accountability Review for children less than 5 years o f age i s 86 per 1,000, high by LAC standards." The l i f e expectancy at birth o f 53 years i s lower than the average for Low-Income Countries. Haiti also faces a high incidence o f HIV/AIDS with U N A I D S estimating that 5.6 percent o f the adult population i s HIV positive. 5.36. Authorities have defined a health sector reform, which aims to improve the coverage and the efficiency of the health sector. The companion National Strategic Plan for Health Sector Reform has been defined for the 2003-2008 period and aims t o enhance the Haitian population's health status in the context o f acheving the MDGs. B o x 8 summarizes the objectives and policy actions (See Appendix 5.1 for more details). But, this strategy does not include spending targets and appears relatively ambitious, according t o the l o w level o f resources devoted to the health sector. Per capita expenditure on health in Haiti i s only US$26, while it averages US$30 in LIC, US$36 in Sub-Saharan A h c a and US$222 in L A C . Furthermore, nearly h a l f o f this total expenditure i s financed by Haitian households, despite their l o w income level. 5.37. Implementing this strategy successfully and achieving higher health outcomes would require increasing the level of resources to the health sector and improving the resource management. However, increased allocations to the sector would not be enough. Ensuring that public resources have impact o n access and quality o f health services i s conditional o n improving the quality o f spending in the sector. Overall, increasing health outcomes to international standards depends on the effectiveness o f public expenditure management in the sector. But given the importance o f the private-sector, it i s also linked t o the ability o f the public sector to createheinforce partnership with the private-sector. The high level o f under-five mortality could b e partly explained by the l o w level o f immunization and the default in the management o f diarrhea and Acute Respiratory Infection (ARI). According t o the preliminary report o f the last DHS ( E M M U S W , 2005-2006), only 41 percent o f children aged between 12 and 23 months have been completely immunized; 48 percent have been partially immunized and 11 percent have not been immunized. According to E M M U S W , 2005-2006, only 20 percent o f children suffering f r o m fever or ARI have their mothers (or relative) request advice or treatment to health services; while 40 percent o f under- five children suffer f r o m these symptoms and more that 2 0 percent o f deaths o f under-five children were caused by pneumonia in 2000 (See WHO database). 108 Haiti: Public Expenditure Management and Financial Accountability Review Child health Infant Mortality (per 1000) 86 83 57 26.5 79.5 100.5 Under 5 Mortality (per 1000) 135 118 86 31.4 121.6 168.2 Prevalence o f immunization (EIP, %) 30 34 41 Prevalence o f Child Malnutrition, 17.2 7.3 38.9 29.4 underweight (%) Reproductive health Maternal Mortality (per 100 000) 476 523 194* 921* 684* Birth attended by skilled staff (%) 60 88.2 40.7 42.3 Prevalence o f contraception (%) 12 28 32 72.3 39.5 21.9 Fertility Index 4.7 4 2.5 3.7 5.3 Diseases AIDS Prevalence (%) 3.8 0.5 1.7 6.1 Malaria Prevalence (“76) Tuberculosis Cases (uer 100 000) 3 86 63.6 224.2 IDOTS detection r a: ( t%) 4.9 61.7 43.8 50.9 * These numbers are modeled estimates which usually overestimate the maternal mortality compare to nation, estimates. For example, for 2000 the modeled estimate for Haiti i s 680, while the national estimate i s 523. Sources: EMMUS I11and IV, and WHO online database. 109 Haiti: Public Expenditure Manaqement and Financial Accountability Review Box 8. Strategy for the Health Sector in Haiti General Objectives of the 2003-2008 National Strategy Plan for Health Sector Reform A reduction o f at least 50 percent in maternal mortality; A reduction o f at least 50 percent in the rates o f infant and under-five mortality; A reduction o f 30 percent in the incidence o f H N / A I D S infection; A reduction o f 30percent in mortality associated with H I V / A I D S ; A reduction o f 10 percent in mother-to-child transmission o f the infection; A reduction of 3 0 percent in the incidence o f Tuberculosis; A reduction o f 50 percent i n mortality associated with malaria; A reduction o f 50 percent in disorders due t o lack o f iodine; and Control o f some diseases (lymphatic filariasis) and elimination (neonatal tetanus) o r eradication o f other diseases (poliomyelitis, measles). Intermediate Objectives 1) Decentralize the health system; 2) Improve the supply o f health care; 3) Revitalize and expand the public hospital network in order t o increase the supply o f health care and improve its quality; 4) Improve regulation of the sector by giving the MPHP tools for managing the entire sector, thus strengthening the leadership o f the MPHP and i t s regulatory role; 5) Modernize the health information system; 6) Develop human resources; 7) Guarantee the population’s access t o essential drugs; and 8) Increase and rationalize financial resources in order t o allow for improvements in the supply service in the context o f sector-based reform. w e : Ministry of Public Health and Population, Strategic Plan for Health Sector Reform, March 2QO4. I - . 5.2.2 T r e n d s and Structure o f H e a l t h Expenditures 5.2.2.1 Total Health Allocation 5.38. Allocation to the health sector doubled in real terms f r o m 1.2 billion gourdes to 2 billion on average between FY2001/02 and FY2006/07, following the government budget increase by 30 percent over the same period. As a result, the share o f Ministry o f Health in total budget increased from 5.4 percent to 6 percent o n average between FY2002-04 and FY2005-07. 5.39. However, allocations to the health sector still compare poorly by international standards. It stands below Madagascar’s figure (8.4 percent)’* and i t i s far below that o f the Dominican Republic (10.3 percent)*3 , Burkma Faso (10.3 p e r ~ e n t ) ’and ~ Zambia (13.1 percent)*’. 5.40. The priorities set for the health sector have not yet translated into a sufficient increase in budgetary allocations to the sector. According t o the World Health Organization, 82 Madagascar - Public Expenditure Review (PER) 2004 - The challenge o f poverty reduction, W o r l d B a n k (2005). 83 Note that the figure for Dominican Republic refers t o executed spending. See Dominican Republic - Public Expenditure Review: Reforming institutions for a more efficient public expenditure management, W o r l d Bank (2004). “Burkina Faso - The budget as centerpiece o f PRSP implementation - Public Expenditure Review, W o r l d Bank (2005). 85 Zambia - Public Expenditure Management and Financial Accountability Review, W o r l d B a n k (2003). 110 Haiti: Public Expenditure Manaqement and Financial Accountability Review Haiti devotes only US$84 per capita t o health services (see Figure 12), less than Ghana (US$98), and Guinea (US$95). In fact at less than 3 percent of GDP, the public sector’s contribution to the health sector accounts for only about 40 percent o f total health expenditures (estimated at about 7.5 percent o f GDP according to the W o r l d Health Organization). Figure 12.Per capita total expenditure on I Figure 13.Total Health expenditure in L I C in 2003 I health-in U S $ in L I C in 2003 (Yo of GDP) I o 50 100 150 200 0 2 4 6 8 10 12 Source: WDI, 2006 ource: WDI. 2006 ‘ 5.41. Thus, the private sector contribution to health expenditure i s quite high by region& standwds and comparable countries. For instance, in Nicaragua, the contribution o f the private sector io the health expenditure i s 10 percentage points lower than that o f Haiti. Furthermore, “out- of-pocket expenditure’’86 represents about 70 percent o f private expenditure (which represents 62 percent o f total health expenditure). In other words, Haitian households directly finance nearly h a l f o f the total expenditure o n health (43 percent), i.e. the same proportion o f spending made by Nicaraguan households whose income i s twice as high. 5.42. These figures suggest that the limited impact of public spending on health outcomes might result f r o m the combination of low levels o f public resource allocation to the health sector and the inability o f the private sector to deliver basic and quality health services affordable to the poor Haitian. This, combined with inefficiency o f spending, lock Haiti in abysmal health conditions. The figures also suggest that under these circumstances, reaching the internatronal goal o f 15 percent o f the budget t o the health sector by 2015 set forth in Abuja (Nigeria) would be a daunting challenge that Haiti’s own resources cannot afford. Pulling external resources to finance health spending i s therefore a priority for Haitian authorities. 5.43. Unfortunately, despite consistent efforts undertaken, external resources allocated to health remain among the lowest within the LIC group. The W o r l d Health Organization estimated that in 2003, the external contribution t o the health expenditure was only about 12 percent o f total expenditure, lower than for Rwanda and Liberia at 54.5 percent and 32.3 percent, respectively. 86 “Out o f pocket expenditure” i s any direct outlay by households, including gratuities and in-kind payments, to health practitioners and suppliers o f pharmaceuticals, therapeutic appliances, and other goods and services whose primary intent i s t o contribute t o the restoration o r enhancement o f the health status o f individuals o r population groups. I t is a part o f private health expenditure (WDI, 2006). 111 Haiti: Public Expenditure Manaqement and Financial Accountability Review Figure 14.Private health expenditure in selected Figure 15.Public health expenditure in selected LIC in 2003 (YO of GDP) L I C in 2003 (YOo f GDP) 0 2 4 6 8 10 1 0 0.5 1 1.5 2 2,5 3 3.5 4 Source: WDI, 2006 Source: WDI, 2006 5.2.2.2 Analysis of Budget Structure 5.44. While recurrent spending accounted for more than 70 percent of Health total budget over 1Fk72002-04, the structure of health budget has shifted toward a larger share of investment, which accounts for more than 60 percent o f total budget in FY2005-07. However, this shift results mainly from the large allocations provided t o the investment budget in FY2006107 (see Figure 16), in the amount o f 4.2 billion gourdes (nearly 80 percent o f total FY 2006107 health budget). . 112 Haiti: Public Expenditure Manaqement and Financial Accountability Review Figure 16. Allocation Distribution, N 2 0 0 4 - 0 5 to FY2006-07 FY2004-05 FY2005-06 FY2006-07 Investment Recurrent Sources: L e Moniteur, Journal Officiel de la RCpublique d’Haiti and Staff Calculations. 5.45. As a result, the increase in sector stemmed mainly from the rise in the investment budget, which saw a three-fold increase in real terms over the period under review: 364.4 million gourdes on average in FY2002-04 to 1.2 billion gourdes in FY2005-07 (see Table 50). This sharp increase in the Total Investment in % o f total 27 63 investment budget more than budget Total Nation Budget 21264.6 35191.0 compensates for the 9 percent decline in Shares o f Health sector in Nation 5.4 6.0 recurrent budget o n average, over the Budget (in %) same period. T h i s decrease in real Total Nation Recurrent Budget 14073.6 11792.5 recurrent expenditure i s mainly due t o a (Excl. Interest Payments) decline in expenditure o n goods and Percent o f Total (Excl. Interest 5.7 6.6 Payments) services, which fell by more than 45 Memo Item: percent o n average over the period under INominalGDP 85,700 94,028 review. Sources: L e Moniteur, Journal Officiel de la Rkpublique d’Haiti and Staff Calculations. 113 Haiti: Public Expenditure Management and Financial Accountability Review 5.46. These figures raise two m a j o r concerns. First, the large increase in the investment budget during the recent years might face limited Figure 17.Recurrent Expenditure (HTG million), absorptive capacity and further result in FY2001/06 inefficiency o f spending. However, this does not mean that investment resources need t o be limited to the sector’s current absorption capacity. Indeed, reaching internationally required levels o f health expenditure implies the increase in allocations o f resources t o the sector. It also implies that the sector’s technical and human capacity as well as management o f WOI-02 F’fO2-03 WO3-04 WM-05 FYO5-06 FY06-07 WO1103 WO4106 Averages resources be significantly improved. Wages and Salaries 0 Goods and senices Others Second, the decline in expenditure o n goods and services i s worrisome as i t Source: L e Moniteur, Journal O f f c i e l de l a RBpublique d’HaYti and Staff Calculations. might have affected the delivery o f health services. Some health centers in Port-au-Prince lack basic functioning resources and cannot provide the basic services t o the population. !PLB 2004-05 PIP 2005-06 PIP 2006-07 136.5 60.3 69.8 303.6 209.1 2996.1 440.0 269.4 3065.9 Sources: Ministry of Planning and External Cooperation. 31.0 22.4 2.3 69.0 77.6 97.7 100 100 100 5.2.2.3 Analysis o f Budget Execution 5.47. Because o f the p o o r quality o f data, the analysis o f the execution o f expenditure T a b l e 52. Health Sector. Budget Execution in the H e a l t h M i n i s t r y should b e interpreted (real HTG million), F?2002/07 with caution. Table 52 shows that in real terms, spending declined from an average o f 92 percent o f the Health Ministry’s total budget Total Budget allocated 1233.6 1125.1 Total Budget executed 1097.9 682.4 over FY2002-04 t o about 60 percent o n average Execution rates (in YO) 92.5 60.3 over the period FY2005-07. The execution o f Total recurrent allocated 869.2 777.9 the investment budget has been particularly Total recurrent executed 954.7 636.8 slow since FY2004/05, reflecting the slow pace Execution rates (in YO) 109.9 81.2 o f execution o f projects and programs, limited Total investment allocated 364.4 347.2 Total investment executed 143.1 45.6 expertise and experienced staff in the Ministry Execution rates (in YO) 45.4 12.6 t o accelerate spending procedures and satisfy iources: L e Moniteur. Journal Officiel de l a procurement requirements. However, it also Rbpublique d’HaYti and Staff Calculations. reflects poor budget planning in particular during the preparation o f the investment budget. 114 Haiti: Public Expenditure Manaqement and Financial Accountability Review 5.48. C o n t r a r y t o t h e investment budget, t h e execution o f the recurrent T a b l e 53. Health Sector. Recurrent Budget budget has been remarkable, varying at high levels between 85-100 percent over the past five years (see Table 53). The execution rates o f recurrent spending are mainly driven by spending o n wages and salaries, which account for on average about 90 percent o f total expenditures executed over the period FY2004-06. This expenditure category i s generally executed in full, reflecting the Government’s resolve t o meet its salary payment obligations to the NA NA c i v i l servants employed in the health sector. “Direct payment” procedures are used t o process the payments o f the wages and salaries, leaving a marginal recourse t o the Sources: L e Moniteur, Journal O f f c i e l de l a Rkpublique d’HaYti and Staff Calculations. use o f comptes courants. Only about 1 percent o f total personnel expenditures were processed through comptes courants during the period FY2001103.87 5.49. T h e execution o f spending o n goods and services has been o n a declining t r e n d since 2004, reflecting mainly tighter control over t h e use o f comptes courants, increased delays in . processing payment documentation (on the part o f t h e Ministry), and the payment order (on the p a r t o f the Treasury). A large share o f these expenditures was executed through the USR of conptes coumnts over that period. For instance, more than 60 percent o f spending o n goods znd , services were processed through comptes courants in FY2001-03. However, the use o f cornptes , ’ courunts has been gradually reduced. In FY2002103, only 17 percent o f those expenditures were executed through comptes courants, as the result o f the Government’s actions to gradually phase out the use o f comptes courants. 5.50. T h e l o w levels o f the execution o f spending o n goods and services i s symptomatic o f the i lfunctioning o f most o f t h e health centers, w h i c h operate m o r e often w i t h o u t the basic l needs (essential drugs, vaccines, small equipments, etc.). I t also highlights the limited impact that health spending has had o n health outcomes in an environment where intra-sectoral efficiency o f spending (“value o f money”) i s weak. 5.51. Table 54 below presents t h e shares and execution rates o f selected items during the FY2003-06 period. Spending o n maintenance o f equipment and building averaged o n l y 0.2 percent and 0.4 percent o f total recurrent budget, respectively. Moreover, t h e i r execution rates are relatively l o w (57 percent and 47 percent o n average) and exhibit a downward trend over the period. This negative trend i s worrisome given the importance o f maintenance spending for investment stock. Small equipment and medical and surgical supplies represent 0.1 percent and 2.5 percent, respectively, o n average o f total recurrent budget. Their average execution rate, although higher than those o f maintenance items, declined over the period FY2005-07 compared to FY2002/04: from 84 percent to 76.4 percent (health supplies and small equipments); and f r o m 89.4 87 Starting in FY2004/05, the execution data do n o t unfortunately recorded the spending made through the ‘’comptes courants ”. 115 Haiti: Public Expenditure Management and Financial Accountability Review percent to 44.6 percent (medical and chirurgical supplies). M o r e troubling is the large decline recorded in the execution o f spending allocated to medical and surgical equipment. Put together, the declining trend might explain the l o w health outcomes of Haiti. Table 54. Health Sector. Share of selected items in Recurrent Budget and execution rate (in %), 4 6 8 Executed 0.18% 0.24% 0.09% 3.56% 3.22% Execution rate 40.98% 3 1.57% 56.46 80.22% 82.45% * z g Allocated Executed 0.14% 0.13% 0.25% 0.18% 0.06% 0.07% 1.26% 1.31% 1.22% 0.72% N Execution rate 85.28% 69.26% 1 1 1.76% 98.57% 56.42% Allocated 0.16% 0.37% 0.05% 2.03% 1.21% 4 8 Executed 0.10% 0.18% 0.04% 1.OO% 0.00% Execution rate 55 14% 45.60% 81 15% 45.22% 0.00% \o Allocated 0 15% 0.34% 0 04% 2.65% 0.00% 3 0 08% 0.17% 0.04% 1.43% I 5 ‘ Execu?ed Execution rate 46.30% 40.12% 71.60% 44.02% I Sources: L e Moniteur, Journal Offciel de la RCpublique d’Ha’iti and Staff Calculations. \ . 5.2.3 Public Investment P r o g r a m s in the Health Sector 5.52. While resources devoted to the health sector’s investment programs (including foreign aid), decreased over the past two fiscal years, they are projected to increase nearly seven-fold, reflecting huge flows of expected resources from the donor community. However, four main issues are worth mentioning. First, the large increase in the projected investment flows reflects the l o w base o f resources in the previous fiscal years 2004-06. Second, the extent to which these resources reflect the required spending levels to implement the National Strategic Plan for Health Sector Reform i s unknown as the strategy does n o t have clear spending targets. Third, a large share o f donor resources i s provided outside the budget and i s not accurately captured in the PIP. Fourth, the level o f resources projected in the PIP takes little or n o account o f the associated recurrent spending that the execution o f the PIP might bring about. But this i s not a specific issue o f the health sector. It i s a more general issue o f a parallel budget preparation process, which i s characterized by the MEF preparing the recurrent budget while the Ministry o f Planning i s in charge o f the PIP. 116 Haiti: Public Expenditure Manaqement and Financial Accountability Review 5.53. Similar t o most p r i o r i t y sectors, investment programs in the Table 55. Resources f o r PIP in the H e a l t h Sector. health sector are mainly financed Comparison to other P r i o r i t y Sectors, by donors. N e a r l y 80 percent o f the sector FY2005/06 PIP was financed by foreign aid (see Table 55). This i s in the same range o f donors’ share o f the PIP in other priority sectors, most notably health, agnculture, and transports. s o f Health in total 5.54. Table 56 below provides Sources: Ministry of Planning and External Cooperation and Staff details o n donors’ interventions in Ca’culations. the health sector t h r o u g h the I C F f o r t h e p e r i o d 2004-006. USA, Canada and UN stand o u t as k e y players in the sector. Altogether, they account for about 85 percent o f the sector’s investment program over the 2004-06 period, with USA representing about 40 percent o f the sector’s investment pledges and 60 percent o f disbursements over the penod June 2004-December 2005. 117 Haiti: Public Expenditure Manaqement and Financial Accountability Review Table 56. Donors’ financing under ICF (in million $US unless otherwise indicated), June 2004- December 2005 Budget allocated in ICF Rationalization o f the 21.15 I I I I I I I 21.15 Sector Fight against I 15.46 I 12.8 2.06 I I I 30.32 HIVIAIDS Fight against Maternal I 10.10 I 6.56 I I I I 16.66 and Infant Mortality Immunization I 5.0 I 4.0 I I I I 9.0 Health emergencies I 1.73 2.5 3.8 I I I I 8.03 Equipment and Material I I I I I 0,07 I I 0.07 Personnel Training I I I 8.95 I J I I 8.95 Health Facility I I I I 0.70 I I f 0.7 Restoration HIViAIDS + Child I I I I I I 91.96 I 91.96 survival Other I 2.03 I 35.3 4.69 4.08 0.5 1.21 47.81 Total 21.15 33.92 2.5 71.44 7.45 4.15 92.46 1.21 234.28 Percent of total 9.03 14.48 1.07 30.49 3.18 1.7 39.47 0.52 100 Contributions SpentBudget2004-05 3.25 23.62 0.42 19,89 3.30 4.30 77.90 0.34 133.02 - Source :Bilan des Financements et des RCalisations PCriode : jhin 2004-dCcembre 2005 ; mars 2006 5.55. I n terms of components, support to fight HIV/AIDS and for child survival account for the largest shares of donors’ interventions in the health sector. They represent about 99 percent o f USA financingggand more than 50 percent o f the total donors’ financing, respectively. 5.56. Although several donors are active in the health sector, there i s s t i l l a significant gap between available resources and needs on the ground. While there has been progress under the ICF, results o n the ground have been slow because conventional modalities for delivering health services to the poor are limited by the lack o f institutional capacity. Furthermore, most o f the external health funding committed under the ICF aims at improving health care quality in public health facilities. M u c h more needs to be done to expand access by addressing both demand- and supply-side constraints, particularly in rural areas. 5.2.4 Weaknesses, Issues, and Challenges in Public Expenditure Management 5.57. Similar with most line ministries, the budget preparation process in the Health Ministry i s weak. There i s n o formal planning and consultation mechanism within the Ministry. There i s also n o specific timetable t o prepare the budget. While there i s a sector strategy and clearly defined actions to improve the health outcomes, they have not served to inform budget gg The different components are n o t disaggregated. 118 Haiti: Public Expenditure Manaqement and Financial Accountability Review proposals. As a result, proposals for allocations made by the Health Ministry do not reflect any strategic objective. They are n o t often substantiated by spending targets. In fact, l i k e f o r most o f the other sectors, budget allocations to the Ministry o f Health are just the renewal o f allocations o f the past year with an increase not often substantiated by clear justification and spending targets. 5.58. The dual preparation process of the budget results in poor planning o f the investment budget in the Ministry of Health. Moreover, recording problems occur. Expenditure recorded as investment under “Programmes et Projets” certainly includes recurrent spending, such as wages o f staff in charge o f the execution o f these projects. However, the lack o f disaggregated figures o n the investment budget does not provide clear information o n items contained in the investment spending. It i s worth mentioning that this i s a general issue o f the budget framework belonging t o H a i t i (and most o f the LICs). 5.59. The sector intends to adopt a programmatic budgeting approach, with the development of MTEF and budget-program tools. Linking public spending with health outcomes would indeed require having programmatic budget tools t o ensure that budget allocation and execution are in line with the multi-year sectoral objectives. However, g w e n the current status o f budget preparation and execution, adopting a programmatic budgeting approach might n o t be a priority. Indeed, firming-up the basis for budget preparation would b e the utmost priority to build the pre-requisites for programmatic budgeting. 5.60. Similar to most line ministries, budget execution in the health sector suffers from a lack of detailed, accurate data. Also, lack o f clear spending targets makes i t difficult to assew the effectiveness o f budget execution in terms o f achieving the objectives set. The recently developed classification o f poverty-reducing expenditures under 10 subcategories offers an opportunity for the Minstry o f Health t o link budget allocation with poverty reduction items and track their exwwtion throughout the fiscal year. However, t o what extent the classification i s consistent with the secror priorities i s questionable. Moreover, as the classification remains broad, the challenge f o r the Ministry o f Health, with limited capacity, i s to design specific poverty-related budget lines with spending targets and follow-up their execution o n a regular basis. The effectiveness o f a traclung mechanism f o r poverty-reducing expenditures i s conditional o n having detailed information o n the execution o f budget lines. 5.61. W e a k capacity o f the Ministry of Health results in poor expenditure management in the sector. Issues o f transparency, accountability and cumbersome budget procedures adversely impact the effective use o f public resources and, ultimately, their impact o n health outcomes. Additionally, the lack o f adequate resources and weak capacity have prevented the Ministry o f Health from effectively overseeing and regulating the private sector. As a result, the quality o f health services provided i s poor and contributes to l o w levels o f health conditions in Haiti. Harmonizing quality standards and coordinating the health policy and reforms are critical to ensure that the limited resources available in the sector translate to improved health outcomes. 5.62. Human resource management i s also a n issue that impacts expenditure management. L a c k of qualified staff well acquainted with budget procedures limit the effectiveness of budget preparation and execution. F e w staff o f the Ministry o f Health i s devoted to the preparation and execution o f the Ministry budget. Because o f the limited human and technical capacity, there i s n o systematic follow-up o f the execution o f the budget and the Ministry does not produce a regular execution report. Data o n the execution i s scarce and more often outdated, reflecting the lack o f a well-equipped statistical unit within the Ministry. Designing a full-fledge 119 Haiti: Public Expenditure Management and Financial Accountability Review human resource management policy i s therefore an important part o f the required policy actions to improve budget management in the Health sector. 5.63. The weaknesses and challenges emphasized above suggest that a strategy to improve health outcomes goes beyond the increase of resources to the sector and embraces a large scope o f issues that would need to b e addressed to create a nexus and positive dynamic between health expenditures and health outcomes. 5.2.5 Policy Recommendations 5.64. Based o n the findings o f the PER, the following policy recommendations are made and will be further discussed with the Government and the donor community: Increase allocations to the health sector, focussing on delivery of basic health services. Allocations to the health sector should b e increased to enable the execution o f the health strategy. However, the increases in expenditures should reflect the sector spending targets, which the current strategy does not contain. This implies that the first step i s for the Ministry o f Health to complement the current strategy with spending targets and clearly defined budget allocation priorities. I t i s also critical that the links between the sector’s strategic objectives (and expected outcomes) and spending targets are reinforced and progress toward achieving the objectives i s closely monitored. Specifically, spending requirements t o achieve the health MDGs should be evaluated. The second step involves defining a l i s t o f poverty-reducing health items that should b e protected fkom any eventual shortfall in resources. A regular follow-up o f th cution, for instance on a quarterly would be needed. T h i s would increase the eness o f a trackmg mechanism o f poverty-reducing expenditures and ultimately improve the impact o f spending o n health outcomes. 0 Improve the budget preparation and planning process. Increasing the allocations to the health sector would be meaningless unless this i s supported by a sound budget preparation process. T o this end, the Ministry o f Health should elaborate a clear timetable and define the responsibilities o f each actor involved in the preparation o f the budget. Three main issues should be addressed at the preparation stage. First, the Ministry o f Health should clearly define the items that should be considered as investment. This means that the budget line “Program and projects” should b e scrutinized to ensure that items included are truly investment spending. Second, the Ministry budget proposals should b e based o n realistic figures, w h i c h should account for the previous year’s actual spending. This will avoid large discrepancies generally observed between budget proposals and allocations. Third, there i s a need to initiate budget scenarii, accounting f o r resource constraints. A first scenario (low case scenario) could be based o n an eventual shortfall in resources and should discuss the spending priority. A second scenario (a baseline scenario) should be based on actual spending during the previous year. A third scenario (high case scenario) would assume that the sector receive a l l the resources required to achieve its objectives. In this context, the Ministry should carry out an analysis o f needs and financing gaps in the context o f an assessment to achieve the M D G s . Such analysis and assessment could be used to engage in a policy dialogue with the donor community to ensure that donors’ interventions reflect the need’s requirements. This will also help avoid overlapping o f interventions. 120 Haiti: Public Expenditure Management and Financial Accountability Review Set the foundations for programmatic budgeting. Achieving higher health outcomes would require strategic budget planning; in other words, adopting a medium-term budgeting approach. However, given the current status o f budget processes and the limited human capacity within the sector, a shift toward a MTEF might be premature. Rather, an effective sequencing o f such shift would require a focus o n creating o r strengthening the prerequisites o f an MTEF. A first step i s t o reinforce technical capacity to improve resource and spending projections, account for aggregate budget constraints, and track with accuracy donors’ investment flows. A second step i s for the Ministry to be able t o produce a full-fledged sectoral PIP, which should translate the strategic objectives o f the health sector into medium-term expenditure requirements. A third step i s t o design a road map toward the preparation o f an MTEF. Only when these prerequisites are in place, should the sector move to programmatic budgeting. Improve transparency and accountability in the use of public resources. Principles o f sound budgeting and financial management should be fully applied t o the health sector. L i k e in most o f the line ministries, there i s a need for increased transparency and accountability through, for example, the issuing o f regular public expenditure reports for the sector to monitor the execution o f spending. Another measure to improve transparency in budget execution i s for the Ministry o f Health t o reduce the number o f comptes courunts and set a clear timetable for completely phasing out use o f these discretionary accounts. Execute a fiscal sustainability assessment o f health programs. Programs developed in the health sectors should be subject to an analysis o f their fiscal sustainability giver, expected levels o f national and international resources available. l[n particular the capacity o f domestic resources t o support recurrent spending associated with investments in the sector should be systematically assessed. Reinforce donors’ financial commitment in support o f the health sector. Similar to most sectors, predictability o f external support i s essential for the success o f reforms in the health sector. The Government and donors should coordinate medium-term financial commitments in support o f the health strategy and programs. Commitment should be based on the sector PIP and in line with the objectives o f achieving the health MDGs. Efforts should also be made o n the part o f the donor community to put more resources into the budget. This would allow the sector to have a better understanding o f resources allocated to the sector and facilitate the monitoring o f the execution o f expenditures. In the meantime, there i s a need for the sector to develop a mechanism t o track donors’ resources provided outside o f the sector’s budget. Strengthen human resource management and reinforce capacity in budget processes and spending management. T h i s requires that the Ministry develop a comprehensive human resource management strategy, including the possibility o f hinng new staff, with associated spending requirements. The strategy and i t s financial feasibility should be discussed with the MEF. In particular, implications o n the Government wage bill should be evaluated closely with the MEF. 121 Haiti: Public Expenditure Manaqement and Financial Accountability Review CHAPTER 6 : JUSTICE AND SECURITY EXPENDITURE REVIEW 6.1. This chapter reviews the broad trends o f allocation and execution o f public spending in the justice and security sector. The lack o f an integrated sectoral strategy and weak budget and expenditure management processes limits the ability o f this review t o assess the quality o f allocations and spending for the justice and security sector. Therefore, the analysis focuses largely o n the recent trends o f allocations o f expenditure, the structure and execution o f the budget, and questions o f fiscal sustainability o f the projected increase in the police wage bill. 6.2. K e y findings o f the chapter are summarized as follows. First, total yearly allocations to the Justice and Security sector nearly doubled in real terms between FY2002-04 period and FY2005-07, in line with the increase in the Government overall budget. Second, at less than 9 percent o f total FY2006/07 budget, the allocations are, however, relatively l o w when compared t o figures in other post-conflict countries and fragile states. Third, the budget structure o f the Justice and Security sector i s dominated by recurrent expenditure, which accounts on average for more than 70 percent o f total resources allocated to the sector over FY2005-07. Fourth, the intra- sectoral expenditures dominated by allocations t o the National Police, which account for more than 85 percent o f the total allocations t o the Justice and Security sector on average over FY2005-07. Partly as a result, allocations to justice are insufficient t o cover basic functioning o f the courts, which lack basic infrastructure and materials, qualified personnel and modem case management capacities. Fifth, the projected increase in the police wage bill might not be sustainable given the Government’s limited resources. Sixth, the justice and security sector i s expected to benefit from significant donor financing. But, little i s k n o w n o f the sector’s ability t o manage huge flows o f resources effectively (absorptive capacity issue), and in a transparent and accountable way (governance issue). Seventh, weaknesses and issues in management o f public resources in the justicehecurity sector would need t o be addressed. 6.3. Based o n these findings, the chapter recommends: (i) preparing a comprehensive and integrated sector strategy with spending targets; ( ii)improving budget preparation and planning; ( iii) prioritizing spending in line with sector strategy; (iv) re-assessing the fiscal implication o f the police wage bill and i t s feasibility; (v) adopting an integrated approach for preparation o f the investment budget; (vi) implementing budget transparency and accountability principles; (vii) accelerating administrative reforms t o strengthen financial management capacity; and (viii) Ensuringpredictability o f external resources. 6.1 Background and Sector Strategy and Objectives 6.4. I n recent years, Haiti has been plagued by widespread, escalating violence and insecurity. Murder, rape, torture, kidnappings, car hijaclungs, drug trafficlung and money laundering are n o w frequent, if not regular, risks for Haitians, particularly those living in the greater metropolitan area o f Port-au-Prince. In the face o f this escalating violence, the state has at its disposal a police force o f some 8,000 ill-equipped and poorly trained officers.89 According to *’ The army was disbanded in 1995, so the police i s the only national security force responsible for a l l aspects o f internal and external security. Aside official security forces, there are numerous private security companies, with approximately 6,000 armed but largely untrained personnel. The United Nations Stabilization Mission in H a i t i (MINUSTAH) currently maintains a multinational force o f some 7,500 122 Haiti: Public Expenditure Management and Financial Accountability Review Haitian authorities and international assessments, the police force cannot provide security for the majority o f Haitians; the judiciary suffers from inefficiency, corruption and a lack o f independence; and the prison system i s overcrowded, insecure and at r i s k o f collapse. As a result, there i s widespread impunity for crime and violence in Haiti. 6.5. Cognizant of the key role that improved security could play in a growth and poverty reduction strategy, Haitian authorities have placed an emphasis o n the justice and security sector for priority policy actions. Justice and security were central components o f the Interim Cooperation Framework (2004-06) and the I-PRSP (2006) identifies justice and security as one o f the key sectors o f the government’s growth and poverty reduction strategy. The Minister o f Finance re-iterated this message during the IDB’s 2007 annual meetings and identified modernization o f the police forces, building prisons and training judges as government priorities for the allocation o f budget resources becoming available through debt relief initiatives. 6.6. Plans for the reform o f the police, justice and prisons provide some important elements of the Government’s strategy for the sector. The National Police Reform Plan i s the most fully developed o f the three institutional strategies and it has been published as an official document o f the UN Security Council.go I t presents overall objectives, specific activities and sequencing for reform and development o f the Haitian National Police (HNP) with an estimated cost o f about US$700M over five years (2006-11). The Plan anticipates that the national budget will be able to support projected staffing increases and basic operating costs, but that capital investments will rely heavily on external donor financing. At a November 2006 international dbnors conference in Madrid, the Minister o f Justice and Public Security outlined the government’s priorities for short, medium and long-term justice reforms (2006-201 l), including institutional reform, increasing access to justice, strengthening the independence o f the judiciary, fighting corruption and impunity, and criminal justice reform. The prison administration, which i s a Department o f the HNP, has developed project proposals for increasing the security o f correctional facilities and improving the conditions o f detention, as well as for the professionalization o f the prison administration. 6.7. However, achieving the Government’s reform objectives will require the development o f an integrated strategy for the justice and security sector with clearly defined objectives and indicators linked to the budget process. I t will also require sufficient allocation o f domestic and external resources t o implement the strategy, efficiency o f intra-sectoral allocations and expenditures, as well as effective and transparent management o f expenditures. The lack o f an integrated sectoral strategy and weak budget and expenditure management processes limits the ability o f this review to assess the quality o f allocations and spending for the justice and security sector. It will therefore focus largely on the structure o f the budget, the eficiency o f expenditures, and questions o f fiscal sustainability. military peacekeepers and 1,500 civilian police with a dual mandate to provide security and to support national police reform. 90 UN Doc. S120061726 (12 Sept. 2006) 123 Haiti: Public Expenditure Manaqement and Financial Accountability Review Box 9. Main Features o f the Justice and Security Sector Objectives and sector mandate. T h e core objective in the justice and security sector i s to ensure and protect the fundamental rights and freedoms o f Haitian citizens. The immediate and urgent task though, is simply to restore basic security and access to justice. Sector structure. The main state institutions o f the justice and security sector are (i) a four-tiered court system, ( ii) a police force and ( iii) a prison system. These all fall under the administrative responsibility o f the Minister o f Justice and Public Security (MJSP). Significant international support was provided during the 1990s to hire and train personnel, provide equipment and infrastructure, particularly for the police. However, much o f the progress achieved in the 1990s was stalled and even reversed during the subsequent period o f political crisis and insecurity. Maintenance was not what it should have been and much infrastructure and equipment was severely damaged during the 2004 events. Today, the sector’s key institutions continue to b e in crisis. The justice system. Haiti’s justice system i s modeled largely o n French civil law in both i t s judicial structure and i t s system o f codified law. The penal and cnminal procedure codes are largely unchanged from the Napoleonic models adopted in the early 1800s resulting in serious inefficiency in the administration o f justice. A recent report on pre-trial detention placed the conviction rate in criminal cases at 3 percent, while 80 percent o f all files referred to prosecutors are dismissed. The judiciary i s organized around 16 geographic jurisdictions. Each jurisdiction has a Court o f First Instance that represents the judicial branch o f power, and a public prosecutor representing the executive branch o f power. At the lowest level o f the system, there are Justices o f the Peace (iuges de paix) with junsdiction over minor civil and criminal matters in each o f the country’s 140 communes, the smallest administrative division. There are five regional Appeals Courts and the Supreme Court (Cow de Cassation) i s the final highest judicial authority. The system also comprises a few special courts o f law. The police force. A Director General leads the HNP and oversees a nlmber o f central and territorial directorates. Each o f Haiti’s nine administrative Departments has a director, and there are 200 commissariats headed by police chiefs (commnsaires) in larger centers, with sergeants (inspecteurs) in smaller towns and communes. At the central level there are separate administrative police (the main police corps) and judicial police (an investigative unit linked #tothe judiciary), an Inspector General, and a number o f specialized units. The Inspector general i s responsible for internal financial accountability and control within the HN”. H e i s also currently the Director o f Central Administration and General Services responsible for the preparation and execution o f the annual HNP budget as well as personnel and asset management. The Haitian National Police had a staff o f close to 8,000 people at the end o f 2006. Part o f t h i s workforce consists o f administrative and support personnel. Theprison system. The prison administration i s under the Department o f a Haitian National Police and thus reports to the HNP Director General. Besides its Central ofices and facilities in Port-au-Prince, the prison administration has four Regional Directorates (North, South, West and Artibonite). At the lower level, all penitentiary facilities report to their local Regional Directorate. There are currently 17 functional prison facilities in Haiti. The Prison administration reports 533 correchons officers and some 600 administrative and support staff in the system. Some 300 persons are actually in training at the Police Academy in order to j o i n the correction officer force later this year. 124 Haiti: Public Expenditure Manaqement and Financial Accountability Review 6.2 T r e n d s and S t r u c t u r e o f Justice and S e c u r i t y Sector E x p e n d i t u r e s 6.2.1 T r e n d s in A l l o c a t i o n s 6.8. T o t a l yearly allocations t o the Justice and Security sector nearly Table 57. Allocations t o Justice and Security Sector doubled in r e a l terms between FY2002-04 p e r i o d and FY2005-07: f r o m an average o f 1.5 b i l l i o n gourdes t o about 2.7 b i l l i o n gourdes. This i s in line with the increase in the overall NA 2,334.7 Government budget over the same period. Including external aid to projects, Total Recurrent Budget 1,413.6 2,041.5 as well as debt service outlays, the justice and security 2006107 budget represents 2.9 percent o f the projected GDP for the Same year, which corresponds t o an expected outlay o f budget Total Nation Budget 2,1264.7 3,2460.3 U S $ 14.6 per capita. Shares o f Justice and Securitv in 7.2 8.3 Nation Budget (in YO 6.9. *lthough levels Of spending !Sources: L e M o d u r . Journal Officiel de l a RCDubliaue might appear ] I w ~ i m p o r t a n t than the d’H&i and Staff Calculations. effectiveness o f spending in relation t o justice and security objectives, a t less than 9 percent o f total FY2006/07 budget; -the allocations can b e considered as relatively l o w since secyrity and justice often account for a much larger share o f between 20-30 percent o f ‘public expenditures, especially in other post. cq&ct countries and fragile states. F o r example, the security sector constituted 33 percent o f total core and external budget expenditures in Afghanistan in 2005106 while in the Central A h c a Republic, security sector expenditures constituted 28 percent o f public expenditures for that same year.g’ 6.10. A budget structure dominated by recurrent expenditures. T h e budget structure o f the Justice and Security sector i s dominated by recurrent expenditures, w h i c h accounts on average f o r m o r e than 70 percent o f t o t a l resources allocated t o the sector over FY2005/07. Investment budgets account for slightly above 25 percent over FY2005-07. 9’It i s however w o r t h noting that unlike Haiti, b o t h these countries retain armed forces in addition to police in response t o internal and regional instability. 125 Haiti: Public Expenditure Management and Financial Accountability Review 6.11. Recurrent budget has also been on a rising trend since FY2001/02. I t increased by Table 58. Recurrent Budget Allocations to more than 40 percent in real terms o n average JusticeLSecurity Sector (In real HTG between the period FY2002-04 and the period FY2005-07 (see Table 58). As a result, the share o f recurrent budget allocated to the Justice and Security sector increased from an average o f 10.5 percent in FY2002-04 up t o 17 percent o f total recurrent expenditure (excluding interest payment) in FY2004-06 (see Table 58). This reflects the new authorities’ efforts to Sources: Le Moniteur,Journal Officiel de la allocate m u c h needed resources t o the Justice RCpublique d’Haiti and Staff Calculations. and Security sector, with a view to improve security, a pre-condition for a stable environment conducive to growth and favorable t o the fight against poverty. At the same time, the investment budget increased significantly in FY2006/07 and accounts for more than 46 percent o f total allocations t o the sector. However, high yearly fluctuations o f the investment budget, with one year o f strong increase followed by another year o f sharp decline, illustrate the difficulty for this sector to conduct a sound, sustainable and predictable investment policy. However, this i s not a specific issue to the justice and security sector. I t reflects a more general concern o f high dependency o f nvestment budget o n volatile donors’ aid flows as discussed in Chapter 2. However, the more acute in the Justice and Security sector because o f the central role that this sector Haiti’s development strategy. Long-term growth and poverty reduction cannot be achieved without significant improvement in security. The current short-term strategy o f increasing the size of the police force has helped improve security (quick win), the challenge i s n o w to develop a medium to long-term investment strategy to ensure a long-lasting security (long-term win). 6.2.2 Intra-Sectoral Allocations o f Expenditures 6.12. Intra-sectoral expenditures are dominated by allocations to the National Police. Because o f lack o f data, the analysis o f intra-sectoral allocation o f expenditure i s limited to FY2005-07 period. M o r e than 85 percent o f the total allocations (or more than 90 percent o f total recurrent allocations) to Justice and Security sector o n average over FY2005-07 go to the Haitian National Police (HNP) in line with the authorities’ efforts to provide adequate resources t o HNP to help ensure i t s mandate o f restoring security, particularly in Port-au-Prince. Recurrent allocations t o the HNP (including prisons department) increased by more than 8 percent in real terms between FY2005/06 and FY2006/07 (see Table 59). As a result, little allocation has been provided t o the daily functioning o f the MJPS. 126 Haiti: Public Expenditure Management and Financial Accountability Review Table 59. Evolution of the intra-sectoral allocation o f recurrent expenditures FY2005-(In real HTG million - excluding aid project, unless otherwise indicated) b’Y2004105 k’Y2OOSIO6 E’Y2006107 Average Shares in % FY2005- of Total 07 recurrent expenditures Ninister’s Office 11.5 8.8 10.8 10.4 0.5 Internal Services 79.3 81 0 83.8 81.4 4 HNP 1,927.6 1,828.7 1,993.0 1,916.5 93 8 UCREF 18 4 19 8 17 9 18.7 09 Public Security Secretariat 5.9 5.5 7.7 6.4 0.3 Identification Office NA 9.5 86 NA 0.5 Justice Secretariat NA NA 6.4 NA 04 Total for MJPS 2,042 9 1,953 3 2,128.3 2,041.5 100 Source: Le Moniteur, Journal Offciel de la Rtpublique d’HaYti and Staff Calculations. 6.13. Insufficient allocations to meet the basic functioning needs. Accounting for about 4 percent of total allocations on average over FY2005-07, allocation to internal services o f the M J S P (100 million gourdes or U S $2.8 million) are not enough to ensure a good quality o f services. Staff o f the MJPS thus lacks the basic material to perform their routine work activity. More troubling i s the l o w level o f resources provided to the Anti-Money Laundering Unit (UCREF), which does not allow the unit to perform i t s planned activities. T h e same applies to the Public Security Secretariat, the Identification Office (responsible for the c i v i l registry) and the Justice Secretariat. Because o f the resource constraints and the pressing need to sh-engthen the HNP, these entities have not been pnonties for allocation o f the MJPS budget. 6.14. Budget albcations to justice are iaasufficient to cover basic functioning of the casmrls, which lack basic infrastructure and materials, qualified personnel, and modern case management capacities. O f the total resources allocated to Justice and Security sector for FY 2004-06, less than 15 percent go to the administration o f justice. According to both Haitian judicial authorities and international experts, this i s insufficient to cover the basic functioning o f the courts and, as a result, negatively impacts the rule o f law in Haiti. 6.15. A difficult trade-off issue in the face of limited resources and emergency security needs. The current structure o f the justice and security budget skewed by allocation to Haitian National Police i s dictated by the limited overall budget allocated to the sector (a macroeconomic issue) rather than a deliberate policy choice by the sector (a microeconomic and sectoral issue). Allocating more to the other entities (Public Security Secretariat, Identification Office, and UCREF) would mean allocating less to the police: a difficult trade-off that the sector could not respond to, given the current status o f emergency security needs. In fact, restoring a minimum level o f security in the country has meant providing sufficient resources to the police force to perform routine operations. 6.16. I n addition to limited allocations of resources, the M J S P and judicial authorities suffer f r o m weak financial planning and management capacities, while internal accountability and oversight mechanisms are largely non-existent. As a result, it has not been possible to gather detailed information regarding budget planning and execution for the judiciary. In order to achieve objectives outlined in the government’s justice reform plans, there i s a clear need to strengthen capacities for financial planning and management in the administration o f justice. Proposed legislation on the status o f the judiciary foresees increased autonomy in the 127 Haiti: Public Expenditure Management and Financial Accountability Review administration and financial management o f the courts, while the MJSP would retain financial responsibility before parliament and the supreme audit institution (Cours des Comptes et Contentieux Administrative). It i s thus important to proceed with the establishment o f a Judicial Audit Unit within the MJSP responsible for verification and oversight o f the administration and financial management o f the court system. Another reform that will contribute to improved transparency and accountability o f the judiciary would be the standardization and official publication o f judicial tariffs (fees for service) for which preparatory work has already been undertaken by the MJSP. 6.17. I n order to meet minimum prison capacity requirements, significant investment in infrastructures will b e needed. The prison system in H a i t i has a capacity for approximately 2,000 prisoners, but currently houses more than 4,800 detainees, the great majority o f whom are in pre- trial detention. By international standards, this rate o f incarceration as a percentage o f population i s s t i l l extremely low.’’ Thus prison overcrowding i s due t o a lack o f capacity rather than from high rates o f incarceration. Before the events o f 2004, there were 21 prisons in operation, albeit in poor condition and all badly overcrowded (less than 1 square meter per inmate). During the 2004 events, there were massive escapes and five prisons were totally destroyed while five others were severely damaged. Since then, some prisons have been rehabilitated, so there are n o w 17 functional facilities. However, most prison facilities are former military barracks that lack health and sanitation facilities and adequate security. Thus the prison system faces serious capacity problems that will require significant capital investment. 6.2.3 Economic composition o f expenditure b 6.18. Over FY2002/2007, a five-year period, ’ salaries have represented on average more than T a b l e 68. Evolution of the cornpasition o f ’ expenditure (in % of total exp., 60 percent o f expenditures (see Table 60). I t i s excl. aid project), F1’2001/06 worth noting the extent to which locally financed 1 1 Averages capital expenditures have fallen to what can only be FY 02/04 F\’05/07 described as an insignificant portion o f expenditures in the sector. However, efforts are being made by the authorities to increase government’s contribution Salaries Goodsandservices Capital expenditures :4 :!ii Total expenditures100.0 100.0 to finance capital expenditures. Sources: Government Data and Staff calculations. 6.19. Personnel expenditures. Based o n the budgeted figures for fiscal year 2006/07, Security and Justice accounts for some 12,000 employees (including the personnel to be hired during the year), o f the roughly 40,000 total c i v i l service force. Personnel in the security and justice sector thus account for 3 1 percent o f all government’s employees and salaries o f the sector represent 26.5 percent o f the government’s payroll. According to figures appearing in the current budget documents, the average salary would stand at US$4,862 per year o n the side o f the Ministry o f Justice and Public Security while the average salary for the whole c i v i l service force i s U S $ 5,558 per year. A review o f salary scales and levels should be a priority for the justice and security sector. 6.20. Capital expenditures. All capital expenditures financed through local resources, both for the ministries’ equipment and the investment projects, amounted t o slightly more than 40 m i l l i o n gourdes (US$ 1 million) in FY2005/06. Yet, they are projected to reach US$18 m i l l i o n in 2006/07. 92 Incarceration rate in Haiti o f approx. 40: 100,000 compares to rates in Canada of approx. 100:100,000 and the U S o f 600: 100.000. 128 Haiti: Public Expenditure Management and Financial Accountability Review I t thus follows that capital expenditures, including capacity building and training, will b e financed f o r the most part by the donor community. Available figures suggest these foreign financed capital expenditures that amounted to US$45 m i l l i o n in 2005/06 as disbursements o f the fiscal 2006/07 are projected to reach US$52.4 million. M o s t o f the contracts financed by foreign a i d are issued and managed directly by the donors themselves. In those circumstances, figures pertaining t o external aid disbursements are difficult t o collect and estimate, even more so f o r Haitian government officials than for donor missions. The current level o f capital expenditures p a i d out o f foreign aid appears t o b e about US$50 million. 129 Haiti: Public Expenditure Manaqement and Financial Accountability Review 6.21. Goods and services. As a whole, the Justice and Security Table 61. Justice and Public Security. Credits and sector was granted a 1.1 billion expenditures for selected goods and services gourdes (US$ 26 million) budget FY2002-FY2007 in real HTG thousand - allocation for recurrent goods and Excluding aid project services for the 2006/07 fiscal year. Averages Most o f these credits (98 percent) are E’Y02101 FY05107 destined t o the MJPS. Again, the police A. Budgets department obtains by far the largest Traveling expenditures 6 370 6 602 share o f credits voted, that i s some 966 Renting o f building 2 967 8 820 million gourdes (US$23 million). The Renting o f vehicles 1 834 5 291 ratio o f goods and services to salaries Maintenance o f office equipment 1990 1435 stands at 46.5 percent in the 2006/07 Maintenance o f vehicles 5 156 3 961 budget, which i s quite close to what i t Maintenance o f building 13 013 15 915 was in 2005/06, as well as the year Office supplies 22 221 18 658 before (2004/05), in terms o f actual Fuel 54 819 69 056 salaries paid and actual expenditures o n Food 58 224 250 876 good and service. As a matter o f fact, Office material & equipment 49 187 24 307 this ratio i s much lower on the side of Transport equipment 16 471 3 283 the courts and tribunals. Given the B. Expenditures actual state o f destitution and the Traveling expenditures 1052 992 general lack o f means in the judiciary, Renting o f building I 528 3 023 i t would seem that the allocation for Renting o f v e h d e s 899 292 ds and services i s well Maintenance o f office equipment 315 292 quired level needed to Maintenance o f vehicles 2 382 2611 ice system functions. Maintenance o f building 6 119 6 515 Office supplies 16 641 12 408 6.22. Food and fuel are the most Fuel 54 273 65 542 significant goods and services Food 57 643 246 662 purchased by the MJSP, respectively Office material & equipment 6 820 7 891 accounting for 44 percent and close Transport equipment 12 836 484 to 60 percent o f the expenditures for C. Execution rates (%) goods and services in FY2004/05 and Traveling expenditures 17 14.8 FY2005/06. The HNP provides a daily Renting o f building 32.1 35.1 meal to its officers and the prison Renting o f vehicles 48.7 5.5 administration provides food for Maintenance o f office equipment 16.4 34.9 inmates. Furthermore, expenditures on Maintenance o f vehicles 47.2 65.5 these two main items show, by far, the Maintenance o f building 47.4 40.4 highest rates o f execution, which i s Office supplies 99.4 66.3 indeed close to 100 percent (see Table Fuel 98.9 94.7 61). T h i s constitutes a very strong Food 99 98.3 performance given the fact that the Office material & equipment 13.1 32.6 budget for fuel doubled in j u s t two _ _ Transport equipment 79.3 14 - ..4 years, and the food budget increased ources: Government Data and Staff calculations. ten-fold during the same short period o f time. All other categories o f goods and services showed quite l o w rates o f execution and in many instances varied significantly over the years. 130 Haiti: Public Expenditure Manaqement and Financial Accountability Review 6.2.4 Projected T r e n d s in Police W a g e Bill and Fiscal Sustainability Issue 6.23. The wage bill for police i s set to increase significantly over the next five years as the HNP intends to hire some 1,500 new officers each year, in order to reach a force o f 14,000 officers in 201 1. T h e National Police Reform Plan’s target i s t o reach a force o f some 20,000 officers - identified as the number o f police and other security officers who would be required to cover the full range o f security needs in Haiti and thus bring the ratio o f police to inhabitants close to 1:500.93 As a result, staff estimates that the total wage bill would almost double within a 3-year period, from US$48.8 million in FY2006/07 to US$85.3 million in FY2009/10.(see Table 62). In the meantime, the Government wage bill i s expected to increase from US$202 million to US$270 million in FY2008/09 (see Table 63). This means that the justice and security wage bill could account for 25 to 30 percent o f total Government wage bill. 5,899 6,194 , Sources: Government Data and Staff projections. Note: All expenditures are expressed in constant 2006107 pnces. The Justice and Public Secunty sector includes the Department of Pnson Administration. The work force i s taken from budget documents. I t 1s implicitly assumed here that new employees are hired on the first day o f the year. For 2006/07, that year’s hiring are already included in the salary budget. The same i s true for salary increases awarded in that same budget. I t i s also assumed that hinng 1,500 new employees a year translates into a net addition of 1,250 employees to the force because of vetting, retirements and deaths. 6.24. The extent to which the Government could afford this larger wage bill i s a matter of concern, and raises three issues. First, the Government’s current revenue i s projected to increase slowly to reach less than US$650 million by FY2008/09 (i.e. 10.9 percent o f GDP) (see Table 63). In other words, more than 40 percent o f Government revenue would be granted to support the justice and security wage bill. Second, perhaps a downward adjustment o f expenditure in goods and services could provide some room o f maneuver for increasing the wage bill. However, spending on goods and services are also projected to increase. Third, recourse to donors’ aid flows could help the Government afford the projected increase in the justice and security wage bill. Unfortunately, grants are expected to decline from US$291 million in FY2006/07 to US$240 million in FY2008/09. T h e issue here i s the financial sustainability/feasibility o f an inflated wage bill o f the Justice/security sector given the Government’s limited room o f maneuver. Dealing with this issue would require both the Government and the donor community to redesign the National Police strategy, including setting the objectives in line with expected increases in domestic revenue and aid flows. 93At i t s current force level o f 8,000, the HNP is one o f the most numerically weak police forces in the world with 100 officers per 100,000 citizens. This compares, for example t o 285 per 100,000 in the LAC region. 131 Haiti: Public Expenditure Management and Financial Accountability Review Table 63. Proiected Justice/Security Salaries Expenditure and Total Government Salaries, FY2007-09 (in i i l l i o n of Constant USS) FY2006i07 FY 2007108 FY 2008109 f Total Justiceisecurity wage bill ( 48.8 60.4 73.8 I Total Government wage bill 202.3 240.6 270.4 Shares o f wage bill in total 24.1 25.1 27.3 Government wage b i l l (in %) Goods and services 26.8 30.4 33.5 Total Revenue and Grants 813.8 825.8 889.5 O.W. Total current revenue 522.5 589.7 647.9 Total grants 291.3 236.1 241.6 Sources; Government and IMF Data, and Staff projections. 6.2.5 Budget Execution Trends 6.25. Overall, the rates of execution have been declining over the period under review, but remain relatively high. Table 64 gives a more precise and detailed view o f intra-allocations for the credits and expenditures o f the MJPS over the years from FY2003/04 up to FY2005/06 for allocations and expenditures. The execution rate stood at nearly 89 percent during the first year and then fell suddenly t o 80 percent during FY2004/05, and remained at that level the following year. 6.26. As indicated in Chapter 1, falling execution rates might result from three factors: (i) cash rationing decision; or ( ii)poor budget planning; and (iii) limited absorptive capacity. The Justice and security sector has not experienced a sudden cut in expenditures during the fiscal years, which might have prevented it from fully executing i t s planned spending. In fact, the fall o f the execution rate i s essentially due to the poor planning o f budget needs and the lirnited absorption capacity of the Haitian police force. This perfomance might not be surprising given that the Police . budget more than doubled between FY2003104'and FY2004105. Indeed, it i s rather surprising in that particular context that the execution rate remained as high as 80 percent. 132 Haiti: Public Expenditure Manasement and Financial Accountability Review Table 64. Intra-allocation of credits and expenditures, and execution (In real HTG million), FY2004/06 A. Budget allocations Minister’s Office 8,4 11,5 8,8 9,6 Internal Services 62,4 79,3 81,O 74,2 HNP 1166,2 1927,6 1828,7 1640,9 UCREF 13,9 18,4 19,8 17,4 Public Security Secretariat NA 5,9 5,5 5,7 Identification Office NA NA 9,5 9,5 Justice Secretariat NA NA NA NA Total for M J P S 1250,9 2042,9 1953,3 1749,O B. Expenditures Minister’s Office 7,7 10,l 5,6 73 Internal Services 59,8 68,l 52,O 60,O HNP 1041,l 1539,9 1490,2 1357,O UCREF 0,3 18,4 12,9 10,5 Public Security Secretariat NA 1,3 2,8 2,O Identification Office NA NA 3,7 3,7 Justice Secretariat NA NA NA NA Total for M J P S 1109,O 1637,7 1567,i 1438,O C. Execution Rates (%) Minister‘s Office 93.7 87.4 64.4 81.1 Internal Services 95.8 85.8 64.2 79.3 HNP 893 79.9 81.5 82.4 UGREF 2.3 100.0 64.4 62.8 Public Security Secretariat NA 21.5 50.7 36.6 Identification Office NA NA 39.0 NA Justice Secretariat NA NA NA NA Total for M J P S 88.7 80.2 79.1 82.0 ource: L e Moniteur, Journal Officiel de la RBpublique d’HaTti and Staff Calculations. 6.3 Public Investment Programs in the Justice and Security Sector 6.27. Figures pertaining to investment programs in the justice and security sector are incomplete, in most cases preliminary, and can be confusing at times. For consistency purposes, one single source o f data was used: the PIP prepared by the Ministry o f Planning and External Cooperation (MPCE). 6.28. Table A.5.8 in Annex shows that investment programs in the justice and security sector have been marginal. Resources devoted to the sector’s investment programs (including foreign aid) over the past two fiscal years are the lowest allocations to b e selected priority sectors. Accounting for about 3.4 percent o f total PIP on average over FY2004-07 (see Table 65), they represent about 17 percent o f total allocations to investment programs in the agriculture sector and only 2 percent o f allocations for the transport sector. 133 Haiti: Public Expenditure Management and Financial Accountability Review 6.29. Also, according to MPCE data, donors’ financial involvement in the justice and security sector Table65. Resources for PIP in Justice and Security PIP has been relatively marginal. Sector. Average over FY2004/05-FY2006-07 Only about 25 percent of total PIP allocations on average over the past two years were financed b y donor resources. This compares poorly with the other priority sectors where more than 90 percent (agriculture and transportssectors), and more than 70 Sources: Ministry of Planning and External Cooperationand percent (education and health sectors) Staff Calculations. o f total PIP allocations originated fi-om foreign aid o n average over the same period. About 75 percent o f total PIP allocations o f the justice and security sector o n average over the past two fiscal years were financed by domestic resources. Actually 100 percent o f the sector investment program was supported by the Government’s o w n resources in FY2005/06. More troubling i s the fact that the allocations (thus the share in total PIP) t o the sector PIP declined in FY2005/06 as a result o f a reduction in domestic resource allocations and a zero flow o f foreign aid. 6.30. However, this picture i s expected to change dramatically in FY2006/07. The justice and security sector i s expected to benefit f r o m significant donor financing (see Table A.5.8 in Annex). The amount o f foreign aid gwen directly to projects in the justice and security sector i s projected to amount to roughly US$50 million in FY2006/07. T h i s signals donors’ strong commitment t o support the security sector, a cornerstone o f Haiti’s growth and poverty reduction strategy. But three issues are worth emphasizing. First, the projected increase in external financing comes from a relatively l o w basis. Second, the extent t o which these resources are backed-up by w e l l designed sectoral projects and programs i s unknown given that the sector does not have a full- fledged integrated strategy and financing plan. Third, it i s true that, ensuring lasting security and improving justice in the medium to long-term will require significant investment in the sector. It i s also true that Haiti’s scarce resources cannot afford the required investments; donors have t o back- up the Government’s efforts. But at the same time, little i s k n o w n o n the sector’s ability to manage huge flows o f resources effectively, and in a transparent and accountable way (governance issue). 6.4 Weaknesses, Challenges and Recommendations 6.4.1 Weaknesses and Challenges 6.31. The security and justice sector has suffered f r o m decades of lack of vision and resources, mismanagement, corruption and political interference. The budget information i s scarce and unreliable, informal practices override rules and procedures, and financial reports are Oversight practically n ~ n - e x i s t e n t . ~ ~ and accountability mechanism are weak or non-functional. Improving sector management will be challenging given the deep-seated informal practices and the lack o f adequate resources. 94 The publication o f a detailed Financial Report for the HNP (2005-2006) i s a notable exception that will hopefully become standard practice. 134 Haiti: Public Expenditure Management and Financial Accountability Review 6.32. Budgetary processes are far f r o m being effective. Information needed for programming and budgeting i s scarce and often unreliable. Because o f a lack o f precise and reliable information o n the actual stocks and assets, and in the absence o f unit prices and technical indicators, the MJPS, like most other departments, basically increases budgetary allocations received the previous year by some given percentage instead o f going through the process o f defining and measuring needs and then estimating their costs. I t i s worth noting that this reflects the general budgetary process, which i s to a large extent a more or less top down mechanical exercise instead o f being an informed bottom up exercise under constraint. Most managers in the sector consider the budget as being imposed by the finance and planning ministries. This might explain why there are only partial cost estimates available from the reorganization and the reconstruction o f the security and justice services and infrastructures. Thus, improving budgetary processes at the level o f MJPS involves addressing weaknesses at the central level o f the central government’s budgetary procedures. 6.33. Expenditure procedures are cumbersome. L i k e in other ministries, capital expenditures are budgeted and monitored in two different accounts, depending on whether they are intended to equip the ministries themselves or if they are destined to finance, or co-finance, projects and programs. Furthermore, those expenditures for small materials and office equipment as w e l l as immaterial expenditures are in some tables and documents classified as recurrent expenditures. The execution o f spending i s affected by the use o f current accounts. As in other departments, the MJPS maintains a certain number o f current accounts at the Treasury in order to facilitate and accelerate the payment o f recurrent as w e l l as capital goods and services. The police department operates four such accounts, the prison administration keeps one o f them open. The number and the value o f current accounts maintained by the judiciary administration as well as by each tier o f courts, at the central as well as the local levels, are unknown. These accounts that operate more or less as cash advance accounts often help process payments o f contracts overlapping two consecutive fiscal years. However, the lack o f transparency and effectiveness in their execution complicate the expenditure process. 6.34. The MJSP i s characterized b y a lack of transparency and accountability regarding expenditures. I t i s virtually impossible to get from the Ministry any information on actual expenditures managed and recorded in the justice sub-sector. B o t h the Police and Prison administrations keep records o f credits and expenditures pertaining to the recent fiscal years. In recent years, HNP has made some improvement regarding budget preparation and expenditure monitoring including the production o f annual expenditure reports. In general, there i s a need t o enhance the capacity o f financial managers and o f oversight mechanisms (both internal and external) t o improve the financial management in the justice and security sector. 6.35. While there are separate plans to reform police, justice and prisons, the sector does not have an integrated sector strategy with a clear implementation plan identifying domestic and external resource requirements. As a result, budget proposals are not backed-up by a sectoral strategy and actions plans. In addition, spending targets do not exist. 6.36. The government has clearly recognized the restoration o f security and the rule o f law as key priorities for achieving growth and poverty reduction in Haiti. However, the system developed for classifying poverty-related expenditures in the FY2006/07 budget identifies spending on justice and security only indirectly (in relation to investments for infrastructures and education). It does not identify recurrent costs o f providing justice and security per se as contributing to the reduction o f poverty. 135 Haiti: Public Expenditure Management and Financial Accountability Review 6.37. While external assistance to the investment budget will increase, investments might not be sustainable if insufficient attention i s given to the development of a sector financial management capacity and to the fiscal sustainability of recurrent costs (especially salaries and asset maintenance). Investment levels and recurrent budgets have t o be consistent, coherent and thus synchronized. Unfortunately, current (and planned) investment programs do not factor in the recurrent costs associated with those investments. Given the scarcity o f resources, this raises the question o f the sustainability o f investments injustice and security sector. 6.38. Human resource management and asset maintenance i s very weak across the sector. While it was possible to obtain some information regarding numbers, pay scales and grades of officers in the police and prisons services, i t was not possible t o obtain similar information regarding judges, prosecutors, clerks and other administrators in the judiciary. There i s also little information available regarding the management o f sector assets, especially security equipment and infrastructures. In order to ensure sustainability o f investments in the police, judiciary and prisons, there i s a need to establish systems for the management o f these assets. 6.4.2 M a i n Policy Recommendations 6.39. Based o n the findings o f the PER, the following policy recommendations are made and will be further discussed with the Government and the donor community: Prepare a comprehensive and integrated sector strategy with spending targets. The justice and security sector should develop a full-fledged integrated, coherent, and forward- - loolung strategy, which encompasses all institutional aspects o f the sector. Such a strategy should be fully costed and comprise o f specific objectives, inputs, outputs and outcome indicators. This strategy should be translated into operational policies and programs with budgets that identify internal and external resources-available and needed. These programs should be subject to an analysis o f their fiscal sustainability given expected levels o f national and international resources available. The strategy should also include associated spending targets t o enable the evaluation and monitoring o f progress achieved toward achieving the objectives and meeting the outcome indicators. The strategy should elaborate an investment program and the needs should be costed. Human resources needs to implement the plan, which should also be costed and the sustainability o f associated costs should be discussed with the donor community. HNP action plan could serve as a basis for designing such a strategy. This medium-term sectoral strategy should be integrated within the National Strategy for Growth and Poverty Reduction currently being prepared. More importantly, the preparation o f the PRSP should offer the opportunity to revise the classification o f poverty-related expenditures in the justice and security sector. 0 Improve budget preparation and planning. Budget preparation and planning needs to be improved. This requires that three elements ensue. First, the Justice and Security sector should produce a clear timetable for preparation o f i t s budget, with clearly defined responsibilities for actors involved in the preparation process. Second, the budget should be linked to the sector’s full-fledged strategy (as envisaged above) and translates the strategy spending targets into yearly budgetary proposals. This would avoid budgetary envelops becoming a renewal o f past year allocations. In addition, the proposals need to be discussed with the Ministry o f Economy and Finance before the launch o f budgetary conferences. This would allow the sector sufficient time to revise its budgetary proposals if need be and submit a realistic budget proposal accounting for the overall budgetary constraints. Third, budgetary planning should be reinforced by reinforcing the technical 136 Haiti: Public Expenditure Manaqement and Financial Accountability Review capacity o f staff of the Ministry in charge o f preparing the budget. Moreover, i t will be important to strengthen the planning capacity in each institution as well as develop the MJSP’s integrated sectoral planning capacity. 0 Prioritize spending in line with sector strategy. Prioritization o f spending in the justice and security sector should be done in the context o f the overall sector strategy and sub- sectoral action plans (police, justice, prisons). Because o f the limited resources, the budgetary trade-offs would need to favor policy actions that consolidate the gains achieved so far in stabilizing the security situation (quick wins). However, it will also be important t o evaluate whether the current prioritization o f the police in intra-sectoral budget allocation has left other institutions without adequate resources t o ensure their basic functioning. In particular, lessons learned f r o m efforts to reform the criminal justice system during the 1990s suggest that while significant external resources were invested, results have been largely lost, in part due t o a lack o f sequencing, prioritization and integration o f reforms and investments across the police, justice and prison institutions. 0 Re-assess the fiscal implication o f the police wage bill and its feasibility. The issue o f an inflated police wage bill should be re-evaluated. As a first step, there i s a need for a thorough assessment o f the projected wage bill and its budgetary implications. Carrying out a fiscal sustainability exercise o f the projected wage bill should provide clarity o n the feasibility o f the police reform. T h i s would require that the Ministry o f Justice and Public Security adopt an integrated approach that should involve the MEF, the Ministry o f planning, and the donor community. 0 Adopt an integrated approach for the preparation o f the investment budget. The investment budget should be prepared jointly with the Ministry o f Planning and involve the donor community at the very outset o f the preparatiopi process. Because o f the central role o f the Justice/Security sector in Haiti’s development agenda, securing donor funding to execute the sector investment program i s critical for a long-lasting security in Haiti. The sector PIP should derive from the overall sector strategy, and implemented through a rolling MTEF so that expenditure targets and sector projected outcomes can be revised according t o resources expected. Inthis perspective, a first priority i s to createheinforce the pre-requisites for a medium-tern budgeting approach, including better budget planning, improved preparation o f annual budget, and spending linked to sector objectives. 0 Implement budget transparency and accountability principles. Principles o f sound budgeting and financial management should be fully applied to the justice and security sector. While discretion might be appropriate in the discussion o f some sensitive security matters, the management o f public resources in this sector should be subject t o the same oversight and accountability requirements as a l l other sectors. In particular, there i s a need for increased transparency and accountability through, for example, the issuing o f regular public financial reports for police, judiciary and prisons. 0 Accelerate administrative reforms to strengthen financial management capacity. Priority administrative reforms must be undertaken t o improve justice and security sector management and oversight capacity. I n particular, for broader institutional reform to succeed, i t will b e essential t o develop solid financial management capacity and practices within the Ministry o f Justice and Public Security, as well as within the institutions i t oversees. The strengthening o f internal audit functions, human resource management and the capacity to manage physical assets in the judiciary, police and prisons, and ensuring that these are linked into central PFM systems are specific areas requiring attention. 137 Haiti: Public Expenditure Management and Financial Accountability Review 0 Ensure predictability o f external resources. Significant and predictable external resources will b e required to attain police, justice and prison reform objectives. Allocations f r o m the Haitian treasury will not b e sufficient to cover investments required to achieve the objectives o f reform o f the police, justice and prisons. In particular, external financing i s required for the training and equipping o f a national police force, rehabilitation and construction o f prison facilities, and modernization o f the judiciary. W h i l e donors are making significant contributions to these areas, this i s largely outside o f the budget and has varied significantly from year t o year. By channeling more o f their assistance through the budget and ensuring that timely and accurate information i s made available t o the authorities for aid provided outside o f the budget, donors could further support strategic management o f the justice and security sector. Donors should establish and coordinate their medium-term financial commitments in support o f a justice and security sectoral strategy and programs. The aid coordination mechanism under preparation should serve as a framework to i n f o r m both the Government and the donor community o f the needs o f the sector. 138 Haiti: Public Expenditure Management and Financial Accountability Review 139 Haiti: Public Expenditure Management and Financial Accountability Review 140 Haiti: Public Expenditure Management and Financial Accountability Review 141 Haiti: Public Expenditure Management and Financial Accountability Review CHAPTER 7 : ASSESSING THE IMPACT OF FISCAL POLICY ON GROWTH AND POVERTY: A MACROECONOMIC FRAMEWORK 7.1. This Chapter provides an overview o f the model used to study the impact o f the composition o f public expenditure on growth and poverty in Haiti in various part o f this report.95A key feature o f the model i s that government spending i s disaggregated into various components, including maintenance, security and investment in education, health and core infrastructure. It also accounts for the externalities associated with infrastructure, in terms o f i t s impact on education and health. In addition, the model accounts for improved political stability and reduction in violence. Improvements in economic security contribute to the rise o f private investment by decreasing downside uncertainty o n the return to investment and securing property rights. Moreover, improved security may enhance the efficiency o f resource allocation and thus growth. In the model, spending o n security lowers violence and increases private sector confidence in the economy’s prospects; this tends to reduce the rate o f time preference and to increase private saving -which in turn stimulates private investment and growth. 7.2. The first part o f the chapter describes the key features o f the model and the second part describes the calibration procedure. 7.1 Deqcription o f the K e y Features o f the Haiti Macro-Model I t 7.1.1 T h e Supply side 7.3. The supply side represents the heart o f the model and i s summanzed in Figure 18. Four categories o f goods and services are produced in the economy: a commodity (produced by the private sector), and three types o f services - education and health (both o f which produced by the Government and the private sector) and infrastructure (produced solely by the Government). The privately-produced commodity i s a tradable good whose price i s taken as given; i t can be used for either consumption or investment. Production i s consumed only domestically and represents the sole source o f supply on the domestic market.96The provision o f education and health services by the public sector are free o f charge, whereas public infrastructure services are sold at a nominal price that i s fully indexed o n the price o f the private good.97 Excess demand for all services prevails; quantities consumed are thus supply-determined. There i s a single household-producer, which includes all workers (educated and non-educated, employed in either the public or private sector) in the economy. 95 A more detailed technical presentation of the model, as well1 as a description o f the calibration procedure, are available in E. Pinto Moreira and N. Bayraktar, “The Composition o f Public Expenditure and Growth: A Small-scale Intertemporal Model for Low-Income Countries.” Unpublished, World Bank (February 2007). 96 We therefore abstract from trade flows and balance-of-payments considerations. Because, as discussed below, borrowing i s fixed as a proportion o f output, and aid (in the form o f grants) adjusts residually to balance the budget, issues o f external debt sustainability do not arise. 97 The price o f the private good i s therefore used as the numkraire. 142 Haiti: Public Expenditure Manaqement and Financial Accountability Review Figure 18.The Supply Side , Public capital Medical Public capital ’ in education personnel in health 1‘ Quality I 1tt 7 T Production Private health Educated labor-, commodities Public capital in / ,,I,~~.n~ L infrastructure Quality Effective labor IPrivate capital Quality A I - Rawlabor L + Private investment K 7.4. First, consider the production o f health services and effective labor. Production o f public health services requires combining inputs at several levels. At the first level, public capital in infrastructure and public capital in health are combined to obtain the “effective” capital stock in the production o f health services. At the second level, the effective capital stock i s combined with medical personnel, which represents a fixed fraction o f the public labor force, to produce public health services. The private sector also produces health services. The value o f that production i s assumed equal to the household’s spending o n health services, which i s a constant fraction o f total private spending. Assuming that private and public services are perfect substitutes, the total supply o f health services i s simply given by the sum o f the two outputs. Effective (educated) labor employed in private production i s produced by combining the supply o f health services to the prevailing stock o f educated labor in that sector. 7.5. Second, consider the production o f commodities. I t i s also specified as a multi-level (Cobb- Douglas) process. At the first level, production requires combining effective educated labor and private physical capital to produce a composite input; at the second level, this composite input i s combined with uneducated labor to produce a composite input, V. At the final level, the supply o f commodities i s obtained by combining the composite input V with (quality-adjusted) public capital in infrastructure. 7.6. Third, consider population, schooling technology and the labor supply. Total population grows at a constant rate. The active population i s a fraction o f the total population. T h e supply o f raw labor i s the difference between the active population and the total supply o f educated labor. 143 Haiti: Public Expenditure Manaqement and Financial Accountability Review The transformation o f raw labor into educated labor requires an accumulation o f skills that takes place in part through a publicly-funded education system, which i s free o f charge. As before, a multi-level nested structure will highlight the role o f infrastructure and health on education. 7.7. At the first level, the stock o f public capital in infrastructure and the stock o f public capital in education produce a composite input, which i s referred t o as “effective” education capital. At the second level, effective education capital and the number o f teachers on government payroll (which represent a fixed fraction o f total public employment) are combined to produce a composite public education input, denoted Z. At the third level, the total number o f students i s combined with the supply o f health services to determine a composite input, which w e refer to as the “effective” supply o f students. At the fourth level, the “production” o f newly-educated workers by the public sector depends o n the fraction o f the effective supply o f students attending public schools, as well as the composite public education input, Z. The value o f that production (measured in terms o f the number o f educated individuals “produced” by private schools) i s assumed t o be proportional to household spending on education services, which i s given as a constant fraction o f total private spending. Assuming that private and public services are perfect substitutes, the total number o f educated workers produced in the economy i s again obtained by adding the two outputs. Given this new flow, the total stock o f educated labor in the economy can be calculated, for a gwen rate o f attrition. Assuming that public sector employment (which consists only o f educated workers) i s fixed as a proportion o f total supply, the supply o f educated labor involved in private production o f commodities i s determined residually. given the shares o f the educated labor force involved in the private production o f education and health services. Finally, wages in the private sector are assumed to be fully flexible; there i s therefore n o open unemployment o f either category o f labor. Box 10. H o w Public Infrastructure can Foster Growth: Lessons from Recent Studies98 Public infrastructure i s usually viewed as promoting growth through its impact on the productivity of private inputs, production costs and private investment (as a result of complementaritybetween public and private capital) whereas it tends to hamper growth and the allocation of resources through crowding-out effects on private capital formation. Recent analytical and empirical research has highlighted the fact that, in addition to these “conventional” effects, core public infrastructure may spur growth through a variety of other channel^.'^ Independent of its direct impact on the marginal product of production factors, public infrastructure may enhance further the productivity of labor if better access to means of public transportation(such as roads or railways) allows workers to get to their jobs more easily, therefore cutting the time spent commuting. By facilitating the reallocation of capital across sectors following from shocks to relative prices (e.g., an increase in the relative price of tradables, which would draw resources away from the nontradables sector), public infrastructure may reduce the magnitude o f adjustment costs associated with increases in private capital formation. An expansion in the road network may not only reduce congestion and facilitate the shipment of goods across the country (thereby reducing unit production costs, as noted earlier) but it may also reduce the cost o f building a new plant or the transportation of heavy equipment for installation to a new location for future production. T h e durability o f private capital may be significantly improved by improving the availability, and quality, of core public infrastructure. Reliable power grids and well-maintained roads tend to reduce the need for the private sector to spend on 98This box dwells largely o n Agknor, Pierre-Richard, and Kyriakos Neanidis, “The Allocation o f Public Expenditure and Economic Growth,” Working Paper No. 69, Centre for Growth and Business Cycle Research, University o f Manchester (March 2006), and Pierre-Richard, Agtnor and Blanca Moreno-Dodson, “Public Infrastructure and Growth: N e w Channels and Policy Implications,” Policy Research Working Paper No. 4064, World Bank (October 2006). 99 Core public infrastructure refers to energy (namely, electricity), transportation (roads, railways, etc.), telecommunications, and water and sanitation (including irrigation in rural areas). 144 Haiti: Public Expenditure Manaqement and Financial Accountability Review maintenance o f i t s o w n stock o f physical capital (for instance, the trucks that are used to move goods across the country). For instance, it has been estimated that for Latin America and the Caribbean, each dollar not spent on road maintenance leads to a $3 increase in vehicle operating costs as a result o f poor road conditions.”’ Better roads, by reducing the rate o f depreciation o f private capital, may raise the rate o f return on physical assets, thereby stimulating private investment and growth. A large body o f microeconomic evidence suggests also that core infrastructure (most importantly, electricity, roads, and sanitation) may have a significant impact on health and education outcomes - particularly in countries where, to begin with, infrastructure assets are low. Access to clean energy for cooking and better transport (particularly in rural areas) may contribute significantly to better health. According to the latest World Developmenf Report o f the W o r l d Bank, the dramatic drop in the maternal mortality ratio observed in recent years in Malaysia and Sri Lanka (from 2,136 in 1930 to 24 in 1996 in Sri Lanka, and from 1,085 in 1933 to 19 in 1997 in Malaysia) was due not only t o a sharp increase in medical workers in rural and disadvantaged communities, but also to improved communication and transportation services - that helped to reduce geographic barriers.”’ Transportation (in Malaysia) and transportation subsidies (in Sri Lanka) were provided for emergency visits to health care centers. Conversely, recent data produced by national Demographic and Health Surveys in Sub-Saharan Africa show that a majority o f women in rural areas rank distance and inadequate transportation as major obstacles in accessing health care.Io2 Studies have also found that access to clean water and sanitation has a significant effect on the incidence o f malaria, and more generally on child mortality. In the cross-section regressions for developing countries reported by McGuire (2006) for instance, average years o f female schooling have a statistically significant impact on under 5-years-old mortality rates.’03 At the same time, infrastructure may have a significant effect on education outcomes. Studies have shown that the quality o f education tends t o improve with better transportation networks in rural areas, whereas attendance rates for girls tend to increase with access t o sanitation in schools. In Morocco, for instance, after rural roads were built, girls’ enrollment rates rose from 28 percent t o 68 per~ent.’’~Electricity allows for more studying and access t o technology, such as computers, which enhance the quality o f human capital. I t i s also worth noting that the impact o f infrastructure o n health and education outcomes can be magnified through interactions between health and education themselves. There i s strong evidence that health has an impact on both the quantity and quality o f human capital. In most developing countries, schools, that lack access to basic water supply and. sanitation services tend to have a higher incidence o f illnesses among their students. In turn, poor health is an important underlying factor for low school enrollment, absenteeism (often the result o f respiratory infections, noted by Bundy and others (2005), poor classroom performance, and early school d r o p o ~ t . ’ ’ ~Inadequate nutrition, which often takes the f o r m o f deficiencies in micronutrients, also reduces the ability to learn. Conversely, healthier and better-fed children tend to perform better in class. Bundy et al. (op. cit.), in their overview 100 G y a m f i , Peter, a n d G u i l l e r m o Ruan, “Road Maintenance by Contract: D i s s e m i n a t i o n o f Good Practice in L a t i n A m e r i c a a n d the Caribbean Region.” L a t i n A m e r i c a and the Caribbean R e g i o n a l Studies P r o g r a m R e p o r t 44, World Bank (October 1996, p. 5). ‘“See World B a n k World Development Report 2006: Equity and Development, World Bank a n d Oxford U n i v e r s i t y Press (Washington DC: 2 0 0 5 ~ ) . ‘”African Union, Transport and the Millennium Development Goals, A d d i s A b a b a (February 2005). 103 See M c G u i r e , James W., “Basic H e a l t h Care P r o v i s i o n a n d under-5 M o r t a l i t y : A Cross-National S t u d y o f D e v e l o p i n g Countries,” World Development ( M a r c h 2006), 405-25. See also McCarthy, Desmond, H o l g e r Wolf, a n d Y i Wu, “The Growth Costs o f Malaria,” unpublished, G e o r g e t o w n U n i v e r s i t y (December 1999), a n d Philip Stevens, The Real Determinants o f Health, International P o l i c y N e t w o r k (London: 2005). 104 See L e v y , Hernan, “Rural Roads a n d P o v e r t y A l l e v i a t i o n in Morocco,” unpublished manuscript, World B a n k ( M a y 2004). 105 Bundy, Donald, and others, “School H e a l t h a n d Nutrition Programs,” in Disease Control Priorities in Developing Countries, ed. by D e a n Jamison a n d others, 2nd ed., Oxford U n i v e r s i t y Press ( N e w York: 2005). 106 See Bloom, D a v i d E., D a v i d Canning, a n d M a r k Weston, “The V a l u e o f Vaccination,” World Economics, 6 (July 2005), 1-13. ‘07See Glewwe, Paul, “Why does Mother’s Schooling Raise Child H e a l t h in D e v e l o p i n g Countries?,“ Journal o f Human Resources, 3 4 (- 1999), 124-59; a n d the cross-country regressions in Wagstaff, Adam, a n d M a r i a m Claeson, The Millennium Development Goals for Heath: Rising to the Challenges, World Bank (Washington DC: 2005). ‘‘See Agenor, Pierre-Richard, “A T h e o r y o f Infrastructure-led Development,” Working Paper No. 83, Centre f o r Growth a n d Business a n d Business C y c l e Research (December 2006). 145 Haiti: Public Expenditure Manaqement and Financial Accountability Review o f experience on the content and consequences o f school health programs (which include for instance treatment for intestinal w o r m infections), have emphasized that these programs can raise productivity in adult life not only through higher levels o f cognitive ability, but also through their effect on school participation and years o f schooling attained. Along the same line, Bloom, Canning and Weston (2005) found that children vaccinated (against a range o f diseases, including measles, polio and tuberculosis) as infants in the Philippines performed better in language and IQ scores as 10-year-olds as compared to unvaccinated children - even within similar social groups.’o6Thus, early vaccination may have a sizable effect on education outcomes and economic growth. At the same time, a number o f empirical studies have found that higher education levels can improve health. Studies have shown indeed that where mothers are better educated infant mortality rates are lower, and attendance rates in school tend to be higher.”’ Better-educated women tend, on average, to have more health knowledge and be more aware o f the myriad o f health risks that their children face. During the period 1970-95, improvements in female secondary school enrollment rates are estimated t o be responsible for 43 percent o f the total 15.5 percent reduction in the child underweight rate o f developing countries. The foregoing discussion suggests that the positive externalities associated with core public infrastructure may be substantial in low-income countries and must be accounted for in the design o f strategies aimed at fostering growth and reducing poverty. I f the production o f health services i s constrained by the lack o f availability o f infrastructure (lack o f electricity to run hospitals and refrigerate vaccines, lack o f roads t o allow easy access t o hospitals and clinics, etc.), a strategy designed to spur growth may need to rely heavily on a large, front-loaded increase in core public infrastructure. Size matters here not only because infrastructure investments are often lumpy in nature, but also because the network externalities associated with infrastructure, which translate into strong increasing returns (at least initially) in the productivity o f public capital, tend to “kick in” only after the stock o f infrastructure assets itself has reached a certain threshold.’08 1 Source: The authors. 7.1.2 Consumption and investment 7.8. Consider n o w consumption and investment decisions. There i s a unique household- producer w h o maximizes the present discounted value o f utility, which in turn depends not only o n consumption o f commodities but also consumption o f health services. The discount rate p i s endogenous as a result o f three factors. First, i t depends negatively o n consumption o f health services, t o capture the idea that better health leads t o a greater weight being attached to fiiture consumption, and therefore tends to lower the degree o f impatience. Second, i t depends also negatively o n total government spending on security, defined as the sum o f spending o n salaries o f public employees involved in security (the army, the police and the judiciary). The view here i s that spending o n security lowers violence, improves political stability, and raises private sector confidence in the economy’s future prospects; this tends to reduce preference f o r the present. B o t h of these effects tend therefore to increase private saving - and thus to stimulate investment and growth over time. Third, the rate o f time preference i s positively related t o wealth - that is, the stock o f private capital in the present context. This tends to lower saving and thus the rate o f economic growth. 7.9. The resource constraint faced by the household-producer equates consumption spending (inclusive o f taxes), physical capital accumulation, and user fees o n public infrastructure services to factor income (net o f taxes), the public sector wage bill, transfers f r o m the government, and net transfers f r o m abroad.’” In the accumulation equation for private physical capital, the rate o f depreciation i s taken to depend inversely o n the ratio o f public spending o n core infrastructure maintenance to the stock o f private capital. Thus, maintenance expenditure o n public infrastructure enhances the durability o f private capital. The budget constraint o f the private sector, together with I09 In addition, g i v e n that the household h o l d s no domestic debt, interest payments on that d e b t do not appear as a resource. 146 Haiti: Public Expenditure Manaqement and Financial Accountability Review the capital accumulation constraint, and the assumption that transfers are fixed as a fraction o f output, i s used to determine residually private investment. 7.1.3 Composition o f public spending 7.10. The government collects taxes (on wages o f educated workers, private capital income, and private consumption), and spends o n goods and services (including for maintenance and security purposes). I t also services its debt and invests in education, health and core infrastructure. Education and health services are provided free o f charge, whereas core infrastructure in i s subject t o fees. I t receives foreign assistance, which serves to balance the budget. 7.11. M o r e explicitly, total government spending, whose composition i s described in Figure 19, i s given by the sum o f consumption (current) spending, capital (investment) spending and interest payment. Current spending consists o f salaries to public sector workers, spending o n maintenance, spending o n security (other than salaries for the army, police, and judiciary), and other spending o n private commodities. Spending o n security and other items are both taken as a fixed fraction of output, whereas maintenance outlays are assumed to b e proportional to total depreciation o f all components o f the public capital stock. Figure 19.Composition of Public Expenditure Investment Investment Investment in infrastructure in education in health Wages t r and salaries L invest Pubsic ment aJL security Efficiency 1 Public Non-interest expenditure + Maintenance 1 capital \ Other spending Interest on goods/services Payments Total expenditure Transfers Depreciation of public capital 4 Depreciation of private capital 7.12. Total public investment i s taken to b e a fixed fraction o f output. Public investment i s allocated to education, health and core infrastructure as well as a residual item. Each component i s given as a fixed fraction o f total investment. Stocks o f public capital in education, health and 147 Haiti: Public Expenditure Manaqement and Financial Accountability Review infrastructure are determined by using the perpetual inventory method, modified to account for partial efficiency o f public investment. The rate o f depreciation o f each public capital stock i s taken to depend inversely o n the ratio o f public spending o n infrastructure maintenance to the relevant stock o f public capital. Thus, maintenance expenditure enhances the durability o f public capital. 7.1.4 Taxes and the government budget constraint 7.13. Taxes are subject t o collection costs; these costs (which are measured in terms o f the private commodities) reduce the y i e l d o f each tax by a fixed proportion.''* User fees are also subject to the same type o f collection costs. Total government revenue i s determined as the (cost adjusted) income and consumption taxes, as w e l l as user fees. With borrowing fmed as a fraction o f output, the government budget balance i s used t o determine the f l o w o f aid (grants). Therefore, in this mode, the model allows one to calculate aid requirements, for a given path o f spending, taxes (net o f collection costs) and borrowing, as illustrated in Figure 20."' F i g u r e 20. Government Budget--Overall Structure Public Public Transfers- Non-interest capita' Efficiency investment wages expenditure Foreign borrowing Tax Overall deficit - revenues (before grants) \ 4 Interest payments - t T Aggregate demand Foreign aid +-- t Domestic borrowing + Consumption investment + Private disposable + income + 2 "'Collection costs refer here only to direct administrative costs incurred by governments. See Bird and Z o l t (2005) for a further discussion. '"Alternatively, the model could b e solved for a specific component o f spending o r taxes, for a given level of aid-as a share, for instance, o f output. 148 Haiti: Public Expenditure Manaqement and Financial Accountability Review 7.1.5 Public capital: quality indicators 7.14. Finally, indicators o f quality o f public capital are all related to indicators o f excess demand (or congestion) on public services. The indicator o f quality o f public infrastructure, for instance, i s related to the ratio o f public infrastructure capital i t s e l f to the stock o f private capital, whereas the indicator o f quality o f public capital in health i s related to the ratio o f the stock o f public capital in health to the size o f the population. Finally, the indicator o f quality o f public capital in education i s related to the ratio o f the stock o f public capital in education itself to the number o f students attending public schools. T h i s specification captures congestion effects in the public education system due to overcrowded classrooms, as discussed elsewhere in the literature. 7.2 Calibration 7.15. The model i s calibrated for 2005, the most recent year for which a complete set o f macro accounts could be constructed. Data on national accounts and fiscal accounts were used to produce estimates. 7.16. Consider first the production o f health services. T h e share parameter PHC, which determines the roles o f public capital in infrastructure and public capital in health in determining the “effective” capital stock in the production o f health services (see equation (1) i s set at 0.3.’12 The share P H o f medical personnel in the public production o f health services (see equation (2)) i s n the same equation, the fraction xGH set at 0.6. I o f the total public labor force that i s employed as medical workers i s set at 0.015, which corresponds to the vahe for Haiti in 2005. I n equation (3), the share xPH o f total privzte spending allocated by households to expenditure on health services i s set at 0.032 for Haiti as well. 7.17. The share parameter PT in equation (4), which determines how the prevailirrg stock of educated labor in the private sector and health services are combined to create effective (educated) labor, i s set at 0.7. 7.18. In the production o f commodities, the share parameter PJ, which determines h o w effective labor and the private capital stock are combined to produce the intermediate input J (see equation (5), i s set at 0.7. Similarly, the share parameter PV, which determines how the composite input J and raw labor are combined to produce the intermediate input V (see equation (6)), i s set at 0.8. 7.19. In equation (7), we normalize output o f commodities, Y at 168 billion gourdes, which corresponds to the value o f Haiti’s GDP in 2005. The stocks o f public capital in infrastructure, health and education are taken to be relatively small to begin with. The infrastructure capital-output ratio i s set at 0.6, the education capital-output ratio at 0.3, and the health capital-output ratio i s set at 0.3. Overall, the aggregate (weighted) public capital-output ratio i s quite l o w by industrial- country standards, but it i s consistent with the average estimate o f the net public capital stock The obtained by Arestoff and Hurlin (2005b, Table 3) for a large group o f developing c ~ u n t r i e s . ” ~ ratio o f private capital to output i s set at 1.4. The resulting private-aggregate public capital ratio i s thus about 1.1. Put differently, o f the two components o f physical capital, public capital i s the relatively scarce factor; this i s consistent with the view (shared by many observers) that lack o f All the equations o f the model are listed in Annex to Chapter 7. Il3The Arestoff-Hurlin estimates are based o n the perpetual inventory method, w h i c h consists essentially in cumulating total capital expenditure flows by central govemments. 149 Haiti: Public Expenditure Manaqement and Financial Accountability Review public infrastructure in low-income countries (including Haiti) i s a major impediment to growth and private capital accumulation. Coefficients pY 1 and pY2, which determine directly the relative importance o f the composite input V and land, are set at 0.7 and 0.15, respectively; by comparison, the value o f pY2, used by Hansen and Prescott (2002), for instance, i s 0.3. 7.20. The estimates o f pY1 and p Y 2 imply that the elasticity o f output o f commodities with respect to public capital in infrastructure, given by l - p Y l - p Y 2 , i s equal to 0.15. This value corresponds to the one estimated by Easterly and Rebelo (1993) and used by Rioja (2005). By comparison, Baier and Glomm (2001) and Rioja and Glomm (2003) use an estimate o f 0.1 which i s close to the figure o f 0.1 1 estimated by Hulten (1996).Il4 Calderon and Serven (2005) also estimate the elasticity o f GDP to infrastructure (proxied by a synthetic index o f physical assets that includes energy, roads and telecommunications) to be 0.138 for a group o f developing countries, whereas Suescun (2005, p. 15) focusing only on Colombia, found a value o f 0.147. By comparison, Esfahani and Ramirez (2003, Table 4) found estimates o f the elasticities o f per capita GDP growth ranging from 0.08 to 0.16, when infrastructure capital i s measured as the number o f telephone lines or power generation capacity, whereas Canning (1 999) estimates an elasticity o f output per worker with respect to infrastructure (as measured by the number o f telephone lines) that i s on average 0.14 for his full sample, and close to 0.26 for higher-income countries. Similarly, Arestoff and Hurlin (2005b, Tables 2 and 7) found elasticities o f output per worker ranging from 0.05 to 0.19 when infrastructure stocks are used, and from 0.04 to 0.22 when estimates o f public capital stocks are used, in the absence o f threshold effects. Thus, the estimate used here i s consistent with the upper range o f the values estimated by Esfahani and Ramirez, and Arestoff and Hurlin, as well as the lower range o f Canning's result^."^ 7.21. I t should also be noted that, gven the multi-level Cobb-Douglas specification adopted here, the "true" elasticity o f output with respect to educated labor i s pTflJ$V.pYl, whereas the elasticity o f output with respect to private capital i s given by (I-pJ)$V.pYl, gwen the above estimates, we obtain, respectively, 0.27 and 0.17. The latter estimate i s significantly lower than the share o f private capital in output used in other studies, which i s 0.36 for Alonso-Carrera and Freire- Seren (2004, p. 852), 0.4 for instance in Ortigueira (1998, p. 337) and %vas (2003, Table l), and 0.45 in Rioja and Glomm (2003, Table 2). 7.22. Consider now population and the production o f education labor. The initial level o f population i s set at 8.5 million, which corresponds to Haiti's population in 2005. The autonomous growth rate o f the total population in equation (9), gNo, i s set at 2.2 percent, which corresponds to Haiti's value in 2005, whereas PN and pNNare set at 0.01 and 0.005, respectively. 7.23. In equation (lo), coefficients aD and as, which measure respectively the share o f dependents and the share o f students (both as a share o f the total population), are set at 0.252 and 0.163. 7.24. The share parameter PEC, which determines how the effective stocks o f public capital in infrastructure and education are combined to produce the composite input KGIE(t) (see equation 114Baldacci,Hillman, and Kojo (2004, p. 533) found an elasticity o f the growth rate per capita with respect to public capital expenditure that ranges f i o m 0.06 to 0.08 for a group o f 39 low-income countries for the period 1999-2001. "'Colletaz and Hurlin (2006, Table 5), using a smooth threshold regression approach, found estimates ranging from 0.07 (for France, Ireland, and the United States, for instance) to values as high as 0.29 for Norway and 0.38 for Portugal. Their average for 21 OECD countries i s -. 150 Haiti: Public Expenditure Management and Financial Accountability Review (12)), i s set at 0.35. Thus, infrastructure plays a relatively important role in determining how much physical capital i s used in the education technology. I n the next section, we will perform some sensitivity tests with respect to the value o f this parameter. 7.25. I n equation (13), the share parameter PZ, which determines how the composite public capital input and the number o f teachers on government payroll are combined t o produce the composite input Z, i s set at 0.8. Thus, physical capital i s as important as teachers in producing n the same equation, the fiaction xGE representing the share o f teachers in total educated labor. I public employment i s calibrated as 0.176, equal to Haiti’s value in 2005. 7.26. T h e share parameter PS, which determines how health services and the number o f students are combined to determine the composite input S H (see equation (14)), i s set at 0.8. I n equation (15), the share parameter PE, which determines how the composite inputs S H and Z are combined to determine the number o f newly-educated workers by the public sector, i s set at 0.6. I n that equation, the fraction xsHG, which measures the share o f the effective supply o f students attending nthis equation, the share o f the “effective” supply o f students enrolled public schools, i s set at 0.3. I G . in the public schools, Xsh , i s set equal t o the share o f actual students enrolled in that sector, which i s 0.185 for Haiti in the base period. 7.27. Given the nested structure o f the model, the “true” elasticity o f the production o f newly- educated workers by the public sector w i t h respect to the public capital stock in education i s given by (l-PEC).(l-PEC).(l-flE). From the estimates above, this value i s 0.052. Although our estimate i s smaller than the value used by Chen (2005), it i s close to the value used by Rioja (2005) and the econometric estimate obtained by Blankenau et al. (2005) for their full sample. Similarly, Perl: and Sakellaris (1998) used a share o f physical capital in final ouitput o f the education sector between 0.11 and 0.17. The estimate used here i s probably quite apprspnate for the group o f lew-income countries where education (at least at the primary and secondary levels) i s to a w r y large extent publicly provided.’16 7.28. I n equation (16), the share xpE o f total private spending allocated by households to expenditure on health services i s calibrated as 0.047 for Haiti in the base year. T h e share qNE, which measures the base-period ratio o f the number o f educated individuals “produced” by private schools and private spending on education, i s set at 0.00002, which represents the figure observed for Haiti in 2005. In (15) and (16), to estimate the flow variables NE(t) and NEG(t),we proceed as follows. We first calculate the change in the total number o f people who become literate within a year, by taking literacy rates in two consecutive years, multiplying them by the total population in that year, and taktng the absolute difference. This gives us an estimate o f NE(t). We then apply to that estimate the actual share o f students enrolled in public schools to obtain an estimate o f NEG(t). 7.29. In equation (17), the rate o f attrition o f the educated labor force i s set at 0.05. T h i s compares to a value o f 0.01 used by Alonso-Carrera and Freire-Seren (2004, p. 852) as an estimate n equation (18), the coefficient proportion aGE, which o f the rate o f depreciation o f human capital. I measures the share o f public sector employment in the total supply o f educated workers, i s calculated as 0.015, whereas apE and apH, the shares o f the educated labor force involved in the private production o f education and health services, respectively, are calculated at 0.0017 and 116 Blankenau et al. (2005) found that the elasticity o f human capital with respect to government spending o n education i s close to zero for low-income countries, but this r u n s counter to intuition. I t also does not account for the heterogeneity in public school enrollment discussed in the conclusion. 151 Haiti: Public Expenditure Manaqement and Financial Accountability Review 0.001 all o f which are the values for Haiti in 2005. By implication, the share o f educated workers employed in private production o f commodities, l-apE-apraGE, i s equal to 0.967. This gives an initial private capital-educated labor ratio in private production o f 78,581, and an overall capital- labor ratio o f 87,480. Keeping in mind that "educated labor" in the present context includes both s h l l e d and unskilled workers employed in production, these ratios (together with the capital-output ratios mentioned earlier) capture fairly w e l l the view that the country considered i s poor and endowed with a relatively abundant supply o f labor (with only part o f it educated), while facing at the same time a relative scarcity o f physical (particularly public) capital. 7.30. I n equation (20), o, the intertemporal elasticity o f substitution, i s set at 0.3. T h i s relatively l o w value i s consistent with the evidence indicating that the intertemporal elasticity o f substitution tends to be l o w at l o w levels o f income (see Ogah, Ostry and Reinhart (1996) and Agenor and Montiel (2007)), a result that may reflect either short planning horizons or liquidity constraints, as discussed for instance by Agenor (2004, Chapter 2).'17 7.3 1. I n the resource constraint o f the private sector, equation (2 l), private consumption i s set at 93.3 percent o f output. This value i s quite sensible for many low-income countries, where limited private resources are allocated t o savings and investment. It corresponds to the value observed for Haiti in 2005. The tax rate on (factor) income, zy, i s set equal to 0.042, whereas the tax rate o n consumption i s set equal to 0.057, both o f which correspond t o the values observed for Haiti in 2005. The first value i s in line with actual ratios for many low-income countries, where taxation (which i s essentially indirect in nature) provides a more limited source o f revenue than in higher- income countries. The coefficient €lo in that equation, which measures the share o f other current government spending allocated to transfers to households, i s calculated as 0.009, which corresponds to the value observed for Haiti in 2005. 7.32. Coefficients pH and pS in the discount rate function (equation (22) tire set at 0.01. The estimate o f the first coefficient i s based o n the results in Lawrance (1991), who identified an (inverse) relationship between the rate o f time preference and the level o f income, with an elasticity o f 0.058. Assuming that spending o n health i s services i s more or less proportional to income (or expenditure, given the l o w degree o f intertemporal substitution), this elasticity can be used as an approximation proportional to pH. T o ensure a reasonable initial value o f the discount rate (given the values o f pH and pS and the initial values o f H(t)/JSp(t)and CGs(t)/Kp(t)), w e set po such that the value o f p in the initial period i s equal to 0.0037. The rate o f time preference, p, i s set at 4 percent, a fairly conventional choice in this literature. This leads t o a discount factor o f approximately 0.96 (see, for instance, Canton (2001, Table l), and Ghosh and Roy (2004, Tables 1 and 2)). 7.33. I n equation (24), the rate o f depreciation o f the private capital stock, Sp(t), i s defined in such a way that i t s initial value i s equal t o 6.8 percent. This value corresponds to the average value estimated by Bu (2006, Table 8) for three low-income countries in Africa. I nthat equation, w e also set xP = 0.002, and then calibrated cp = 0.94. 7.34. Turning now t o the government budget, in equation (27), the effective interest rate o n the public debt i s calculated as 0.024. The shares o f current and capital spending in total government Ii7Of course, using even lower values o f the ntertemporal elasticity o f substitution would "flatten" the response o f consumption to shocks. However, they would not affect the direction o f the effect discussed below. 152 Haiti: Public Expenditure Management and Financial Accountability Review spending are set equal to 0.64 and 0.3, respectively, implying a share o f interest payments o f 0.06. These shares correspond to those observed for Haiti in 2005. 7.35. In equation (28), the shares o f salaries to public sector workers, maintenance, spending on security (other than salaries) and transfers in total current spending are set equal to 0.38, 0.11,O. 15, and 0.27 respectively, implying a share o f spending on other categories o f 0.25. These shares imply that, the coefficients 0, in equations (29), which measure the shares o f spending o n security (excluding salaries o f security personnel), transfers, and other items in GDP, are equal to 0.014, 0.009, and 0.024, respectively. In equation (30), the coefficients eM1,eMH,and are set so that the value o f total spending on maintenance for each item i s indeed equal to i t s share in total current spending, that is, 0.40,0.23, and 0.21, respectively. 7.36. Coefficient € Iwhich I, measures in equation (31) the share o f total public investment in GDP i s set equal to 0.04, the value for Haiti in 2005. This number i s consistent with the share o f public investment in total government spending. The allocation o f public investment between education, health, infrastructure, and other categories, i s determined by the coefficients KE, KH, K ~ K , ~ which , are calculated as 0.053, 0.054, 0.36, and 0.53, respectively. These ratios correspond to those observed for Haiti in 2005. 7.37. In equation (34), the degree o f (in) efficiency o f public investment, that is, (Ph for h = E,H,I, i s set uniformly at 0.5. Arestoff and Hurlin (2005) found values o f cp rangmg from 0.4 to 0.6 for a group o f developing countries. In the experiments reported below, we use at f i r s t the uniform value o f 0.5, and perform subsequently a sensitivity analysis. In equation (35), the rates o f depreciatmn o f each capital stock, sGh(t>, i s set so that the ratios are equal to 3.5 percent in the base period. Similar values are used by Agknor, Bayrdktar, and El Aynaoui (20061, and Pinto h h r e i r a and Bayraktar: (2006). The coefficients EGh and XGh i s set at 0.001, and &Gh i s calibrated as 0.979. 7.38. In equation (36), the coefficients measuring the tax collection costs, qc and qy, are set at 0.03 and 0.06 respectively, whereas the cost o f collecting infrastructure fees q, i s set at 0.06. Thus, I collecting income taxes and fees are assumed to be twice as costly as collecting consumption taxes."' The value o f 0.03 corresponds to the average o f administrative costs (in proportion o f taxes collected) estimated by Gallagher (2005, p. 127) for a group o f low-income developing countries. In equation (38), the coefficient OD, which measures the ratio o f borrowing (both domestic and foreign) as a fixed fraction o f output i s set at 0.007, which corresponds to the value observed for Haiti in 2005. 7.39. I n the private investment equation (40), the share o f foreign transfers as a proportion o f GDP, eR, i s set at 0.21, which corresponds to the value observed for total transfers for Haiti in 2005. Finally, for the quality indicators defined in (41), (42) and (43), coefficients Oh, are chosen so that the initial values o f these indicators i s relatively low, at 0.4. lisNote that we assume that the cost o f collecting taxes on both components o f factor income (wages and profits) is the same. In practice, however, collection costs m a y b e higher for non-wage income. See Agenor and Neanidis (2006b) for a more detailed discussion. 153 Haiti: Public Expenditure Manasement and Financial Accountability Review CHAPTER 8 : ADDRESSING GROWTH AND POVERTY CHALLENGES: FISCAL REFORMS AND AID REQUIREMENTS 8.1. T h i s chapter presents policy experiments camed out using H a i t i macro-model, including: (i) increase in public investment with l o w and high efficiency o f investment; ( ii) budget-neutral reallocation o f spending toward health; ( iii) an improvement in fiscal management; (iv) an increase in user fees on core infrastructure; (v) an increase in security spending; and (vi) a composite fiscal package. 8.2. The key findings are the following. First, an increase in public investment leads to higher growth rates compared to the baseline. However, although growth i s higher, it i s not enough to entail a substantial drop in the poverty rate. But when efficiency o f public investment i s higher, growth rates are much higher, and the poverty rate drops much faster and t o lower levels than in the case o f l o w efficiency. Second, a reallocation o f spending toward investment in health has limited impact o n growth and poverty reduction. Third, a reduction in collection costs combined with an increase in effective direct tax rate has a negative impact on growth and worsens poverty, assuming that public spending remains unchanged. However, i f the additional revenues generated by higher income taxes and lower collection costs, are reallocated (across the board) t o public investment, the growth rate o f output per capita increases and the poverty rate declines. Fourth, a permanent increase in user fees combined with the reallocation o f additional revenues across the board to spending o n maintenance results in a small positive impact o n growth and lower poverty rate. Fifth, an increase in security spending (per se) has almost n o discernible effects o n growth and poverty. However, when spending o n security have a relatively large impact o n agents’ rate o f preference for the present; in such conditions, improvements in security, t o the extent that they translate into reductions in crime and violence leads to higher growth rates and lower poverty rates. Sixth, a composite fiscal package combining several o f the previous experiments, results in higher growth rates and lower poverty rates. Seventh, the different components o f the fiscal reform package point out the dynamic trade-offs between growth-poverty and improvement in public finances. 8.3. The main policy lesson drawn from our analysis i s that a fiscal reform package should be designed not only for the objective o f improving fiscal performance but also with a view to assess i t s impact o n growth and poverty. T h i s requires a dynamic macro-model. 154 Haiti: Public Expenditure Manaqement and Financial Accountability Review 8.1 Baseline Scenario, 2007-2015 Figure 21.Baseline scenario, 2007-15 8.4. W e need t o build a baseline scenario to be able to 1 l2 conduct policy experiments with the model. I n the baseline scenario, budget deficit as a share o f GDP i s taken fixed to make aid endogenous. I t should be noted that this closure rule can be changed. The shares o f all other components o f spending remain constant at base period values. The efficiency parameter o f ’1 04 2007 2008 2009 2010 2011 2012 2013 2014 2015 public investment i s equal to the -Real GDP per capita at market prlces (% change) uniform value o f 0.5. 8.5. Table A.8.1 in Annex - .......Foregn ad (56 of GW) _ _ _ _ Total revenues (% of GDP) mverty rate-Growth elasbcity of -1 0 (nght scale) presents the baseline scenario. Source :Authors’ Calculations. Given that current conditions continue, the growth rate o f real GDP per capita at market prices i s estimated to increase only slightly, which i s basically caused by the supply side effect o f increasing public investment as a share o f GDP from 4.5 percent to 5.7 percent, and private investment from 32.3 percent in 200’7 to 38.3 percent in 2015. The l o w growth rate leads to a n;inor drop in the poverty rate whether we use Ravallion’s adjusted elasticity or the growth elasticity o f -1.O. F o r example, the poverty rate with * Ravallion’s adjusted elasticity decreases from 55 percent in 2087 to 52.8 percent in 2015. If the current trends were t o be maintained, the prospects o f reducing poverty would not be realized and the MDGs o f halving poverty by 2015 would not be achieved. T h u s the results indicate that foreign aid at the level o f five percent o f GDP would not be enough to obtain desired growth rates to reduce poverty. 8.2 Policy Experiments 8.6. This section illustrates the properties and implications o f the model by considering six different policy experiments: an increase in public investment (financed by aid); a budget-neutral reallocation of spending toward health; an improvement in fiscal management that takes the alternative forms o f a reduction in collection costs, an increase in user fees o n core infrastructure, and an increase in security spending; and a composite fiscal package, that combines elements o f all the individual experiments listed above. I n line with the favorable international environment that H a i t i faces currently, all experiments are conducted under the assumption that the overall budget deficit i s constant and aid i s the balancing item in the government budget.”’ “’As indicated earlier, one could also consider the case where aid is fixed in proportion o f GDP, with the balancing item in the budget being either a component o f non-interest expenditure, or a tax rate. 155 Haiti: Public Expenditure Manaqement and Financial Accountability Review 8.2.1. Increase in Public Investment 8.7. Our first experiment consists o f a temporary increase in total public investment as a share o f GDP by 5 percentage points starting in 2008 until 2011, then dropping by 1 percentage point each year after that, to eventually return to the initial baseline value. W e consider two variants: first, the case where the efficiency parameter o f public investment i s constant throughout at 0.5, and the second case where the efficiency parameter (for all categories o f public investment) improves gradually over time. In both scenarios, investment i s totally financed by foreign aid, due t o the closure rule described earlier. 156 Haiti: Public Expenditure Manaqement and Financial Accountability Review 8.8. Simulation results for the first variant are shown in Table A.8.2. As Figure 22.Higher Total Public Investment, 2007-15, in a l l subsequent tables, they are (Deviation from the Baseline scenario) displayed as absolute differences 0.5 , fkom the baseline scenario. 8.9. The direct effect o f the increase in public investment i s o n the stock o f public capital in infkastructure, which tends to stimulate output. Because the growth in output exceeds the increase in consumption, private capital -0.5 J 2007 2008 2009 2010 2011 2012 2013 2014 2015 formation expands.l2O By 2015, private investment increases by nearly 1.7 percentage points o f GDP. - - MGop per cap- at market pres (% chawe) mvWh/ rate-Grow th Blxt~Py - - - - TMal revenues (% of GW of -1 0 ....... For&" a a (K d GW, right scale) Thus, the rise in public investment crowds in private investment through Source :Authors' Calculations. an indirect complementarity effect. In turn, the increase in private Figure 23.Higher Total Public Investment and Higher investment raises the stock o f private Efficiency of Public Investment, 2007-15, capital over time; this, combined with (Deviation from the Baseline Scenario) the increase in the stock o f public 1 - - 6 capitdl in infrastructure, tends t o increase the marginal productivity o f a l l other production inputs. At the same time, the rise in public investment in education leads t o an increase in the stock o f capital in -0.4 education and the public education input, and therefore to a higher supply o f educated workers.'21 In -1 - : - 1 -1.2 J : l o addition to improvements in the 2007 2008 2009 2010 2011 2012 2013 2014 2015 public infkastructure and education --Real WPpercapitaatmarketprkes(%change) capital stock, the increase in the stock --c-mverty rai~Growthelastidy of -1.0 o f public capital in health raises the ---- Total revenues ("/.of GDp) .......Foreign ad (%of GCf, nghtscale) efficiency o f educated labor in production. The productivity gains Source : Authors' Calculations. associated with the combined effect o f improved effective labor, and increased margmal productivity o f a l l inputs, contribute to higher output. In terms o f growth rates, output per capita remains o n a sustained basis at 0.3 percentage points above its level in the baseline case. However, although growth i s higher, it i s n o t enough t o cause a substantial drop in the poverty rate; even in the case o f neutral growth elasticity, the poverty rate drops by only 1.1 percentage points by 20 15. 120 In the model, given that the estimate o f the intertemporal elasticity o f substitution that we use i s relatively l o w (in line with the evidence for low-income countries), consumption smoothing i s significant, implying relatively small changes over time in private expenditure. 121 The ratio o f educated workers to population changes only slightly, however, due to a relatively l o w degree o f substitution between teachers and public capital stock in the production o f education services. 157 Haiti: Public Expenditure Manaqement and Financial Accountability Review 8.10. In the second variant, in addition to higher public investment as described above, efficiency o f all categories o f public investment i s assumed t o improve uniformly over time. Specifically, it i s assumed that the efficiency parameter remains constant at 0.5 in 2006 and 2007, and then increases to 0.8 by 0.1 point each year between 2008 and 2010. After 2010, it remains constant at 0.8. This case may represent the reforms aimed at improving governance and eliminating mismanagement o f public resources -a key policy challenge in Haiti, as in many other developing countries. 8.1 1. Table A.8.3 summarizes the simulation results. Now, because o f improved efficiency o f public investment, the rate o f accumulation o f a l l categories o f public capital i s higher, thereby magnifying productivity effects on private inputs. In turn, higher rates o f factor accumulation lead to higher growth rates o f GDP per capita, relative to the first variant. For example, Table A.8.3 shows that the growth rate o f output relative to the baseline value rises t o 0.6 in 2015, whereas it was only 0.2 in Table A.8.2. This higher growth rate o f GDP per capita translates into a lower poverty rate. With a growth elasticity o f -1 .O, the drop in the poverty rate reaches 2.5 percentage points by 2015 (relative to the baseline scenario) in case o f higher efficiency o f public investment, compared to 1.1 percent drop with the first variant o f the experiment. 8.2.2 Spending reallocation 8.12. Our third experiment i s a typical “fiscal space” experiment. It consists o f two Figure 24.Reallocation of spending to health ,2007- 15 (Deviation from the Baseline scenario) components. First, starting in 2008, there i s a permanent reduction o f 1 percentage point OO81 o f GDP in “other” public spending which i s reallocated t o investment (across the board, that is, keeping constant the initial shares in _________............. public capital formation). Second, the share o f the residual category “other” in public investment i s reduced permanently by 5 -0.04 1 percentage points, with the whole amount -0.06 4 reallocated to investment in health. Given 4.08 1 2007 2008 2009 2010 2011 2012 2013 2014 2015 that the experiment consists o f a spending --ai m p r capia at mlprces 1% change) reallocation, foreign aid requirements do not --c~ethl,ale-GiowInelastr;miof -1 0 ....... Fmegnad(%of Gcp) change o n impact (although they may ___ - Tolal (% d TBYBWLIBE Gcp) change subsequently, given the dynamics of Source : Authors’ Calculations. the model). 8.13. The simulation results are presented in Table A.8.4. The impact o n output growth per capita i s relatively weak. I t increases only by 0.1 percentage points by 2015. A s a result, the effect on poverty i s weak as well. The proportion o f poor drops by only 0.2 percentage points in the case o f a neutral-growth elasticity. Thus, the results indicate that, by itself, and given the magnitude o f the shock, higher public health investment i s not enough to have a strong impact on growth, despite i t s positive effect o n the productivity o f the labor force and i t s positive effect on the incentive to save (which results from a reduction in the rate o f time preference). 158 Haiti: Public Expenditure Management and Financial Accountability Review 8.2.3 Tax Reform Figure 25.Lower collection cost, higher direct tax rate, 8.14. In this set o f experiments, 2007-15 (Deviation from the Baseline collection costs are reduced by h a l f scenario) starting in 2008 and the effective direct tax rate i s increased by 1 percentage point over three years, starting in 2008. W e ..' r _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - - - - - consider again two variants: first, w e assume that public spending does not change; second, w e consider the case where the additional resources generated by tax reform are allocated t o investment. .......________...___________________________. -3 i -4 J 8.15. The results o f the first variant are 2007 2008 2009 2010 2011 2012 2013 2014 2015 shown in Table A.8.5. The direct effect -%ai G W p m capda a1 m k c l WEBS (% chsngsj o f the tax reform i s an increase in tax --cRNarty rats-Oowlh eI=ll~tty of -1 0 .......Focegnad (% of Gopl revenues. But given that public spending ----T~ldrw~n-(%d GW does not change, and that the budget Source :Authors' Calculations. deficit i s constant, the increase in revenues translates almost one t o one into Figure 26.Lower collection cost, higher direct tax rate, a f a l l in foreign aid requirements. new revenue to investment in infrastructure, Moreover, the impact o n growth i s 2007-15 (Deviation from the Baseline largely negative: higher income taxes scenario) tend to induce households t o consume 35 more today and to reduce saving rates. As ~ a result o f this intertemporal effect, private investment drops, thereby lowering the marginal productivity o f a l l production inputs. As a result, output drops and poverty worsens. -0.5 - 8.16. n the I second variant, we 1 - reallocate (across the board) the -1.5 J additional revenues generated by higher 2007 2008 2009 2010 2011 2012 2013 2014 2015 income taxes and lower collection costs, -&ai GCPper capita at m r b t pfices 1% change) rate-Growth e b t k i i of -1.0 t o public investment. The results are --cNverty .......F o M n aid ( X of G W shown in Table A.8.6. This time around, ---- Total revenues (X of G W private investment increases as w e l l due Source : Authors' Calculations. t o the improvement in the public infrastructure capital stock w h i c h stimulates output (and thus after-tax income) and savings. As a result, the impact o n growth i s positive. The growth rate o f output per capita increases by 0.8 percentage points by 2015, compared to a 0.1 percentage point drop in Table A.8.5. Thus, the poverty rate declines n o w by 2.4 percent by 2015, in the case o f a neutral growth elasticity. T h i s last experiment illustrates w e l l the importance o f reallocating additional revenue f r o m fiscal reforms to investment (assuming that external financing conditions do n o t change). Otherwise, the disincentive effects o f taxation on private savings and investment m a y w e l l lead to lower growth and higher poverty. 159 Haiti: Public Expenditure Manaqement and Financial Accountability Review 8.2.4 Increase in security spending Figure 27.Higher security spending, 2007-15 (Deviation from the Baseline scenario) 8.17. This experiment consists o f an T increase in security spending by 3 0,004 3.5 percentage points o f GDP between 2008 and 201 1, followed by an increase o f 2.5 nnn? .,___ 0,002 0.001 J / .......................... ’.. -------I t3 percent in 2012, 2 percent in 2013, and 1.5 percent in 2014 and 2015. Given the budget closure rule, higher security spending i s essentially financed by -. ----- -0.004 -. --_--- foreign aid, which rises at about the same rate. Table A.8.7 presents the simulation 2007 2008 2009 2010 2011 2012 2013 2014 2015 results when the elasticity o f security --3 GCPpsr cspna 81 wake1 pwes (%change) spending, pS, in Equation (22) i s taken to --eFbvertyrdaUowVIsiasl~nyof ____ Told revenu~p( X of G W -1 0 be 0.01. In this case, there are almost n o ....... Faalgn a d (% of O W iohl scale) discernible effects o n growth and Source : Authors’ Calculations. poverty. Figure 28.Higher security spending and Elasticity of 8.18. By contrast, Table A.8.8 shows Security Spending, 2007-15 (Deviation from the results when the elasticity o f security the Baseline scenario) spending, pS, i s equal to 0.1. 0.08 , Conceptually, this case corresponds to a situation where security concerns have a relatively large impact o n agents’ rate o f , 3.5 3 - - 4. T t 25 preference . f o r the present; in such conditions, improvements in security, to the extent that they translate into reductions in crime and violence, may translate into greater incentives to “think about” the future, and thus to save. 4.06 6.1 : 1 --. --- Indeed, the results show that the impact 2007 2008 2009 2010 2011 2012 2013 2014 2015 o f higher security spending on growth i s -- I G W w r c a m ai m r k l prces (% chawe) d m v e r t y ratPGrow1h clasbcnY of -1 0 now quite significant. The growth rate _ _ _ _ Total revenues (46 of G D q rises by about 0.1 percentage points by .......Faregnad (% of GCP rgMscale) 2015, whereas the poverty rate drops by Source :Authors’ Calculations. about 0.12 by 2015 when the growth elasticity i s neutral. Again, the higher value o f pS leads t o a higher growth rate because private sector confidence in the economy’s future prospects improves more with higher security spending. By inducing a greater reduction in preference for the present, private saving increases. Therefore, this generates a stronger effect on private investment and growth. 8.2.5 A composite fiscal package 8.19. W e n o w consider a composite fiscal package, which combines several o f the previous experiments. 0 An increase in total public investment in GDP by five percentage points beginning in 2008 until 20 11, then dropping by 1 percentage point each year after 20 11; 160 Haiti: Public Expenditure Manaqement and Financial Accountability Review A permanent reduction o f one percentage point o f GDP, starting in 2008, in the “other” category o f public spending which i s reallocated (across the board) t o investment, with at the same time a permanent reduction in the share o f the category “other” in public investment by five percentage points, reallocated in i t s entirety t o health; An increase in the effective indirect tax rate t o six percentage points begmning in 2008; An increase in the direct tax rate by 1 percent for three years, beginning in 2008; An increase in security spending by three percentage points o f GDP between 2008 and 2011,2.5 percent in 2012,2 percent in 2013, and 1.5 percent in 2014 and 2015. A reduction in collection costs by half, starting in 2008. Figure 29.Combined shock with lower Figure 30.Combined shock with higher security collection cost, higher direct tax and expenditure, direct taxes, and elasticity security spending, 2007-15 (Deviation of security expenditure, 2007-15 from the Baseline scenario) (Deviation from the Baseline scenario) 5 , T 57 2001 2008 2009 2010 2011 2012 2013 2014 2015 ’ -2 1 -.- -bel ____ GDPper cwpila el r baty iete-~ii- Told i c y m u 1% ~ ~oi G m lprres 1% change) of -I o mekirr~y W b 2007 2008 -RealGCQper - 2009 2010 2011 capka at market prIies 1% change) cmverty raRGrowth ebstc$ of -1 0 2012 2013 2014 2015 ... ...Foregoad 1% of GW, rwhl ssde) _ _ _ _ Total revenues 1% of G w ) ..... Forepn a d (% of GCQ, rlgM scak) Source :Authors’ Calculations. Source : Authors’ Calculations. 8.20. Table A.8.10 presents the simulation results. W e observe t w o opposite effects o n growth and poverty. As the tax rates increase, people start saving less, as discussed earlier. As a result private investment and private capital accumulation slows down. Thw leads to an initial negative impact o f the fiscal package o n growth and poverty. But at the same time, the higher tax rates and lower collection costs raise government revenue, which increases public investment and thus the various components o f public capital. Over time, the larger public capital stock, directly and indirectly, raises saving and investment, increases output, and lowers poverty. In the medium term, the impact o n growth turns out to b e positive. W h i l e the growth rate o f real GDP per capita increases by 0.6 o n average, the poverty rate with a growth elasticity o f -1.0 drops by 2.2 percent in 2015. 8.21. If the elasticity o f security spending, pS, rises, the effect o n growth improves slightly. Table A.8.10 shows the simulation results when pS i s taken as 0.1 instead o f 0.01, as in Table A.8.9. In this case, private investment increases more, because (as discussed earlier) private sector confidence in the economy’s future prospects improves. T h i s leads to a higher rate o f output growth and lower poverty. 161 Haiti: Public Expenditure Management and Financial Accountability Review 8.22. The impact o n growth improves even more i f the government applies a lower increase in the direct tax rate and makes it effective later. Tables A.8.11 shows the case when the direct tax rate increases by 0.5 percentage points only between 2010 and 2013 and then stays constant. The lower rise in the direct tax rate leads t o a lower drop in savings during the initial phase o f adjustment, and therefore t o a smaller negative effect o n private investment. In this case, the growth rate o f real GDP per capita increases by 0.7 percentage points o n average, compared t o the 0.6 increase shown in Table A.8.9. As a result, the poverty rate (with a growth elasticity o f -1.0) decreases by 2.6 percentage points in 2015, instead o f 2.2. Table A.8.12 shows that the effect o f changes in fiscal policy o n growth and poverty get even better when the elasticity o f security spending i s raised t o 0.1, given that in this case i t leads to higher private investment. Figure 31.Combined shock with lower collection Figure 32.Combined shock with lower collection cost, higher security spending, and cost, higher security spending, elasticity direct tax increasing later, 2007-15 of security spending, and direct tax (Deviation from the Baseline scenario) increasing later, 2007-15 (Deviation from the Baseline scenario) 3.5 , ........... ... T 8 3- 2.5 2- 1.5 - .... i : 3.5 , T 8 1- . 0.5 - 0 - -0.5 - 1 - -1.5 - ...... 2007 2008 2009 2010 2011 2012 2013 2014 2015 -F$alG[Ppercapnaalnsiketpiasr (%change) --"erhlrsle-~owIhda31*;1RIai-l 0 . . - - - - - T o U R V ~ ~ U B L ( % ~G w ) ....... b e g n aid 1% Of GW, ight $ d e ) Source :Authors' Calculations. 8.23. Overall, the results o f our experiments illustrate that the impact o f changes in fiscal policy will depend o n h o w i t affects the behavior o f the private sector (notably through incentives to save and invest) and h o w additional revenue i s allocated by the government. Regarding the latter, i t i s worth pointing out that even tax reforms that are considered t o b e highly regressive (involving sharp increases in indirect tax rates) m a y end up being beneficial to the poor, t o the extent that the resources that they generate are allocated t o productive capital accumulation. Moreover, these effects may vary in opposite direction over time: effects o n poverty m a y b e negative in the short- term but may become favorable over time. Thus, dynamic trade-offs m a y emerge in the design o f fiscal reforms and using a dynamic model (such as the one developed in this chapter) i s essential to capture them. 162 Haiti: Public Expenditure Management and Financial Accountability Review REFERENCES P-R. AgCnor et alii. Adjustment Policies, Poverty, and Unemployment. T h e IMMPA Framework. Blackwell Publishing. Forthcoming. Banque de l a Republique d’ Haiti. 2004. Rapport Annuel. Banque Mondiale. Public Expenditure and Financial Accountability. DCpenses publiques e t responsabilitk financihe. Cadre de mesure de l a performance de l a gestion des finances publiques. Secretariat PEFA. Banque Mondiale. Juin 2005. Briceiio-Garmendia, C., A. Estache, and N. Shafik. 2004. “Infrastructure Services in Developing Countries: Access, Quality, Costs and Policy Reform.” Policy Research Working Paper 3468. World Bank, Washington, D.C. Cadre de Cooptration Inttrimaire (CCI). Bilan des Financements et des RCalisations. PCriode : Juin 2004-DCcembre 2005. Ministkre de 1’Education Nationale, de l a Jeunesse et des Sports. Mars 2006. Calderh, C., and L. ServCn. 2004 “The Effects o f Infrastructure Development on Growth and Income Distribution.” Policy Research Workmg Paper 3400. World Bank, Washington, D.C. COCCI. 2005. “Rapport d’Evaluation de l a M i s e en Oeuvre du Cadre de Cooperation Interimaire.” COCCI. Collier, P and A. HoeMer. 2002. “Aid, Policy and 6rowth a n Post-Conflict Societies.” Policy Research Worlung Paper 2902. World Bank, Washington, D.C. B. Eifert and A. Gelb. Improving the Dynamics o f Aid. Towards More Predictable Budget Support. World Bank. Policy Research Working Paper No. 3732. October 2005. Fisher, C. et al. 2006. Haiti: Rapport de l a Mission Multisectorielle, Fonds Monetaire International, 14 fevrier 2006. Fay, M. (Ed.). 2005. “The Urban Poor in Latin America.” World Bank, Washington, D.C. Fay, M. and M. Morrison. 2005. “Infrastructure in Latin America and the Caribbean: Recent Developments and K e y Challenges” World Bank, Washington, D.C. Gajdeczka, P., Jaramillo, L., Everaert, G.,Sancak, C., Dalsgaard, T., and Maticen, J. 2005, Haiti, Selected Issues, International Monetary Fund. Government o f Haiti. L o i de Finances 200 1-2002; 2002-2003; 2003-2004; 2004-2005; 2005-2006; 2006-2007. L e Moniteur. Journal Officiel de l a Rhpublique d’Ha’iti. NumCros SpCciaux. IDB and IMF. Examen des dCpenses publiques d’Ha’iti. Notes 1-9. mars 1997. 163 Haiti: Public Expenditure Management and Financial Accountability Review IMF. 2005. Staff Report for the 2005 Article I V Consultation and Review o f the Program Supported by Emergency Post-Conflict Assistance. April 29,2005. I n s t i t u t Haitien de Statistique et d’Informatique, Enquete sur les Conditions de Vie e n Haiti (ECVH-2001). IMF. Programme de Renforcement de l a Politique Fiscale et des Administrations Fiscale et Douanii-re. 16 mai 2005. IMF. 2006. Fiscal Adjustment for Stability and Growth. Pamphlet Series No. 55. Interim Cooperation Framework (ICF). Summary Report. July 2004 IMF. Haiti. Request for Three-Year Arrangement Under the Poverty Reduction and Growth Facility. November 7,2006. Jha, A. K. (Ed.). 2005. “Institutions, Performance and the Financing o f Infrastructure Services in the Caribbean.” World Bank - Workmg Paper No. 58, Washington D.C. KPMG. 2004. Corporate Tax Rates Survey. January 2004. Lundahl., M. 2004. Sources o f Growth in the Haitian Economy. Inter- American Development Bank. Ministry o f Agriculture and Worldbank. 2005. Diagnostic and Proposals for Agnculture and Rural Development Policies and Strategies, October 5,2005. Ministi-re de l a Planification et de l a CoopCration Externe (Direction d’Evaluation et de Contrble, DEC). Programme d’Investissement Public 2004/2005. Bilan d’Execution. DCcembre 2005. Ministere de l a Planification et de l a CoopCration Exteme. Carte de PauvretC d’HaYti. Version 2004. Ministkre de l a SantC Publique et de l a Population. Plan StratCgique National pour la RCforme du Secteur de l a SantC, 2005-2010. Novembre 2005. Nexant and Econergy (for USAID). Haiti - Energy Sector Assessment, March 2005. Orozco, M. 2006. Understanding the remittance economy in Haiti, Inter- American Dialogue, World Bank. PNUD. Situation Cconomique et sociale d’HaTti en 2004. Ratha, D. 2005. “Workers’ Remittances: An Important and Stable Source o f External Development Finance”. In: Remittances. Development Impact and Future Prospects. Edited by Samuel Munzele Maibo and Dilip Ratha. Republic o f Haiti. ConfCrence Internationale pour l e DCveloppement Economique et Social d’HaTti. Document de Travail. Port-au-Prince. 25 Juillet 2006. 164 Haiti: Public Expenditure Management and Financial Accountability Review - InterimPoverty Reduction Strategy Paper (IPRSP). - September 2006. Schiavo-Campo, S. 2003. Financing and Aid Management Arrangements in Post-Conflict Situations. World Bank - CPR Paper No. 6, Washington D.C. Schwartz, J., S. Hahn, I.Bannon. 2004. “The Private Sector’s Role in the Provision o f Infrastructure in Post-Conflict Countries: Patterns and Policy Options.” Social Development Papers: Conflict Prevention and Reconstruction Paper No. 16. World Bank, Washington, D.C . Verner, D. 2005. Making Poor Haitians Count, Labor Markets and Poverty in Rural and Urban Haiti, Based on the First Household Survey for Haiti, October 25, World Bank. World Bank. 1998. Haiti: The Challenge o f Poverty Reduction, Volume I. PREM. - Database. Various years. - Colunbia Public Expenditure Review. Report No. 25163-CO. April 16,2005. - Sierra Leone. Public Expenditure Review: From Post-Conflict Recovery to Sustained Growth. Report No. 29075-SL. August 31,2004. - Republic o f Niger. Public Expenditure Management and Financial Accountability Review (PEMFAR). Report No. 29752- NE. December 2004. - Improving the Dynamics o f Aid. Towards More Predictable Budget Support. - Haiti Country Economic Memorandum. Options and Opportunities for Inclusive Gro-wth. June 1,2006. - Development Policy Grant. Economic Governance Reform Operation 11. December 12, 2006. Report No. 38325-HT. - Assistance to the Republic o f Haiti Under the Enhanced HIPC Debt Initiative. October 25, 2006. Report No. 37394-HT. - Social Resilience and State Fragility in Haiti. A Country Social Analysis. April 2006. 165 Haiti: Public Expenditure Management and Financial Accountability Review 166 Haiti: Public Expenditure Manaqement and Financial Accountability Review 167 Haiti: Public Expenditure Manaqement and Financial Accountability Review 168 Haiti: Public Expenditure Management and Financial Accountability Review ANNEX 1 Table A1.1. VAT in Selected Countries Bolivia 14.9 5.7 25 Brazil 20.5 8.6 34 Chile 18.0 8.5 17 Colombia 15.0 4.9 35 Costa Rica 15.0 6.5 30 Dominican Republic 8.0 3.1 25 Ecuador 12.0 4.4 36.2 El Salvador 13.0 5.3 25 Guatemala 10.0 3.7 31 Haiti 10.0 2.2 35 Honduras 12.0 3.8 25 Mexico 15.0 3.2 33 Nicaragua 15.0 10.0 Panama 5.0 2.0 30 Paraguay 10.0 4.5 30 Peru 18.0 6.3 30 Uruguay 23.0 8.4 35 Venezuela 15.0 3.2 34 Average (LAC) 14.3 5.3 Source: Keen, et al. (20011, KPMG (2004), and IMF (2005) Table A1.2. Average Annual GDP growth between 1995 and 2004, H a i t i compared to selected Sub- Saharan Africa countries. Haiti 0.9 Guinea 4.0 Rwanda 10.0 Gambia 3.9 Mozambique 8.2 Seychelles 3.5 Uganda 6.5 Niger 3.5 Mali 6.1 Togo 3.4 Cape Verde 6.0 Namibia 3.3 Botswana 5.8 Swaziland 3.1 Benin 5.3 Malawi 3.0 Mauritius 5.2 Senegal 5.0 Tanzania 4.8 Cameroon 4.7 Ethiopia 4.6 Burkina Faso 4.5 Ghana 4.3 Mauritania 4.2 Sources: Government Statistics, World Bank Database, African Action Plan, and Staff calculations. 169 Haiti: Public Expendi'ture Management and Financial Accountability Review Current revenue 8.1 7.6 8.3 9.0 8.9 9.7 10.0 8.3 9.6 Domestic taxes 5.6 5.3 5.9 6.2 6.2 6.3 6.4 5.8 6.3 Customs duties 2.1 2.1 2.2 2.3 2.5 2.5 3.0 2.2 2.7 Other current revenue 0.3 0.3 0.1 0.4 0.1 0.8 0.6 0.3 0.5 Transfers from public enterprises ... ... ... ... ... ... 0.0 ... ... Grants 0.3 0.4 0.1 0.1 1.3 3.5 3.8 0.2 3.0 Budget support ... ... ... 0.0 0.3 1.4 1.0 ... 0.9 I Project grants ... ... ... ... 1.0 2.0 2.9 ... 2.1 Sources: Government statistics, IMF, and Staff calculations. T a b l e A1.5. Shares of r e c u r r e n t spending categories in t o t a l r e c u r r e n t , FY2000-06 (in p e r c e n t unless otherwise indicated) A \ erages FY FY FY FY FY FY FY FY FY 2000 2001 2001 2003 2004 2005 2006 2000-03 2004 06 E Wages and 56.0 47.4 39.3 34.6 37.4 35.3 34.1 44.3 35.6 salaries Net operations 28.2 36.7 45.2 43.4 40.7 26.6 27.7 38.4 31.6 Interest 10.8 10.7 8.2 9.4 10.5 7.0 8.9 8.8 payments Transfers and 5.0 5.2 8.7 12.5 11.4 31.1 29.3 23.9 subsidies Totalrecurrent 100 100 100 100 100 100 100 100 Sources: Government statistics, IMF, and Staff calculations. T a b l e A1.6. Composition of G o v e r n m e n t Expenditure. in P e r c e n t o f GDP, FY 2000-06 Averages FY FY FY FY FY FY FY F~ FY ZOO0 2001 2002 2003 2004 2005 2006" 2ooo-03 2o0406 Total expenditure 10.1 10.2 11.5 12.6 11.4 13.7 14.8 11.3 13.5 Current expenditure 7.5 8.3 9.4 9.3 7.9 9.6 9.5 8.7 9.1 Wages and salaries 4.2 4.0 3.7 3.2 2.9 3.4 3.2 3.7 3.2 Net operations' 2.1 3.1 4.3 4.0 3.2 2.6 2.6 3.5 2.8 Interest payments 0.8 0.9 0.8 0.9 0.8 0.7 0.8 0.8 0.8 External 0.4 0.5 0.4 0.5 0.5 0.4 0.4 0.4 0.4 Domestic 0.4 0.4 0.4 0.4 0.3 0.3 0.4 0.4 0.4 Transfersand subsidies 0.4 0.4 0.8 1.2 0.9 3.0 2.8 0.7 2.3 Capital expenditure 2.7 1.8 2.0 3.3 3.5 4.1 5.3 2.5 4.4 Domestically financed ... ... ... ... 2.2 1.3 1.0 ... 1.4 Foreign financed 1.3 2.9 4.4 ... 3.0 Iurces: Government statistics, IMF, and Staff calculations. 1 Includes statistic discrepancy. 171 Haiti: Public Expenditure Management and Financial Accountability Review 1 Average PY FY F\' FY PY Fj' F\ FY FY 2000 2001 2002 2003 2004 2005 2006* 2ooo-03 2o04-06 Overall Balance Including grants -1,699 -2,070 -2,849 -4,167 -3,441 -1,041 -1815 -2,696.3 -881 Excluding grants -1,964 -2,440 -2,962 -4,338 -5,265 -6,841 -9525 -2,926.0 -7,210.3 Excluding grants and NA NA NA NA -1671 -2036 -777 NA -1,494.7 externally financed projects Domestic financing 1,974 2,265 2,561 3,470 2,825 -9 -695 2,567.5 707 BRH 1,935 2,248 2,897 3,720 2,838 -21 -333 2,700 828 Banks -94 7 -103 -108 -32 -29 -290 -74.5 -1 17 O.W. Other nonbank -11 -9 0 0 0 0 -120 -5.0 -40.0 financing Arrears (net) 144 19 -233 -142 18 41 48 -53.0 35.7 External financing -274 -195 288 697 592 1,050 2,509 129 1,383.7 Loans 139 50 102 1,583 897 4,010 3,729 468.5 2,878.7 Amortization -429 -531 -584 -953 -1,136 -1,274 -1,606 -6,24.3 -1,338.7 Sources: Government of Haiti, IMF and Staff Calculations. 172 Haiti: Public Expenditure Management and Financial Accountability Review Appendix 1.1 Creating Fiscal Space: Old Wine in New Bottles? Fiscal space can be defined as “the availability o f budgetary r o o m that allows a government to provide resources for a desired purpose without any prejudice t o the sustainability o f a government’s financial position.”’22 There has been much debate in recent years o n h o w countries can create fiscal space, to help them achieve the Millennium Development Goals (MDGs). While much o f the early discussion was focused on public investment o n inhastructure (given the associated large externalities, as discussed in B o x l), i t has been recognized that some components o f current spending (such as spending o n health and education, for instance) may have significant effects on growth as well. To the extent that governments borrow in the short run to finance productive spending, higher growth may w e l l raise the tax revenues needed t o repay in the longer run. I n a sense, therefore, higher public spending may eventually “pay for itself,” creating thereby dynamic trade-offs among various components o f expenditure. H o w can low-income countries create fiscal space? Essentially, there are three ways to make additional resources available for productive government spending, without compromising the sustainability o f public finances in the medium term: through improvement in expenditure efficiency and reallocation o f spending, through higher tax revenues, and through greater domestic or foreign financing at relatively l o w cost. Regarding expenditure efficiency and reallocation, a first option i s t o create fiscal space by reducing waste in public spending, particularly when it comes t o public investment. This can be achieved by improving governance in a broad sense, that is, with tighter management and greater accountability. Another option i s t o reduce “unproductive” expenditures (or those considered to n doing so, have a lower priority) and reallocate available resources to more productive outlays. I however, care must be taken in classifying expenditures as “unproductive.” In an environment where crime and violence deter individuals from saving arid investing, for instance, spending on security should retain a high priority-despite the fact that i t may not appear prima facie to be directly productive. Similarly, attempting to create fiscal space by cutting spending on maintenance may be an ill-advised strategy; i t may “work” in the short term (in the sense that consequences for the quality o f infrastructure may not be immediate), but only at a greater economic cost in the long term. Regarding tax revenues, it i s possible that by broadening the tax base and strengthening tax administration (through, for instance, a reduction in collection costs and measures to reduce tax evasion) additional revenues can be raised t o finance productive spending. In many poor countries, however, the capacity t o raise resources through taxation remains limited by the very fact that incomes are l o w to begin with; and if the components o f spending that are being contemplated for expansion require large or lumpy investments (as i s often the case for infrastructure), relying solely o n higher taxation to create fiscal space may be unfeasible. Regarding domestic or foreign financing, resources can be borrowed or received in the f o r m o f foreign grants. In many poor countries, the capacity to borrow domestically i s often limited by the size o f the financial sector.’23 As for foreign borrowing, unless i t i s at highly concessional terms, i t may be a source o f concern because o f its implications for the sustainability o f public debt. Debt relief, by contrast, hees resources that were previously earmarked for debt service ‘**Peter S. Heller, “Understanding Fiscal Space,” IMF Policy Discussion Paper No. 05/4 (March 2005), page 3. See also Peter S. Heller, Menachem Katz, Xavier Debrun, The0 Thomas, T a k e Koranchelian, and Isabel Adenauer, “Making Fiscal Space Happen!” World Economics, Vol. 7 (July-Sept. 2006), 89-132. ‘23The degree of financial development, together with other institutional factors (such as the nature o f the country’s exchange rate arrangement) also limits the ability to raise revenues through seigniorage. 173 Haiti: Public Expenditure Management and Financial Accountability Review operations, and may in principle provide a significant boost to outlays in some specific productive spending c a t e g o r i e ~ . ’ ~ ~ An alternative financing option i s foreign grants. What has become clear in recent years, however, i s that grants tend to display a high degree o f volatility.’25 This i s particularly problematic if a country must finance a multi-year public investment program that requires (due t o the lumpy nature o f some outlays, for instance) a sustained i n f l o w o f resources; an unexpected shortfall could derail the program and annihilate i t s potential effects o n growth and future tax revenues.’26 Another potential problem with foreign aid (in addition t o short-term D u t c h disease effects) i s an adverse moral hazard problem-the fact that a i d m a y weaken revenue performance and lead t o higher spending. If so, then the fiscal space initially created by higher aid flows m a y vanish fairly quickly. There are two main lessons that can b e drawn from the foregoing discussion. The f i r s t i s that most o f the issues that arise in discussing h o w “fiscal space” can b e created are not new; they a l l b o i l down, to a significant extent, o n ways t o allow governments to expand expenditure, without jeopardizing fiscal sustainability. This suggests that considerations related t o fiscal space are best made in the context o f an explicit medium-term budget framework, with a clear outline o f the government’s expenditure priorities-particularly in the area o f public investment-and available financing options. In that sense, the current debate i s “old wine in n e w bottles.” M o r e specifically, there are two reasons why a medium-tern perspective i s essential. The first i s because o f the need to assess the impact o f the higher public expenditure o n the economy’s growth rate and the country’s capacity t o generate the revenue needed t o service any debt that was occurred in the short term. The second i s that, to the extent that higher expenditure in the short term leads to higher future recurrent expenditures (such as maintenance spending, in the case o f core infrastructure investment, for instance), one must assess if these future outlays can b e financed from future revenues. Put differently, ensuring public debt sustainability implies that the spending programs f o r which fiscal space i s created must be evaluated not only in terms o f their short-term impact o n the budget, but also in terms o f their medium-term effect as well. The need t o provide future budgetary resources may magnify requirements t o create fiscal space. The second important lesson that can b e drawn f r o m the foregoing analysis i s that assessing ways to create fiscal space, while ensuring sustainability o f the fiscal stance, i s inherently country specific and requires the use o f a quantitative model that accounts for the country’s k e y structural economic features, i t s revenue and expenditure structure, the composition o f its debt and its cash f l o w implications, and i t s prospects for greater access t o foreign resources. Such models are essential to discuss h o w a reallocation o f spending from “unproductive” outlays to investment in health and education, for instance, m a y affect growth and the budget, and whether “true” fiscal space can be achieved. As discussed elsewhere in this report, issues o f efficiency and strengthening o f fiscal institutions can also be framed, at least in part, in the context o f these models. Iz4Seefor instance John Weeks and Terry McKinley, “Does Debt R e l i e f Increase Fiscal Space in Zambia? The MDG Implications,” Country Study No. 5, International Poverty Centre (September 2006), for a case study o f the impact (or lack thereof) of debt relief o n fiscal space. lZ5SeeAles Bulk and A. Javier Hamann, “Volatility o f Development Aid: F r o m the Frying Pan into the Fire?,” Working Paper No. 06/65, International Monetary Fund (March 2006), and UNCTAD, Doubling Aid: Making the Big Push Work, Geneva, 2006. Iz6In fact, as argued for instance by Pierre-Richar Agenor and Joshua Aizenman (“Aid Volatility and Poverty traps”, unpublished, University o f Manchester, M a y 2007) a high degree o f a i d volatility m a y deter governments f r o m implementing (costly) reforms aimed at raising domestic resources, and m a y well lead to a poverty trap. 174 Haiti: Public Expenditure Manaqement and Financial Accountability Review ANNEX 2 INSTITUTIONS MIN PUB HEALTH & POP Source : L e Moniteur, Journal Officiel de la Rkpublique d’Haili and Staff Calculations. 175 Haiti: Public ExDenditure Manaaement and Financial Accountabilitv Review Table A2.2 Budget Allocation Trends (million of nominal US$) ' MIN FEMALE CONDITION .9 .4 .4 .9 .9 2.3 .6 1.4 t ' MIN YOUTH, SPORTS & SOCIAL ACTION 7.0 CULTURAL SECTOR '7.0 . ' 6.9 6.0 10.4 30.1 13.9 6.6 11.5 MIN o f WORSHIP .a 1.6 1.8 1.9 1.7 1.9 1.4 1.a MIN o f CULTURE 6.2 5.4 4.2 8.5 8.4 12.0 5.2 9.6 ELECTORAL COUNCIL 2.2 .a 1.4 1.5 1.6 10.0 1.5 4.4 CITIZEN SAFETY OFFICE .2 .3 .1 .2 .2 .2 .2 .2 STATE UNIVERSITY o f HAITI 6.6 6.2 4.7 7.1 7.5 9.5 5.8 8.0 5. PUBLIC DEBT AMORTIZATION 42.5 TOTAL 490.3 379.0 475.1 600.0 900.4 1537.3 448.2 1012.6 176 Haiti: Public Expenditure Management and Financial Accountability Review Table A2.3. Total Spending by Broad Categories (in Million of gourdes), FY2002-06 FY FY FY FY FY Average 200Il02 200210.3 2003104 200-1105 2005106 F’ FY 2002f04 2005f06 TOTAL 18,080.53 19,606.78 17,159.32 18,563.02 17,278.60 18,282.21 17,920.81 1. EXECUTIVE POWER 17,064.31 18,481.11 16,468.15 17,867.16 16,491.75 17,337.86 17,179.46 ECONOMIC SECTOR 2,805.49 3,828.16 2,136.29 3,083.68 3,470.49 2,923.31 3,277.08 POLITICAL SECTOR 6,231.38 6,105.58 3,965.19 3,798.89 3,984.48 5,434.05 3,891.69 SOCIAL SECTOR 4,308.58 4,735.61 2,390.90 3,376.29 3,331.41 3,811.70 3,353.85 CULTURAL SECTOR 282.14 489.54 210.51 393.17 303.29 327.40 348.23 OTHER PUBS ADMINISTRATIONS 3,436.72 3,322.22 7,765.27 7,215.12 5,402.10 4,841.40 6,308.61 2- LEGISLATIVE POWER 437.07 539.18 204.95 74.35 210.96 393.73 142.65 3- JUDICIARY POWER 246.70 238.21 155.70 213.53 197.23 213.54 205.38 4- INDEPENDENT ORGANISMS 332.45 348.28 330.52 407.98 378.65 337.08 393.32 Sources: L e Moniteur, Journal Officiel de la Rkpublique d’Haili and Staff Calculations. Table A2.4. FunctionalDistribution of Actual Total Expenditures-Payment basis (in million of Gourdes unless otherwise indicated) FY PY FY FY FY Average tOOllO2 2002103 200.3104 2004105 2005106 FY FY 2002-04 2005-06 Aagriculture, Natural Res. 258.8 324.6 303.9 372.7 360.3 295.8 366.5 and Rural Development I n millions of nominal US$ 9.6 80 7.7 9.6 8.6 84 P I I n millions of real gourdes 446.6 442 0 322.5 338.7 286.0 403 7 3123 In %o f GDP 0.3% 0 3% 0.3 96 0.3% 0.2% 0 -2% 0 2% Public Works, Transport 686.4 1,396.70 841.7 1,268.90 1,576.30 974.9 1422,6 and Communications I n millions of nominal US$ 25.3 34 5 21.2 32.5 37.5 27 0 35 0 I n millions of real gourdes 1184.4 1901 7 893.I I 153.2 1251.1 13264 12021 In %o f GDP 0.8% 15% 0.7% 0.9% 0.9% 10% 0 9% Justice and Security 1,087.90 1,496.20 1,131.50 1,991.30 2,365.70 1,238.5 2,178.5 I n millions of nominal US$ 40.2 37 0 28.5 -71.1 56.3 35 2 53 7 I n millions of real gourdes 1877.2 2037 1 1200.6 1809 7 1877.6 17050 18437 In %o f GDP 1.3% 16% 0.9% 1.4% 1.4% 13% 14% Education (Incl. University 1,679.50 2,221.50 1,619.40 2,904.50 2,466.30 1,840.1 2,685.4 of Haiti) I n millions of nominal US$ 62.0 54 9 40 8 74.5 58.7 52 6 66 6 I n millions of real gourdes 2,898.0 3024 7 I,718.3 2,639.6 1,957.5 2,547 0 2,298 -7 I n %of GDP 2.0% 2 4% 1.4% 2.1% 1.5% 19% 18% Health 771.8 930.3 655.3 890.4 1,003.20 785.8 946.8 I n millions of nominal US$ 28.5 23 0 16.5 22.8 23.9 22 7 23 4 I n millions of real gourdes 1,331.8 1,266 6 695.3 809.2 796.2 1,0979 802 7 In %o f GDP 0.9% 10% 0.5% 0.6% 0.6% 0 8% 0 6% Total priority sectors 4,484.40 6,369.40 4,551.90 7,427.70 8,571.40 5,135.2 7,999.6 I n millions of nominal US$ 165.6 1574 114.7 190.5 204.1 14-79 1973 I n millions of real gourdes 7,738.0 8,672 2 4,829.9 6,750.2 6,803.I 7,080 0 6,776 6 In %o f GDP 5 2% 6 8% 3.8% 5.3% 5.1% 5 3% 5 2% Total other sector 5,993.90 8,031.1 0 11,619.70 12,998.40 13,198.50 8,548.2 13,098.5 I n millions of nominal US$ 221.3 198 5 292.8 333.4 314.3 2376 3238 I n millions of real gourdes 10,342.7 10,934 6 12,329.4 11,812.8 10,475.6 11,202 2 11,144 2 I n %ofGDP 7.0% 8 5% 9.7% 9.3% 7.9% 8 4% 8 6% TOTAL SPENDING 10,478.30 14,400.50 16,171.60 20,426.10 21,769.90 1,3683.5 21,098.0 I n millions ofnominal US$ 386.9 356.0 407.5 524.0 518.3 383.5 521.1 I n millions of real gourdes 18.080.6 19.606.8 I 7.1 59.3 18.563.0 1 7.2 78.6 18.282.3 17.920.8 I n % o f GDP 12.2% 15.3% 13.5% 14.5% 13.0% 13.7% 13.8% Source: L e Moniteur, Journal Officiel de la Republiqued’Ha’iti and Staff Calculations. 177 Haiti: Public Expenditure Manaqement and Financial Accountability Review Table A 2 5 FunctionalDistribution of Actual Public Recurrent Expenditures-Payment basis (million of HTG) FY FY FY FY FY Averages 2001i02 2002103 2003t04 2004f05 2005106 FY FY 2002-04 2005-06 AGRICULTURE, NATURAL R E S . AND 176.9 217.8 209.5 286.3 289.5 201.4 287 9 RURAL DEVELOPMENT I n millions ofnominal US$ 6.5 54 53 73 6.9 5.7 7.I I n millions of real gourdes 305.2 296 5 222 3 260 2 229.8 274.7 245.0 %o f GDP 0.2% 0 2% 0 2% 0 2% 0.2% 0.2% 0.2% PUBLIC WORKS, TRANSPORT AND 198.8 276 171 5 314.1 404.8 215.5 359.4 COMM~JNICATIONS I n millions ofnominal US$ 7.3 68 43 81 9.6 6.2 8.8 I n millions of real gourdes 343.0 375 8 182 0 285 5 321.3 300.3 303.4 %o f GDP 0.2% 0 3% 0 1% 0 2% 0.2% 0.2% 0.2% JUSTICEAND SECURITY 905.2 1,234.20 1,026.40 1,841 60 2,323.80 1,055.30 2,082.70 I n millions of nominal US$ 33.4 30 5 25 9 47 2 55.3 29.9 51.3 I n millions of real gourdes 1561.9 16804 1089 I 1673 6 1844.4 1443.8 1759.0 %o f GDP 1.1% 13% 0 9% 13% 1.4% 1.1% 1.3% EDUCATION (INCL. UNIVERSITY OF 1,520.3 1,997.8 1,455.8 2,737.1 3,124.5 1,657.9 2,930.30 HAITI) I n millions of nominal US$ 56.I 49 4 36 7 70 2 74.4 47.4 72.3 I n millions of real gourdes 2623.3 2720 1 1544 7 2487 4 2479.9 22960 2483.7 %o f GDP 1.8% 2 1% 12% 19% 1.9% 1.7% 1.9% HEALTH 686.8 790 3 568.2 798.8 930.2 681.8 864.5 I n millions of nominal [IS$ 25.4 19.5 14 3 20 5 22.1 19.7 21 3 I n millions of real gourdes 1185.1 1076 0 602 9 725 9 738.3 954.7 732.1 %of GDP 0.8% 0 8% 0 5% 0 6% 0.6% 0.7% 0 690 TOTAL PRIORITY SECTORS 3,488.1 4516.2 3,431.5 5,976.9 7,072.7 3,811.9 6,524.80 I n millions of nominal US$ 128.8 111.6 86.5 153.3 168.4 109.0 160.9 n millions o I f real gourdes 6018.8 6149.0 3641.1 5431.7 5613.6 5269.6 5522.6 %oj GDP 4.1% 4.8% 2.9% 4.3% 4.2% 3.9% 4.2% OTHER SECTORS 5,079.6 5,857.1 9,491.6 7,915.7 10,826.6 6,809.4 9,371.28 I n millions of nominal US$ 187.6 144 8 239 2 203 0 257.8 190.5 230.4 I n milhons of real gourdes 8765.0 7974 7’ 10071 3 7193 7 8593.0 8937.0 78933 % of GDP 5.9% 6 2% 7 9% 5 6% 6.4% 6.7% 6.0% TOTAL RECURRENT SPENDING ALL 8,567.6 10,373.4 12,923.1 13,892.7 17,899.3 10,621.4 15,895.90 SECTORS I n millions of nominal US$ 316.4 256 4 325 7 356 4 426.2 299.5 391.3 I n millions of real gourdes 14783.7 14123 8 13712 4 12625 5 14206.5 14206.6 13416.0 % of GDP 10.0% I1 0% 10.8% 9 9% I O . 7% 10.6% 10.3% TOTAL RECURRENT SPENDING (Exel. 8,330.0 10,201.6 11,725.6 12,806.9 16,655.2 10,085.7 14,731.10 Interest Payments of the debt) I n millions of nominal US$ 307.6 252 2 295 5 328 5 396.6 285.1 362.5 I n millions of real gourdes 14373.7 13889 9 12441 8 11638 8 13219.1 13568.4 12428.9 % of GDP 9.7% 10 8% 9 8% 9 I% 9.9% 10.1% 9.5% SHARE OF PRIORITY SECTORS I N 41.9 44.3 29.3 46.7 42.5 37.8 44.3 TOTAL RECURRENT SPENDING (ExCl. Interest Payments of the debt) Source: Le Moniteur, Journal Offcief de la RBpublique d’HaYti and Staff Calculations. 178 Haiti: Public Expenditure Management and Financial Accountability Review Table A2.6. Execution Rates in Percent, FY2002-06 FY FY FY FY FY A~crages 2001l02 2002f03 2003104 2004105 2005106 FY FY 2002104 2005/06 Economic 41.6 05.7 109.0 69.3 24.1 71.2 33.7 sector Political 141.4 155.2 111.6 92.4 99.2 134.6 96.0 sector Social 62.2 125.0 69.8 90.4 78.7 82.2 83.8 sector 1 Total 80.5 93.9 97.2 87.3 57.6 91.3 68.9 Sources: Government Statistics and Staff Calculations. 179 o9 - m9w- o r- ooq= 0 ? .- a 0 0 0 0 Q CY LA- 009- oc: I a2 Q W O r- r- e- O t ? ? c!oor+o Y W N O 0 o i = 0 I m ZOZZg N LAwr- ooas 00 Y Nr- O ~ 6 0~ O - 0 m O N .- c - r l i .. .- + .- 5 Haiti: Public Expenditure Manaqement and Financial Accountability Review ANNEX 3 Table A3.1. PEFA Indicators Indicator I Score I Comments A. PFM-OUT-TURNS: Credibility o f the budget PI-1:Aggregate expenditure out-turn Based o n domestically-financed primary expenditure, deviations compared t o original approved budget between original budget and outturns were: 2.8% i n 200312004; 12.7% in 2004l2005 and 42.4% in 200512006. Thus expenditures deviated only in 200512006 for more than 15% f i o m original budget. PI-2: Composition of expenditure out-turn Based o n domestically-financed primary expenditure, deviations compared t o original approved budget between original budget and outturns were: 32.6% in 200312004, 9.1% in 200412005 and 1% in 200512006. Thus variance in expenditure composition exceeded overall deviation inprimary expenditure by 10 percentage points in no more than one o f the last three years PI-3: Aggregate revenue out-turn Actual revenue collections as a percentage o f budgeted domestic compared to Original approved budget revenxe were: 63% in 200312004; 57% in 2004/2005 andc148% in 200512@06. Thus Actual domestic revenue collection was below 92% o f budgeted domestic revenue estimates in two ofthe last three years. PI-4: Stock and monitoring of expenditure D+ payment arrears (i)Stock o f expenditure payment arrears (as a C T h e accounting system used by the Directorate o f the Treasury i s a percentage o f actual total expenditure for the single-entry system that does not take accruals into account. I t i s corresponding fiscal year) and any recent therefore not possible, under this system, to establish a treasury change in the stock. balance, and it does not identify payment arrears. Many expenditures-around 80 percent-are made according to a reverse procedure, i.e., before delivery o f goods and services as stated above. T h i s type o f practice obviously l i m i t s the amount o f arrears. Nonetheless, a large number o f arrears, totaling a fairly substantial sum, have accumulated over the past few years upstream o f the payment authorization [ordonnuncement] phase (e.g., commitments made for unauthorized expenditures, or even expenditures made ( ii) Availability o f data for monitoring the D Data o n the stock of arrears has been generated only recently and i s stock o f expenditure payment arrears. s t i l l incomplete. 181 Haiti: Public Expenditure Manaqement and Financial Accountability Review Indicator I Score I Comments B. KEY CROSS-CUTTING ISSUES: Comprehensivenessand Transparency PI-5 “Budget classification” B Budget classification has improved over the past few years. The classification now used consists o f institutional, sectoral, administrative, and economic (i.e., by nature o f expenditure and revenue) classifications. PI-6 “Comprehensiveness of information C The annual budget document contains information on: (i) included in budget documentation” macroeconomic assumptions; ( iithe budget deficit; ( ) iii) funding o f the deficit; and (iv) an explanation o f the impact o f any new budget initiatives by the public authorities. PI-7: Extent of unreported government operations (1) The level o f extra-budgetary expenditure (other than donor fmded projects) which i s unreported Le. not included in fiscal reports. (ii) Incomeiexpenditure information o n donor- hnded projects which i s included in fiscal reports. PI-8: Transparency of inter-governmental fiscal relations (i) Transparent and rules based systems in the horizontal allocation among S N governments of unconditional and conditional transfers from central government (both budgeted and actual allocations) ( , ii I )Timeliness o f reliable information to S N governments on their allocations from central government for the coming year (iii) Extent to which consolidated fiscal data (at least o n revenue and expenditure) i s collected and reported for general government according to sectoral categories. PI-9: Oversight of aggregate fiscal risk from other public sector entities (i) Extent o f central government monitoring o f NR D I N o annual monitoring o f Autonomous Agencies takes place. AGAs and PES. Autonomous Agencies are not producing financial reports o n a regular basis and are not submitting their reports as required by the law to the MEF trough the State Enterprises Commission. (ii) Extent o f central government monitoring o f S N governments’ fiscal position. PI-10: Public Access to key fiscal Information NR C II As described in the table 18 bellow, the government makes available 182 Haiti: Public Expenditure Management and Financial Accountability Review Indicator I Score I Comments I I to the Dublic onlv one o f the listed t w e o f information. 183 Haiti: Public Expenditure Manaqement and Financial Accountability Review Indicator I Score I Comments C. BUDGET CYCLE C(i) Policy-Based Budgeting PI-11: Orderliness and participation in the B annual budget process Existence o f and adherence t o a fixed budget B n its A clear annual budget timetable exists, but delays often occur i calendar implementation. The timetable gives the sectoral ministries enough time (two and one-half months) t o draw up their detailed estimates in an adequate and timely manner. Clarity/comprehensiveness o f and political A circular i s issued by the Prime Minister. I t informs the ministers o f involvement in the guidance o n the preparation the broad outlines o f the draft budget as w e l l as the macroeconomic o f budget submissions (budget circular or framework and the broad budget outlook. But it does n o t specify a equivalent) ceiling envelope for each MDA. Timely budget approval by the legislature o r Over the past three years, only one Annual Budget Act. (2003/2004) similarly mandated b o d y (within the last three was voted after the beginning o f the fiscal year. years) PI-12: Multi-year perspective in fiscal planning, expenditure policy and budgeting (i> Preparation o f multi -year fiscal forecasts There are n o medium-term perspectives. The Public Investment and functional allocations Program (PIP) covers one year only. (ii) Scope and frequency o f debt sustainability There hasn't been any debt sustainability analysis during the last analysis three years. (iii) Existence o f sector strategies with multi- Some MDA (Education, Health, Public Transportation, year costing o f recurrent and investment Environment, and Agriculture) are working o n the elaboration of expenditure; their strategies. But this i s at a n early stage and they haven't been costed. (iv) Linkages between investment budgets and D The investment and recurrent budgets are elaborated separately. A s a forward expenditure estimates. result, recurrent costs estimates arising f r o m the investments are n o t taken into account. PI-13: Transparency o f Taxpayer B Obligations and Liabilities (i)Clarity and comprehensiveness o f tax Regulations concerning customs and excise taxes are being liabilities reformed. Codes are being developed to incorporate a l l o f the content o f recent reforms. Legislation and procedures pertaining t o a l l the m a i n categories o f customs and excise taxes might b e considered. ( ii) Taxpayer access t o information o n tax Taxpayers have access to some information. DGI has made available liabilities and administrative procedures on the W e b site tax liabilities and procedures. F o r the customs Authority, access t o information i s limited t o taxpayers involved in the customs process. Information published o n the W e b site. (iii) Existence and fimctioning o f a tax appeals B Taxpayers might also seek redress through the administrative mechanism tribunal represented by the CSCCA. They also have the option o f amicable recourse t o the AGD and T a x Directorates. However, the efficiency o f this system has n o t been verified. 184 Haiti: Public Expenditure Manaqement and Financial Accountability Review Indicator 1 Score Comments taxpayer registration and tax assessment Controls i (i) n the taxpayer registration system C A unique Taxpayer Identification Number (TIN) exists. The f i l e containing this information i s linked to taxpayer rolls. There i s a commission responsible for checking registrations, but it i s not operational due to the lack o f capacity. (ii) Effectiveness o f penalties for non- C Penalties for non compliance exist in the law but are not enforced due to the lack o f capacity o f RA as well as the recent political troubles. There i s a need o f substantial changes to give them a real impact or ensure compliance. T h e revenue agencies have annual plans o n tax audit. However, fraud investigation programs these audits are not based o n clear r i s k assessment criteria. The audit plans are not comprehensive and well documented. payments I (i) Collection ratio for gross tax arrears, being I C The concept o f arrears does not exist at AGD, since it i s not possible the percentage o f tax arrears at the beginning to make any deferred customs payment arrangements. W i t h regards o f a fiscal year, which was collected during to DGI, the Large Enterprise Management Unit achieves recovery that fiscal year (average o f the last two fiscal rates o f about 85 percent, but the average overall recovery over the vast two vears has been about 60 vercent. Revenue collections are transferred daily to the Th-easury Account in Port AILPrince whereas, in the provinces, it i s done monthly. Reconciliations o f tax transferred are done monthly between the RA assessments. and DT. However reconciliation o f tax assessments, collections and and receipts by the arrears are done less frequently (more than 3 months delay). (i)Extent to which cash flows are forecast and D Cash Flow planning and monitoring are not undertaken monitored ( ii) Reliability and horizon o f periodic in-year D According to the Constitution, ceiling for expenditures commitment information to MDAs on ceilings for is limited to one twelfth o f the annual budget allocation. expenditure commitment Authorization o f expenditures commitment in SYSDEP i s thus limited to one twelve. However, in practice, the system can be changed. (iii) Frequency and transparency o f C A Supplementary Budget A c t i s not voted regularly. However, few adjustments to budget allocations, which are adjustments on budget allocations were made except for this fiscal decided above the level o f management o f year. MDAs PI-17: Recording and management of cash D+ balances, debt and guarantees (i) Quality o f debt data recording and reporting D Debt data records are incomplete, particularly domestic debt. The government i s making e n effort to identify salary arrears. (ii) Extent o f consolidation of the D T h e Treasury cannot draw up a report o n the cash status o f government’s cash balances govemment accounts because it does not receive all the data it would need to ensure reconciliation o f the Government’s bank accounts, an essential aspect o f the reporting o n funds encashed and their utilization. (iii) Systems for contracting loans and issuance C Central Government contracting o f loans and issuance o f debt and o f guarantees guarantees are always approved by the Ministry o f Economy and Finance and submitted to the varliament. However. there isn’t an 185 Haiti: Public Expenditure Manasement and Financial Accountability Review Indicator Score Comments assessment o f the sustainability f the debt. PI-18: Effectiveness o f payroll statements D+ (i) Degree o f integration and reconciliation T h e Directorate o f the Treasury (DT) calculates the monthly pay o f between personnel records and payroll data. the Government’s 45,000 civil servants and issues salary statements. T h e DT uses a single computerized database that theoretically prevents duplicate payments. I t should be pointed out t (ii) Timeliness o f changes to personnel records The personnel roster i s updated a long time after hiring and changes and the payroll in the personal and administrative status o f employees have occurred. ( iii) Internal controls o f changes to personnel D Measures in place to monitor the wage bill are inadequate. Indeed, records and the payroll. even though appropriations are a limiting factor, the ministies often take hiring initiatives that exceed budget authorizations. Payment arrears have thus accumulated, and are currently (iv) Existence o f payroll audits to identify C Payment verification i s performed on a monthly basis by the control weaknesses and/or ghost workers. Oversight Directorate o f the DGB. This oversight i s inadequate, however. Instances have occurred o f payment to deceased employees and o f payment that failed to take promotions into account. PI-19: Cornpetition, value f o r money and NR controls in procurement PI-20: Effectiveness o f internal controls f o r C non-salary expenditure (i) Effectiveness o f expenditure commitment C Measures to supervise commitments are i nplace: the D C B exercises controls. ex ante controls over expenditures, and payment authorizations are issued by the MEF after being reviewed by the Financial Comptroller in the SYSDEP system. However, that these controls do ( ii) Comprehensiveness, relevance and C Other internal control procedures exist, but they are inadequately understanding o f other internal control rules/ documented and are neither widely disseminated among users nor procedures. thoroughly understood by them. ( iii) Degree o f compliance with rules for Procedures are respected with regard to expenditures introduced into processing and recording transactions the normal circuit. As indicated above, however, many expenditures are executed via exceptional procedures (e.g., through comptes courants or using “own hnds”.) PI-21:Effectiveness o f internal audit (i)Coverage and quality o f the internal audit The internal audit body, IGF was created very recently and is s t i l l function. not operational. Some M D A s have internal audit unit but they are basically duplicating the ex ante controls performed by D C B and don’t have professional standards, such as ISPPIA. Within (ii) Frequency and distribution o f reports. As a result o f the lack o f proper internal audit mechanism, there is no report produced. (iii) Extent o f management response to internal D There i s no action taken, i nthe absence o f internal audit reports. audit findings. PI-22: Timeliness and Regularity o f D Accounts reconciliations 186 Haiti: Public Expenditure Management and Financial Accountability Review Indicator Score Comments (i) Regularity o f banking reconciliations D Banking reconciliation i s not performed because the Treasury does not have the necessary banking information. The only kind o f monitoring i s a l i s t o f checks issued and receivable maintained at BRH. Reconciliation and Clearance o f suspense accounts and travel o f suspense accounts and advances. comptes courants maintained at DT take place at least quarterly, within a month from end of period and with few balances brought forward. However, reconciliations o f comptes courants located i nthe MDA As was mentioned in the assessment o f indicator noPI-7, substantial resources received by service delivery units funds are collected by service-providing u n i t s and kept by those units. These sums are used to cover the operating costs o f these units, and sometimes to pay staff. Over the past three PI-24: Quality and timeliness of in-year D+ budget reports (i)Scope o f reports in terms o f coverage and C Comparison to budget i s possible only for main administrative headings. Detail o f on-requisition expenditures as well as expenditures executed under comptes courants are not provided. T h e quarterly report i s summarized in one page. The in-year information In-year reports are produce with more than 8 weeks delay. As some o f the expenditures are executed outside the system there ase some issues with data accuracy. A consolidated Annual Financial Staternent is prepared for central goverament, covering MDAs. However, the information i s prepared by D T does not include some information o n revenues (own resources o f MDA) and expenditures comptes courants, other expendi (ii)Timeliness o f submission o f the financial D The annual financial statements are submitting with delays to the statements Auditor General. The table x shows the submission dates for the last three years. (iii) Accounting standards used C Statements are presented in consistent format over time with some disclosure o f national accounting standards, which are however different from the IPSAS. PI-26: Scope, nature, and follow-up o f D+ external audits (i)Coverage and quality o f the internal audit C Although the law requires that all govemment financial transactions function. be audited, only centra1 administration entities, which represent at least 50% o f total expenditures, were audited in reports pertaining to FY02-03 and FY03-04 (draft). The reports raise (ii) Frequency and distribution o f reports. - D Audit reports are submitted to Parliament beyond the 12-month after the end o f the fiscal period under review. For example, the most recent report available (FY02-03) has not yet been submitted to Parliament. (iii) Extent o f management response to internal - D There i s very l i t t l e proof that recommendations have been audit findings. implemented, or documentation o f responses from the entities audited. 187 Haiti: Public Expenditure Management and Financial Accountability Review Indicator Score Comments PI-27: Examination o f annual appropriation c+ law by the legislative b r a n c h (i) Scope o f the legislature’s scrutiny. C The Parliament’s exercise o f its authority t o approve appropriation laws i s relatively recent. Parliamentary oversight covers the details o f expenditures and revenues, but only at the stage when detailed proposals have been finalized by the Government. (ii) Extent t o which the legislature’s B Simple procedures exist for the Parliament’s examination o f the procedures are well-established and respected. budget. These procedures are generally followed. ( iii)Adequacy o f time for the legislature to B Although the legal timeframe ( over two months, f r o m June 30 to the provide a response to budget proposals b o t h Znd Monday in September) i s adequate, substantial delays have the detailed estimates and, where applicable, occurred in practice. The situation with fiscal 2006/07 i s still an for proposals o n macro-fiscal aggregates exception, however, since the n e w Parliament h a d just bee earlier in the budget preparation cycle (time allowed i n.practice for a l l stag (iv) Rules for in-year amendments t o the B Procedures for preparation and approval o f Supplementary Budget budget without ex-ante approval by the A c t are set forth clearly in the Decree o f M a y 23,2005, but they do legislature. n o t impose limits o n administrative reallocations. PI-28: Legislative scrutiny of external a u d i t reDorts 188 Haiti: Public Expenditure Manaqement and Financial Accountability Review Indicator I Score I Comments D. DONOR PRACTICES D-1: Predictability of Direct Budget D Support (i)Annual deviation o f actual budget support D Donor projections o f budget support were more than 25% to actual from the forecast provided by the donor outturns for the last three years: 31% in 200312004, 29% in agencies at least six weeks prior to the 200412005 and 43% in 200512006. government submitting i t s budget proposals to the legislature (or equivalent approving body) ( ii) In-year timeliness o f donor D Data were not available to obtain the quarterly distribution o f actual disbursements budget support inflows compared to the plan. D-2: Financial information provided by NR donors for budgeting and reporting on project and program aid (i) Completeness and timeliness o f budget estimates b y donors for project support. (ii)Frequency and coverage o f reporting by donors o n actual donor flows for project support. D-3: Proportion of aid that i s managed by D i use of national procedures (i)Timeliness o f examination o f audit reports D Over the last three fiscal years n o audit reposts were reviewed by the by legislatwe (for reports received the last parliament, given the political circumstances. The newly elected three fiscal years) -- parliament was kastalled on last July- 2006. - ( ii) Extend o f hearings on key findings D As a result, no hearings were conducted. undertaken by legislature ( iii) Issuance o f recommended actions by D N o recommendations were issued by the legislature. T h i s i s expected legislature and implementation by the to change with the new parliament. Executive 189 Haiti: Public Expenditure Manaqement and Financial Accountability Review ANNEX 4 Table A4.1. Budgetary allocations (HTG millions, otherwise indicated) in % of GDP 1.0% 0.4% 0.2% 0.4% 0.7% 1.9% 0.5% 1.0% Total 13278.4 15333.3 16634.7 23390.0 37816.6 64566.7 15082.1 41924.4 in real H T G 22912.3 20876.8 17650.7 21256.6 30014.8 46202.5 20479.9 32491.3 in nominal US$ 490.3 379.0 419.2 600.0 900.4 1537.3 429.5 1012.6 in US$/cap 59.0 44.7 48.5 68.0 100.3 168.3 50.7 112.2 in % of GDP 15.5% 16.3% 13.9% 16.7% 22.5% 32.2% 15.2% 23.8% Agriculture Share 6.7% 2.3% 1.5% 2.2% 3.2% 5.9% 3.3% 4.4% Source : L e Moniteur, Journal Officiel de la Rkpublique d’HaYti and Staff Calculations. Table A4.2. Budgetary Allocations Structure (HTG million, unless otherwise indicated), FY2001/06 Averages FY FY FY FY FY FY FY Fl, 2001~02 2002103 200310.1 2004105 2005106 2006107 01,03 04106 Total allocated 887.0 348.8 251.1 507.6 1208.6 3777.6 495.6 1831.3 in real HTG 1530.6 474.9 266.4 46 1.3 959.3 2703.1 75 7.3 1374.4 in nominal IS$ 32.8 8.6 63 13 0 28.8 89.9 15.9 43.9 Personal 168.4 148.8 174.13 246.1 252.1 304.8 163.8 267.7 in real HTG 290.5 202.6 185.0 223.7 200.1 218.1 226.0 214.0 in nominal US$ 6.2 3.7 44 6.3 6.0 7.3 4.8 6.5 % 19.0% 42.7% 69.4% . 48.5% 20.9% 8.1% 43.7% 25.8% Other current exp. 58.3 51.2 43.2 57.0 51.3 56.1 50.9 54.8 in real H T G 100.6 69.8 45 8 51.8 40.7 40.1 72.1 44.2 in nominal US$ 2.2 1.3 I1 1.5 1.2 1.3 1.5 1.3 Immob. 27.5 8.6 1.5 10.2 27.5 6.8 in real HTG 47.5 .o .o 7.8 1.2 7.3 15.8 5.4 in nominal US$ 1.0 .o .o .2 .o .2 .3 .2 Total current exp. 254.2 200.0 217.5 311.8 304.9 371.1 223.9 329.2 in real HTG 438.6 272.3 230.8 283.3 242.0 265.5 313.9 263.6 in nominal US$ 9.4 4.9 5.5 8.0 7.3 8.8 6.6 8.0 % 28.7% 57.4% 86.6% 61.4% 25.2% 9.8% 57.5% 32.2% Programs and projects 632.8 148.7 33.6 195.8 903.7 3406.5 271.7 1502.0 in real HTG 1091.9 202.5 35.6 178.0 717.3 2437.6 443.4 1111.0 in nominal US$ 23.4 3.7 .8 5.0 21.5 81.1 9.3 35.9 % 71.3% 42.6% I3.4% 38.6% 74.8% 90.2% 42.5% 67.8% Sources: Le Moniteur, Journal Officiel de la Rkpublique d’HaYti and Staff Calculation 190 Haiti: Public Expenditure Management and Financial Accountability Review Table A4.3. Structure of MARNDR’s Operating Budget (allocations), (HTG million, unless otherwise indicated), FY2005-2006 PERSONNEL EXPENSES 252.1 83% 304.8 82% in constant US$ 4.764 5.192 CONSUMPT. GOODS AND SERVICES 17.7 6% 23.1 6% in constant US$ 0.335 0.393 GENERAL HARDWARE 22.3 7% 31.8 9% in constant US$ 0.421 0.542 TANGIBLE ASSETS 1.4 0% 10.2 3% in constant US$ 0.02 7 0.173 INTANGIBLE ASSETS 0.1 0% 0.1 0% in constant US$ 0.002 0.001 SUBSIDIES AND SHARES 11.1 4% 1.o 0% in constant US$ 0.210 0.01 7 OTHER PUBLIC EXPENDITURES 0.2 0% 0.2 0% in constant US$ 0.003 0.003 TOTAL 304.9 100% 371.1 100% in constant US$ 5.8 63 Sources : MAWNDR data ;Le Moniteur, Journal Offiiciel de la RCpublique d’Ha’iti and Staff Calculations Table A4.4. Trends in MARNDR’s Expenditures, [HTG million, unless otherwise indicated), FY2004-2005 .. , Sources: MARNDR data; Le Moniteur, Journal Officiel de la RCpublique d’Ha‘iti and Staff Calculations 191 Haiti: Public Expenditure Management and Financial Accountability Review Table A4.5. Detailed Expenses by Budget Line (Fiscal year 2005-2006) (HTG thousand, unless otherwise indicated) 2005-2006 BIID<;ET BY AhlOflNI’ SPENT Allocated Spent Hate Share 312 Fuel and lubricants 12734.9 9877.0 78% 29,0% in real terms 10107.6 7839.3 709 Subsidies to other beneficiairies 11116.1 9129.4 82% 26,8% in real terms 8822.7 7245.9 222 Transport o f personnel; Expenses for travel abroad 5 147.6 3535.1 69% 10,4% in real terms 4085.6 2805.8 304 Parts and accessories for vehicles 3567.1 2099.9 59% 6,2% in real terms 2831.2 1666.7 240 Rent 1975.4 1248.5 63% 3,7% in real terms 1567.8 990.9 251 Vehicle maintenance 1897.5 1112.1 59% 3,3% in real ternis 1506.0 882.7 221 Transport of personnel ; Expenses for local travel 2321.0 1062.1 46% 3,1% in real terms 1842.2 843.0 TOTAL 28064.2 82,5% in real terms 22274.3 Source: MARNDR data Table A4.6. Detailed Expenses by Budget Line (Fiscal year 2002-2003) (XTG thousand, unless otherwise indicated) 002,2003 BIIDGET BY AXIOIJN’I’SPENI’ Allocated Spell? Rate Part 3 12 Fuel and lubricants 78% 20% I 46246 57% 12% in real terms 10958 8 6296 6 3 10 Fertilizer, plant material and seeds 5094.6 4323.6 85% 11% in real terms 6936 5 5886 8 47 1 Hydraulic works 15270.0 4295.7 28% 11% in real terms 20790 7 5848 7 25 1 Vehicle maintenance 3432.0 1923.8 56% 5% in real terms 4672.8 2619.3 303 Veterinary tools and supplies 2023.7 1858.5 92% 5% in real terms 2755.4 2530.5 240 Payments to training institutions abroad 2555.4 1534.2 60% 4% in real terms 3479.2 2088.8 702 Subsidies for autonomous organizations 3000.0 1412.8 47% 4% in real terms 4084.6 1923.5 3 11 Insecticides, disinfectants, and other chemical inputs 1400.8 1235.8 88% 3% in real terms 1907.2 1682.6 TOTAL 28787.5 76% in real terms 39195.3 Source: MARNDR data 192 Haiti: Public Expenditure Management and Financial Accountability Review Table A4.7. Shares in Budget Execution (HTG million, unless otherwise indicated) -- FY FY P\’ FY FY A! erages 2001102 2002103 2003l0-1 2004iOS FY 2005106 FY I in real terms 142.9 246.6 149.4 203.5 168.1 178.4 240.5 218.5 246.9 196.0 2001103 200406 153.5 209.5 243.7 207.2 1 % o f total expenditure 55% 46% 55% 65% 69% 52% 67% Current Expenditure 34.0 68.4 41.4 45.9 42.6 47.9 44.2 I in real terms 58.7 93.2 43.9 41.7 33.8 65.3 37.8 I % of total expenditure 13% 21% 14% 12% 12% 16% 12% Projects 81.8 106.8 94.4 86.3 70.9 94.4 78.6 in real terms 141.2 145.4 100.2 78.5 56.2 128.9 67.4 % o f total expenditure 32% 33% 31% 23% 20% 32% 22% Total Expenditure 258.8 324.7 304.0 372.7 360.3 295.8 366.5 in real terms 446.6 442.0 322.5 338.7 286.0 403.7 312.3 Sources : MARNDR data ;L e Moniteur, Journal Officiel de la Republique d’HaYti and Staff Calculations Table A4.8. Investment Budget (PIP), yearly allocations (HTG million, unless otherwise indicated) I IFY20041051 FY2005/06 1 F Y2006107 Source: MARNDR and MPCE data Table A4.9. Investment budget, Treasury funds (HTG million, unless otherwise indicated) uthorized in real ternis 79.0 65.1 in real terms 57.7 z 5-7. /Spent/ budget alloc.1 32% I86% I I (*) up to January 31,2007 Sources: MARNDR and MPCE data 193 Haiti: Public Expenditure Management and Financial Accountability Review Table A4.10. Comparison of Budgets Requested and Allocated (HTG million, unless otherwise indicated) Sources: 3IARNDR and MPCE data. Table A4.11. Generated Funds in Ministry of Agriculture (HTG million, unless otherwise indicated) Current expend. 209520.9 286347.3 289470.0 in real terms 222318.1 260229.0 229750.2 1% o f gen. funds I 0,4% I 1,3% I 0,4% 1 Source: MARNDR 194 Haiti: Public Expenditure Management and Financial Accountability Review TableA4.12. Allocations to the Ministry of Public Works, Transports and Communication (MTPTC) Sources: L e Moniteur, Journal Officiel de la RCpublique d’Ha’iti and Staff Calculations. 195 Haiti: Public ExDenditure Manaaement and Financial Accountabilitv Review TableA4.13. Infrastructure Sector. Budget Execution (HTG million, unless otherwise indicated), FY2001/06 Sources: L e Moniteur, Journal Officiel de l a RBpublique d’fla’iti and Staff Calculations. TableA4.14. Infrastructure Sector. Recurrent Budget Execution (HTG million, unless otherwise indicated FY FY FY Averages Total recurrent allocated O real terms f which Wages and salaries 2001102 2002103 207.8 358.6 122.9 175.7 239.2 122.8 220.8 234.3 130.8 379.2 344.6 196.1 - 2005!06 329 261.1 187.9 ~ FYO1103 FY04106 201.4 277.4 125.5 1 362.3 302.9 210.9 real terms 212.1 167.2 138 8 178.2 149.1 172.7 163.7 Goods and services 67.5 52.9 90 53.3 70.1 55.3 real terms 116.5 72.0 95.5 48.4 94.7 43.7 Others 17.4 0 0 129.7 5.8 96.1 real terms I 30.0 I 0.0 0.0 117.9 0.4 2.4 Total recurrent executed I 198.8 1 276 171.5 314.1 404.8 215.5 359.4 real terms 343.0 375.8 182.0 285.5 321.3 300.3 303.4 Of which Wages and salaries 121.2 130 132.6 185.3 184.2 184.7 real terms 209.1 177.0 140.7 168.4 157.3 Goods and services 146 38.9 128.8 220.6 174.7 real terms 133.9 198.8 41.3 117.1 146.1 Others n n n n Execution rates (in %) 157.1 77.7 82.8 110.2 102.9 Wages and salaries 105.9 101.4 94.5 96.3 Goods and services I 91.5 1 275.8 43.2 70.3 113.4 Sources: L e Moniteur, Journal Officiel de l a RCp blique d’ a’iti and ! iff Calculations. 196 Haiti: Public Expenditure Manaqement and Financial Accountability Review ANNEX 5 Table A5.1. Allocations to Education Sector (HTG million, unless otherwise indicated), FY2002107 FY FY E’Y FY FY FY Averages 2001i02 2002103 2003104 2004lOS 2005106 2006107 Fk’02104 FY05i07 Sources: L e Moniteur, Journal Officiel de l a Rhpublique d’Ha’iti and Staff Calculations. Educahon sector comprises Mimtry o f National Education, Youth and Sports and University o f Haiti. In FY2006/07, it includes Mimstry o f National Education and Professional Training, M i n i s t r y o f Youth and Sports, and Civil Action, University o f Haiti. We also include “interventions publiques” in favor o f education sector. Table A5.2. Composition of Spending in Education, (HTG million, unless otherwise indicated), FY2002-07 FY FT FT FY FT FY Averages 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 FY 2002-04 W2005-07 Wages and salaries 1,895.3 1,780.7 1,342.4 1,578.5 1,497.2 1,687.1 1,672.8 1,587.6 Goods and services 602.7 417.0 691.6 228.2 250.5 363.1 570.4 280.6 Others 26.5 NA NA 1,205.9 939.1 1,236.4 NA 1,127.1 Total 2,524.5 2,197.7 2,033.9 3,012.7 2,686.8 3,286.6 2,252.1 2,995.4 Share o f wages and salaries in total (in 75.1 81.0 66.0 52.4 55.7 51.3 74.3 53.0 %) Share o f goods and services and others 24.9 19.0 34.0 47.6 44.3 48.7 25.7 47.0 in total (in %) Sources: L e Moniteur, Journal Officiel de la RCpublique d’Ha’iti and Staff Calculations. 198 2 N vl x v, 2 0 U .I d - e I c tz d iz U :s .e x 2 x w s .I I cc 2 d I a2 P .- I a2 8 0 U I co i e a 0 c, 0 $ E G- 0 a a2 U .I U . E I H 2 51 W a2 3 .. 9 a2 E .I u E L a 0 rT, L 3 V L a 2 s: d 2 $ a2 Haiti: Public Expenditure Manaqement and Financial Accountability Review Appendix A5.1. Strategy for the Health Sector in Haiti General Objectives of the 2003-2008 National Strategy Plan for Health Sector Reform 1 A reduction o f at least 50% in maternal mortality; 1 A reduction o f at least 50% in the rates o f infant and under-five mortality; 1 A reduction o f 30% in the incidence o f H I V / A I D S infection; 1 A reduction o f 30% in mortality associated with HIVIAIDS; 1 A reduction o f 10% in mother-to-child transmission o f the infection; 1 A reduction o f 30% in the incidence o f Tuberculosis; 1 A reduction o f 50% in mortality associated with malaria; 1 A reduction o f 50% in disorders due to lack o f iodine; 1 Control o f some diseases (lymphatic filariasis) and elimination (neonatal tetanus) or eradication o f other diseases (poliomyelitis, measles). Intermediate Objectives I) Decentralize the health system 1 Strategy I :Revision of the Legal and Institutional Framework for Decentralization P Action 1 : Develop a legal framework assigning new functions to the local level 3 Action 2: Revise the organic law o f the Ministry o f Public Health P Action 3: Introduce structures to allow the population t o participate in managing the system 3 Action 4: Develop an urban health policy P Action 5: Give the sector tools for regulating the introduction o f CHUs > Action 6: Strengthen the management capabilities o f the M P H P at all levels 1 Strategy 2 :Expansion o f health care coverage through the introduction offunctional and eflective CHUs P Action 1: Planning, support and follow-up o f the expansion o f CHUs F Action 2: Strengthening o f the management capabilities o f C H U offices and health care institutions (particularly the CRHs) P . Action 3: Preparation o f the C H U development plans as the are eligible . . . . P Action 4: Jntroduction o f tbols for evaluating the operational development o f CHbs. 2). Zmprove the supply o f health care 9 Sfrategy 1: Improvement o f the quality of health care )1 Action 1: Introduction o f Basic Package o f Services (BPS) in the health zones served b y C H U health care facilities, giving priority to combating maternal, mortality 3 Action 2: Definition and implementation o f a national policy on the subject o f health care quality 3 Action 3: Development o f the laboratory network 1 Strategy 2: Development o f universal access to services in response to specific priority health concerns > Action 1 : Evaluation o f management cost 3 Action 2: Definition o f financial mechanism P Action 3: Application and follow-up 1 Strategy 3: Gradual integration ofprograms designed to manage priority health problems, taking the millennium objectives into account 3 Action 1: Studies conducted on efficiency and effectiveness o f integrating priority programs P Action 2: Introduction o f a national children’s office t o coordinate EPI, IMCI, nutritional deficiencies, etc. k Action 3: Introduction o f a national women’s health committee > Action 4: Revision o f sub-sectoral strategic plans from a perspective o f integration in terms o f the new directions in the sector. 1 Strategy 4: Consideration of traditional medicine as a key player in the health sector 3 Action 1: National entity put in charge o f relations between modem and traditional medicines P Action 2: Studies and research done on traditional medicine sector 3 Action 3: Mechanisms introduced for partnering w i t h traditional practitioners Strategy 5: Improve managemennl o f mental disorder P Action 1 : Training o f nursing staff in the first and second tiers o f the CHU, in the management o f simple cases and referral o f complex cases, according t o the BPS 3 Action 2: Strengthening o f specialized services in the hospitals at the secondary and tertiary levels P Action 3: Promotion o f changes in behavior for family and community integration o f the mentally ill Strategy 6: Health promotion and protection 200 Haiti: Public Expenditure Management and Financial Accountability Review 9 Development o f health promotion activities as defined in the BPS 9 Development of policies for the most vulnerable population groups (women, pregnant women, children, street children, orphans, adolescents, families in difficult situations) 3) Revitalize and expand thepublic hospital network in order to increase the supply of health care and improve . its quality Strategy 1: Improvements in the management ofpublic health 9 Action 1: Training in hospital management for staff responsible for medical care, nursing care and administration 9 Action 2: Development o f standardized management tools, including establishment plans, for hospitals 9 Action 3: Installation o f a central agency responsible for organizing the upkeep and maintenance infrastructures, equipment and logistical resources 9 Action 4: Securing at the national level o f financing for the revitalization and expansion o f public hospitals 9 Action 5: Legislation on administrative and financial autonomy o f secondary and tertiary level hospitals . 9 Action 6 : Follow-up and evaluation o f managerial performance Strategy 2: Improved supply of services 9 Action 1: Gradual implementation o f establishment plans 9 Action 2: Installation o f quality transfusion capability in all hospitals 9 Action 3: Implementation o f continuing training policy in the area of pathology management 9 Action 4: Strengthening o f departmental and university hospitals’ maternity wards b Action 5: Promotion o f service quality (through application o f management standards, improved reception, supervision) 9 Action 6 : Improved hygiene and cleanup in the hospital setting 9 Action 7: Organization o f a system for hospital accreditation and inspection 9 Action 8 : Installation of an integrated medical emergency system (IMES) 9 Action 9: Installation o f a national cancer research center 9 Action 10: Installation o f hospital coordination structures and mechanism = rtegy 3: Development o Strr, f the hospital-university axis 9 Action 1 : Modernization and autonomy o f Haitian State University Hospital 9 Action 2: Expansion o f hospital-university axis, based on a process that incorporates transformation o f existing facilities . 9 Action 3: Development o f research and partnership agreements Strategy 4: Expansion of thepublic hospital network 9 Action 1: Expansion o f CRHs b Action 2: Networking o f hospital facilities in the metropolitan area 4) Improve regulation o f the sector by giving the MPHP toolsfor managing the entire sector, thus strengthening . the leadership of the M P H P and its regulatory role Strategy 1: Strengthening o f legal and institutional framework for regulation 9 Action 1: Preparation and utilization by the public sector o f legal and administrative instruments necessary t o fulfill its mission 9 Action 2: Updating o f the country’s health legislation taking into account the national and international situation 9 Action 3: Installation o f training facilities for the health professions (physicians, pharmacists) 9 Action 4: reactivation o f the national ethics commission 9 Action 5: Installation o f accreditation systems for health institutions and training institutions b Action 6 : Preparation and/or completion and/or dissemination o f rules and standards for the delivery o f quality health care services 9 Action 7: Improved effectiveness and development o f contractual approach 9 Action 8: Strengthening o f central and departmental offices responsible for ensuring the application o f laws, rules and standards 9 Action 9: Creation o f a national public health laboratory 9 Action 10: Strengthening o f health inspection 9 Strategy 2: Definition of a research policy 9 Action 1: Introduction o f an institutional framework for research b Action 2: Development o f an applied research program b Action 3: Implementation o f a research training plan 9 Action 4: Development o f research strategies in the hospital-university axis 201 Haiti: Public Expenditure Management and Financial Accountability Review 9 Action 5: Human resources encouraged to conduct research 9 Action 6 : Mobilization o f resources for conducting research projects s Strategy 3: Regular inspection and evaluation o f health actions at all levels 9 Action I : Strengthening o f health inspection b Action 2: Observance o f established standards and rules . Strategy 4: Strengthening ofpartnership b Action 1: Preparation o f a partnership charter P Action 2: Strengthening o f intersectoral coordination (water, sanitation, communication facilities and routes, agriculture, environment, education, women's affairs, social sector, law, etc.) b Action3: Strengthening o f the ability o f the M P H P t o develop partnerships 5) Modernize the health information system Strategy 1: Revision of the information systemfor management b Action 1 : Evaluation o f existing health information system for management 9 Action 2: Strengthening o f central team in charge o f guiding the information system for management F Action 3: Strengthening o f departmental offices for management and utilization o f information for management b Action 4: Updating o f objectives, procedures and tools o f the information system for management at each level, taking into account the hospital's information requirements P Action 5: Introduction o f methods and means for gathering, collecting, processing and disseminating information (including feedback) t o all levels in real time b Action 6: Traininghaking awareness as to the use o f new tools at all levels b Action 7: Traininghaking awareness as to the utilization o f data at all levels - b Action 8: Monitoring the quality o f information collected Strategy 2: Strengthening of the epidemiological surveillance and warning system b Action 1: Strengthening the central team in charge o f epidemiological surveillance b Action 2 : Centralization o f all epidemiological data at level o f a single central entity > Action 3: Strengthening o f departmental offices in charge o f epidemiological surveillance > Action 4: Updating o f the list o f illness and syndromes requiring active surveillance 9 Action 5: Review and ation o f collection and reporting procedures, definition o f illness and - l i s t o f illness and synd be reported, based on specific surveillance objectives Strategy 3: Improvements in the availability and accessibility o b Action 1: Creation o f a documentation center f information for development of the system b Action 2: Publication and dissemination o f health sector data b Action 3: Studies, surveys and applied research projects carried out 9 Action 4: Installation o f the health observatory k Action 5: Training for media and journalists in health intervention 9 Action 6: Introduction o f health education programs 6) Develop human resources Strategy 1: Introduction of the political and institutional framework for management of human resources appropriate to the needs of the sector b Action 1 : Developmenb'adaptation o f institutional and political framework for HR management b Action 2: Strengthening o f HR planning by category and by level in system 9 Action 3: Planning for integrating professionals in training in Cuba b Action 4: Integration o f human resources in the system according to a career plan 9 Action 5: Registration o f professionals 9 Action 6: Organization o f a State examination for health professionals . Strategy 2: Bringing the qualijkation of operational personnel up to standard b Action 1: Introduction o f a continued training system for the sector Strategy 3: Production o f human resources consistent with the sector's needs 9 Action 1: Development o f a framework for partnerships between the M P H P and private and public training institutions k Action 2: Strengthening o f initial training b Action 3: Conversion o f Health Management Information and Training Center into National School o f Public Health 3 Action 4: Planning o f specialization needs P Action 5: Adaptation o f social service taking public-private partnership into account 8 Strategy 4: Geographically equitable allocation of HR according to needs 202 Haiti: Public Expenditure Management and Financial Accountability Review > Action 1: Deployment o f human resources based on needs . Strategy 5: Improvement in the appeal o > f the sector Action 1 : Development o f career management profiles and tools > Action 2: Improvement o f working conditions > Action 3: Promotion o f equitable compensation 7) Guarantee the population’s access to essential drugs Strategy I : Preparation of an oficial document on the National Pharmaceutical Policy, including policy in essential drugs > Action 1: Completion and submission o f National Pharmaceutical Policy document to the process o f validation by ad hoc groups . Strategy 2: Introduction o f an effective systemfor supply o philanthropic health facilities f essential drugs (INN)for state and > Action 1: Revision o f existing supply (based on PROMES) and distribution system 8 Strategy 3: Strengthening ofpharmaceutical inspection and control services > Action 1 : Strengthening of the Pharmaceutical Control Division in terms o f human resources, training and equipment > Action 2: Development o f a charter on donated medication > Action 3: Installation o f national quality control laboratory 9 Strategy 4: Improved accessibility of essential drugs on the national list by level, at an affordable and standardized cost > Action 1: Review o f existing system for setting prices for medications > Action 2: Rationalization o f prescriptions > Action 3: Universal access t o certain medications > Action 4: Improved access to essential drugs for certain vulnerable groups 1 Strategy 5: Development of natior?alcapacity io manufacture medications 8) Increase and rationalize financial resources in order to allow f i r improvements in the supply service in the context of sector-based reform. Strategy 1: Secure thefinancial resources neededfor proper operation o f the health system > Action 1 : Securing regular availability of M P H P resources compared to system needs > Action 2: Mobilization o f funds > Action 3: Promotion o f solidarity mechanisms > Action 4: Development o f resources allocation mechanisms 8 Strategy 2: Regulate the financing subsystem o f the health sector > Action 1: Introduction and monitoring o f regulations adapted to financial system > Action 2: Introduction o f a decentralized management system based on results Source: Ministry o f Public Health and Population, Strategic Plan for Health Sector Reform, March 2004. 203 Haiti: Public Expenditure Manaqement and Financial Accountability Review Table A5.5. Allocations to Health Sector (In million of gourdes, unless otherwise indicated), THB in real term 2,033.54 965.74 701.48 1,231.77 1;018.39 31889.45 1233.6 2046 5 THB in nominal US$ 43.52 17.53 16.66 34.77 30 55 129.41 25.9 64 9 THB in US$/capita 5.24 2.07 1.93 3.94 3 40 14.17 72 THB i n %of GDP 1.38% 0.75% 0.55% 0.97% 0 76% 2.71! % 0.89% 148% Total Recurrent Budget 685.9 587.7 587.9 871.2 962 7 1,150.90 620.5 994.9 TRB in real term I,183.54 800.18 623.81 791.74 764 09 823.56 869.2 793 I TRB in nominal US$ 25.33 14.53 14.81 22.35 22 92 2 7.40 18.2 24 2 TRB in US$/capita 3.05 1.71 I .71 2.53 2 55 3.00 2.2 2.7 Total Investment Budget 492.6 121.6 73.2 484.2 320.4 4,284.50 229.2 1,696.40 TIB in real term 850.00 165.56 77.67 440.04 254 30 3,065.89 364.4 1253 4 TIB i n nominal US$ 18.19 3.01 1.84 12.42 7 63 102.01 7.7 40 7 TIB i n US$/capita 2.19 0.35 0.21 1.41 0 85 11.17 0.9 4.5 Total Recurrent in % o f total budget 58.2 82.8 88.9 64.3 75 21.2 73 37 Total Investment in % o f total budget 41.8 17.2 11.1 35.7 25 78.8 27 63 Total Nation Budget 13,278.4 15,333.2 18,853 4 32,404.5 37,816 5 64,436.8 15,821.7 44,885.9 TNB in real term 22,912.3 20,s 76.7 20,004 9 29,448 8 30,014 7 46,109.5 21264.6 35191.0 TNB in nominal IJS$ 490.33 3 79.02 475.10 831 23 900 39 1,534.21 448.2 1088 6 TNB in US$/capita 59.05 44.71 5493. 9422 100 29 167.98 52.9 120 8 Shares o f Health sector in Nation 8.9 4.6 3.4 8.4 5.4 60 Budget (in %j Total Nation Recurrent Budget 6,908.10 9,012.30 16,992.3 7,808.50 19,597.0 17,786.1 10,970.9 15,063 9 (Excl Interest Payments) TNR zn real term I 1920.1 12270.6 18030.2 7096.3 15554 0 12727.3 14073.6 I 1 792 5 THR in nominal 440.2 303.3 454.4 182 0 370 3 303.0 399.3 285 I IJS$ THR in US$/capita 53.0 35.8 52.5 20 6 41 3 33.2 47.1 31.7 Percent o f Total (Excl Interest 9.9 6.5 3.5 11.2 49 6.5 5.7 6.6 Paymentsj Memo I t e m Nominal GDP 85,700 94,028 119,758 140,387 168,034 200,456 85,700 94,028 Sources: L e Moniteur, Journal Officiel de l a Rkpublique d’Ha’iti and Staff Calculations. 204 Haiti: Public Expenditure Management and Financial Accountability Review Total Budget allocated 1,178.5 709.3 661.1 1,355.4 1,283.1 849.7 1,319.30 TBA in US$ 43.5 175 16 7 34.8 30.6 25.9 32 7 in real term 20335 9657 701.5 12318 1018.4 1233.6 1125.1 Total Budget executed 771 8 930.3 655.3 890.4 700 785.8 795.2 TBE in US$ 28 5 23 0 165 22.8 16 7 22.7 19.8 in real term 1331.8 12666 6953 809.2 555.6 1097.9 682.4 Execution rates (in YO) 65.5 131.1 99.1 65.7 54.6 92.5 60.3 Total recurrent allocated 685.9 587.7 587.9 871.2 962.7 620.5 916.9 TRA in US$ 25.3 14.5 14.8 22.3 22.9 18.2 22.6 in real term 1183.5 800.2 623.8 791.7 764.1 869.2 777.9 Total recurrent executed 686.8 790.3 568.2 798.8 689.9 681.8 744.4 TRE in constant US$ 25.4 19.5 14.3 20.5 16.4 19.7 18.5 in real term 1185.1 1076.0 602.9 725.9 547.6 954.7 636.8 Execution rates (in YO) 100 134.5 96.7 91.7 71.7 109.9 81.2 Total investment allocated 492.6 121.6 73.2 484.2 320.4 229.2 402.3 TIA in US$ 18.2 3.0 1.8 12.4 7.6 7.7 10.0 in real term 850.0 165.6 77.7 440.0 254.3 364.4 347.2 Total investment executed 84.9 139.9 87.1 91.5 10.2 104 50.8 TIE in US$ 3.1 3.5 2.2 2.3 0.2 2.9 1.3 in real term 146.5 190.5 92.4 83.2 8.1 143.1 45.6 Execution rates (in YO) 17.2 115 118.9 18.9 3.2 45.4 12.6 Sources: L e Moniteur, Journal Qfficiel de l a RCpublique d’Haiti and Staff Calculations. Table 85.7 Health Sector. Recurrent Budget Execution (MTG million, unless otherwise indicated), Sources: L e Moniteur, Journal Officiel de l a RCpublique d’Hai’ti and Staff Calculations. 205 Haiti: Public Expenditure Management and Financial Accountability Review ANNEX 6 Table A6.1. Allocations to Justice and Security Sector (HTG million, unless otherwise indicated), P Y - - -~ . - O -O - ~ / O ~ FY FY FY FY FY FY Aberages 2001102 2002103 2003104 2004/05 2005/06 2006/07 FYO2104 F\ 05/07 Total Justice and 1,054.4 1,153.6 1,180.1 2,247.9 2,491.0 5,599.9 1,129.4 3,446.3 Security Budget In million real I 1,819.4 1 1,570.7 In million of US$ In % o f GDP Of which HNP NA NA In million o f real NA NA YOof total budget Total Nation Budget 13,278.4 15,333.2 I n millions of gourdes 2,291 2.3 2,0876.7 In million nominal Shares of Justice and 6.3 9.6 6.6 8.7 7.2 8.3 Security in Nation Sources: L e Moniteur, Journal Officiel de I RBpublique d’Ha’iti and Staff Calculations. Total Nation Recurrent 6,908.1 9,012.3 13,831.1 7,808.5 19,597.0 17,786.1 9,917.1 15,063.9 Budget (Excl. Interest Payments) Percent o f Total (Excl. 13.9 10.9 8.5 28.8 12.6 16.7 10.5 17.0 Interest Payments) Sources: L e Moniteur, Journal Officiei de l a RBpublique d’Ha‘iti and Staff Calculations. 207 Haiti: Public Expenditure Management and Financial Accountability Review Table A6.3: Evolution of the composition of expenditure (in Percent of total expenditures, excluding services Capital 14.0 14.4 6.7 6.9 1.6 15.9 11.7 8.1 expenditures Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 expenditures FY FY FY FY FY Averages 2002t03 2003104 2004i05 200906 2006i07 FY02104 FY05107 A. Budgets Traveling expenditures 8,806 707 5,504 10,334 6,687 4,757 7,9 19 Renting o f building 3,303 1,354 13,625 6,625 6,625 2,329 10,125 Renting o f vehicles 1,250 1,852 7,912 4,274 191 1 3 51 6,093 Maintenance o f office equipment 2,784 178 2,125 1,183 1,483 1,48 1 1,119 Maintenance o f vehicles 5,519 2,637 7,226 1,708 3,730 4,078 4,467 Sources: Government Data and Staff calculations. 208 - d 0 . mI *E E. W $ 0 0 E m 9 cn s 0 0 E" v1 L. 0 0 L . W r/) .h c, .I L. 0 .I & L 8 9 0 0 +I E 5: .I L. m a V i 0 w 0 4 0 .I L. 1 0 al r W V .I +I v1 1 c, W 9 .* E Haiti: Public Expenditure Management and Financial Accountability Review ANNEX 7 Summary List o f Equations Production of Health Services and Effective Labor - KGIH(t) = AHC. [el ( t ) KGI( t )] pH'* [OH ( t ) * KGH(t )] T ( t ) = AT. [LEp(t)]". [H(t)]''P' Production of Commodities Y ( t ) = AY* [V( t)lpY'* [el (t)' KGl(t)]'-pY'-PY2 Population, Labor Supply, and Schooling Technology &(t) = AEC. [el (t). KGl(t)IPEC* [BE(t)* KGE(~)}'-'~' 210 Haiti: Public Expenditure Management and Financial Accountability Review 211 Haiti: Public Expenditure Management and Financial Accountability Review IG(t) = 81*Y(t) A(t) = G(t) - T ( t ) - 8D*Y(t) Stocks o f Public Capital: Quality Indicators ei(t) = eio/{eio + (I - elo).exp[-(KG,(t-l)/Kp(t-l))]] BH(t) = BHo/{BHo + (1 - BHo).exp[-(KG~(t-l)/N(t-l))]] 0E(t) = BEo/{BEo + (1 - eEo).exp[-(KGE(t-l)/~5HG-STU(t-1)]) 212 Haiti: Public Expenditure Management and Financial Accountability Review ANNEX 8 Table A.8.1. Baseline scenario, 2007-15 Years 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real GDPper capita at market pnces (% change) 0.7 0.7 0,6 0,6 0.6 0.5 0.5 0,4 0.3 Poverty rate-Favallion's adjusted elasticity (Gini = 66 0) 55.5 55'3 55.2 55.1 55.0 54.8 54.8 54,7 54.6 Poverty rate-Growthelasticity of -1 0 55.0 54.6 54.3 53.9 53,6 53.4 53.1 52,9 52,8 Government Sector 1% of GDP) Toal reso~rces (.ncludlnggran's) 13.5 13,s 14.3 14.7 14.7 14,7 14,7 14.7 14.7 Total rerenues 9.6 9.6 9.5 9.5 9,4 9,4 9.4 9.5 9,5 Direct taxes 3.9 3,9 3,9 3,9 3,9 3.9 3.9 3.9 3.9 Indirect taxes 4,9 4.8 4.7 4.7 4,6 4.6 4.6 4,s 4.6 User fees 03 0,s 0.8 0.8 0.9 0.9 0,9 0.9 0.9 Fore.gn ard [granis) 3,s 4.3 4.8 5.2 5.2 5.2 5.2 5.2 5.2 Total exDenalture 11.8 12.2 12.6 12,9 12,9 12.9 12.9 13,O 13.0 Spenoing on goods and serwces (iotal) 6.6 6.6 6,s 6.5 6.5 6.5 6.5 6.6 6.6 Spendlng on maintenance 0.9 0.9 0.8 0.8 0.8 0.7 0.7 0,7 0.7 Vvages an0 maries 3.4 3.5 3,5 3.5 3.5 3.6 3.6 3.6 3,7 Security 1.4 1.4 1.4 1.4 1.4 14 1.4 1.4 1.4 Olner 0.9 09 0.9 0.9 0,9 09 09 0.9 03 Investment 4.5 4.9 5,3 5.7 5,7 5.7 5.7 5.7 5.7 lnieresl payments 0.7 0.7 0.7 0.7 0.7 0,7 0,7 0.7 0,7 Overai f scal Da ance including grants (caw bass) 1,7 1.7 1.7 1.7 1.7 1.7 1,7 1.7 1,7 Memorandum items Private investment (% of GOP) 32,3 34,l 35,6 36.7 37.6 38,l 38,4 36.3 38.1 Public investment ( % of total public expenditure) 32.1 34.0 35,8 37.5 37,5 37.5 37,5 37-5 37.4 Health ( % of public investment) 5.4 6.4 7,4 8,4 9.4 9.4 9.4 9.4 9.4 Infrastructure(% of public investment) 36.4 36.9 37.4 37.9 38,9 39.9 39.9 39.9 39.9 Eduwtion (% of public investment) 7.3 9.3 11.3 13.3 15.3 15.3 15.3 15.3 15,3 Other (Yoof public investment) 50.9 47,4 43,9 40.4 36.4 35.4 35,4 35,4 35.4 Aid (% of total revenue) 40,3 45.3 50.3 55.0 55.4 55.5 55.5 55,3 55.0 Total Aid (%of public Investment) 85.3 87.E 89.3 90.5 90.9 91.0 91.1 91.0 90.9 Total debt (% of GDP) 29.0 28.8 28.7 28.6 28.5 26.5 28,5 26,6 26.7 Educatedlabur (in % of population) 37.4 37.8 38.3 38,s 39.3 393 40.4 40,9 41.5 Note: The "adjusted" elasticity formula proposed by Ravallion (2004) is -9.3'(1-Glni]A3 = -1 13 where Gini index is 50.5 for Haiti 213 Haiti: Public Expenditure Manaaement and Financial Accountability Review Table A.8.2. Higher Total Public Investment, 2007-15 (Deviation from the Baseline scenario) Years 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real GDP per capita at market prices (% change) 0,o 0.0 0.3 0.3 0.3 0.4 03 0.3 0.2 Poverty rate-Ravallion'sadjusted elasticity (Gini = 66.0) 0.0 0,o -0.1 -0.1 -0.2 -0.3 -0.3 -0,4 -0.4 Poverty rate-Growthelasticily 01-1.0 0.0 0,O -0.2 -0.3 -0.5 -0,7 -0.9 -1,0 -1.1 Government Sector (% of GDP) Total resources (including grants) 0.0 5.0 5,O 5,O 5.0 4.0 3.0 2.0 1,o Total revenues 0.0 0.0 0,o 0.0 0.0 0,o 0-0 0.0 0,0 Direct taxes 0,o 0.0 0,o 0.0 0,o 0.0 0.0 0.0 0.0 Indirect taxes 0,o oxo 0,o 0.0 0.0 -0.1 -0,l -0,l -0,1 User fees 0.0 0.0 0.0 0.0 0,o 0.0 0.1 0.1 0.1 Foreign aid (grants) 0,o 5.0 5.0 5.0 5.0 4.0 3.0 2.0 1.o Total expenditure 0.0 5.0 5.0 5.0 5.0 4,o 3.0 2,o 1.o Spending on goods and services (total) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Spending on maintenance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,o 0.0 Wages and salanes 0.0 0.0 0.0 0.0 0,o 0.0 -0,l -0.1 -0,1 Security 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Other 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0,0 Investment 0.0 5.0 5.0 5,O 5,O 4,O 3,O 2.0 1,o Interest payments 0.0 0,o 0.0 0.0 0.0 0.0 0,o 0,o 0,0 Overall fiscal balance including grants (cash basis) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Memorandum items Private investment ( Oh of GDP) 0.0 0.0 0.3 0.5 0.8 1.1 1.4 1,6 1,7 Public investment (%of total public expenditure) 0.0 16.9 16.1 15,4 15.4 13.0 10-3 7.3 3.9 Health ( % of public investment) 0.0 0,o 0,o 0.0 0,o 0.0 0.0 0,o 0.0 Infrastructure(% of public investment) 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 Education ( % of public investment) 0.0 0,o 0,o 0,o 0.0 0.0 0.0 0,o 0.0 Other (% of public investment) 0.0 0,o 09 0.0 0.0 0.0 0.0 0.0 0,0 Aid ( % of total revenue) 0.0 52,3 52,8 53.1 53.2 42.7 31,Q 21.2 10.5 Total Aid (% of public investment) 0.0 6.2 5.2 4.4 4.2 3.6 2.9 2.1 1.o Total debt (% of GDP) 0.0 0.0 -O,? -0.2 -0.3 -0.4 -0.4 -03 -0.5 Educated labor (in % o f population) 0,000 0,000 0,000 0.000 0,001 0,002 0,003 0.004 0,006 214 Haiti: Public Expenditure Management and Financial Accountability Review Table A.8.3 Higher Total Public Investment and Higher Efficiency o f Public Investment, 2007-15 (Deviation from the Baseline scenario) Years 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real GDP per capita at market prices (% change) 0.0 0,o 0.5 0.7 0.8 0.8 0.7 0.7 0.6 Poverty rate-Ravallion's adjusted elasticity (Gini = 66.0) 0.0 0.0 -0.1 -0.3 -0.4 -0.6 -0.7 -0.8 -1.0 Poverty rate-Growth elasticity of -1.0 0.0 0.0 -0.3 -0,7 -1.1 -1.5 -1.9 -2.2 -2.5 Government Sector (% of GDP) Total resources (inciudinggrants) 0.0 5,O 5,O 5.0 4.9 3.9 2.9 1.9 0.9 Total revenues 0.0 0.0 0,o 0.0 0.0 -0.1 6.1 -0.1 -0.1 Direct taxes 0.0 0.0 0.0 0.0 0.0 0,o 0.0 0,o 0.0 Indirect taxes 0.0 0,o 0,o -0.1 -0.1 -0.1 -0.2 -0.2 -0.2 User fees 0.0 0.0 0,o 0.0 0.0 0.1 0.1 0,l 0.2 Foreign aid (grants) 0.0 5,O 5,O 5.0 5.0 4.0 3.0 2.0 0,9 Total expenditure 0.0 5,O 5.0 5,O 4.9 3.9 29 1.9 0,s Spending on goods and services (total) 0.0 0.0 0,o 0.0 0.0 -0.1 -0.1 -0.1 -0.1 Spending on maintenance 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0-1 0,l Wages and salaries 0.0 0.0 0.0 0.0 -0.1 -0,l -0.1 -0.1 -0.2 Security 0.0 0,o 0.0 0.0 0.0 0.0 0,o 0.0 0,o Other 0-0 0.0 0.0 0.0 0.0 0.0 0.0 0-0 0.0 investment 0.0 5.0 5,O 5.0 5.0 4,O 3.0 2.0 1.o Interest payments 0,o 0.0 0.0 0,o 0,o 0.0 0.0 0.0 0.0 Overall 6scal balance including grants (cash basis) 0.0 0.0 0.0 0,o 0,o 0.0 0,o 0.0 0.0 Memorandum items Private investment ( O h of GDP) 0.0 0.0 0.5 1.1 1.7 23 2.9 3.4 3.8 Public investment (Yo of total public expenditure) 0.0 16,9 16.2 15.5 15.5 13,l 10.4 7.4 4.1 Health (O h of public investment) 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0.0 Infrastructure ( O h of public investment) 0,o 0.0 0.0 0,o 0.0 0.0 0,o 0,o 0.0 Education (%of public investment) 0.0 0.0 0.0 0,o 0,o 0,o 0,o 0.0 0.0 Other (% of public investment) 0.0 0.0 0.0 0.0 0.0 0,o 0,o 0.0 0.0 Aid (% of total revenue) 0.0 52.3 53.0 53,4 53,5 42,9 32,l 21.2 10.3 Total Aid ( % of public investment) 0,o 6,2 5.2 4.4 42 33 2,8 1.8 0.6 Total debt ( O h of GDP) 0.0 0,O -0.2 -0.3 -0.5 -0.7 -0,9 -1.1 -1.2 Educated labor (in % of population) 0.000 0,000 0,000 0,001 0,002 0,004 0,006 0.010 0,014 215 Haiti: Public Expenditure Manaqement and Financial Accountability Review Table A.8.4 Reallocation of spending to health ,2007-15 (Deviation from the Baseline scenario) 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real GDP per capita at market prices (X change) 0,o 0.0 0.0 0.0 0.0 0.1 0.1 0.1 0.1 Poverty rate-Ravallion'sadjusted elasticity (Gini = 66 0) 0.0 0.0 0.0 0,o 0.0 0,o 0.0 -0.1 -0.3 Poverty rate-Growth elasticity o f - I 0 0.0 0.0 0.0 0.0 -0.1 -0.1 -0.1 -0.2 -0.2 GovernmentSector (X of GDP) Total resources (including grants) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 Total revenues 0.0 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 Direct taxes 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 indirect taxes 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 User fees 0.0 0,o 0.0 0.0 0.0 0,o 0.0 0.0 0.0 Foreign aid (grants) 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0.0 Total expenditure 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0.0 Spending on goods and services (total) 0.0 -0.5 -0.5 -0.5 -0.5 -0.5 -0.5 -0.5 -0.5 Spending on maintenance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Wages and salaries 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Security 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,O 0.0 Other 0.0 -0.5 -0.5 -0.5 -0.5 -0.5 -03 -0.5 -0.5 Investment 0.0 0.5 0.5 0.5 03 0,s 03 0.5 0.5 Interest payments 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 Overallfiscal balance including grants (cash I iasis) 0.0 0.0 0.0 0,o 0,o 0.0 0,o 0.0 0.0 Memorandumitems Private investment (%of GDP) 0,o 0,o 0.0 0,l 0.1 02 0.2 0,2 0.3 Public investment ( % of total public expenditure) 0.0 3.4 3 >4 3.3 3.3 3.3 3.3 3.3 3.3 Health (46 of public investment) 0.0 5.0 5.0 5.0 5.0 5.0 5.0 5,O 58 Infrastructure(% of public investment) 0.0 0.0 0.0 0,o 0.0 0.0 0,o 0.0 0.0 Education( % of public investment) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,0 Other (46 of public investment) 0.0 -5.0 -5.0 -5.0 -5.0 -5.0 -5.0 -5.0 -5.0 Aid ( % of total revenue) 0.0 0.0 0,o 0.0 0.0 0.1 0.1 0.1 0.1 Total Ald (% of public investment) 0,O -8.0 -7.6 -7.2 -7.2 -7,2 -7.2 -7.2 -7.2 Total debt (% of GDP) 0.0 oxo 0.0 0.0 0.0 0.0 -0.1 -0.1 -0,1 Educated labor (in % of population) 0.0 0.0 0,o 0.0 0,o 0.0 0.0 0.0 0.0 216 Haiti: Public Expenditure Management and Financial Accountability Review Table A.8.5 Lower collection cost, higher direct tax rate, 2007-15 (Deviation from the Baseline scenario) Years 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real GDP per capita at market prices (% change) 0.0 0,o -0,1 -0.2 -0.2 -0.2 0,2 -0.2 -0.1 Poverty rate-Ravallion's adjusted elasticity (Gini = 66.0) 0,o 0,o 0,o 0.0 0.1 0.1 0.2 0,2 0,2 Poverty rate-Growthelasticity of -1.0 0.0 0.0 0.0 0.1 0.2 0.3 0.4 03 0,6 Government Sector (% of GDP) Total resources (includinggrants) 0.0 0.0 0,o 0,o 0,o 0.0 0.0 0.0 0.1 Total revenues 0,o 1.2 2.2 3.2 32 3.2 3.2 3.2 3.2 Direct taxes 0.0 1.1 2.1 3.0 3.0 3.0 3.0 3.0 3.0 Indirect taxes 0.0 0.1 0.1 0,l 0.1 0.1 0.1 0.1 0.1 User fees 0.0 0.0 0.0 0-0 0.0 0,o 0,o 0,o 0.0 Foreign aid (grants) 0.0 -1.2 -2.2 3.1 -3.1 -3.1 -3.1 -3.1 -3.1 Total expenditure 0.0 0,o 0,o 0.0 0.0 0.0 0.0 0.0 0.1 Spending on goods and services (total) 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 Spending on maintenance 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0,0 Wages and salaries 0.0 0.0 0.0 0.0 0.0 0,o 0,o 0,o 0.0 Security 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,o 0,0 Other 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Investment 0,o 0.0 0,o 0,o 0.0 0.0 0,o 0,o 0.0 Interest payments 0.0 0.0 0.0 0.0 0-0 0,o 0.0 0,o 0.0 Overallfiscal balance includinggrants (cash basis) 0,o 0.0 0,o 0,o 0.0 0.0 0.0 0,o 0.0 Memorandumitems Private investment ( % of GDP) 0.0 -1.0 -2.2 -3.3 -3.5 -3.7 -3.8 -3,9 -4.0 Public investment(% of total public expenditure) 0,o 0.0 0.0 0,o -0,l -0.1 -0.1 -0.1 -0.1 Health ( % of public investment) 0.0 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 Infrastructure(% of public investment) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Education(% 0%public investment) 0.0 0.0 0.0 0,o 0,o 0.0 0.0 0.0 0.0 Other (% of public investment) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Aid (% of total revenue) 0,o -16.2 -28.0 -38,7 -38,9 -39,O -38.9 -38.8 -38.6 Total Aid (% of public investment) 0.0 -24.2 -40.6 -54.7 -54.7 -54.7 -548 -54.7 -54.7 Total debt (% of GDP) 0,o 0.0 0.0 0,l 0.1 0.2 0.2 0.3 0.3 Educated labor (in % of population) 0.0 0.0 0.0 0.0 0.0 0,o 0,o 0,o 0.0 217 Haiti: Public Expenditure Management and Financial Accountability Review Table A.8.6 Lower collection cost, higher direct tax rate, new revenue to investment in infrastructure, 2007-15 (Deviation from the Baseline scenario) 2007 2008 2009 2010 2011 20t2 2013 2014 2015 Real GDP per capita at market prices (% change) 0.0 0,o 02 0.5 0.7 0.7 0.8 0.8 0,s Poverty rate-Ravallion'sadjusted elasticity (Gini = 66.0) 0.0 0.0 -0.1 -0.1 -0.3 -0.4 -0.6 -0.8 -0.9 Poverty rate-Growlhelasticity of -1 0 0,o 0,o -0.1 -0.4 -0,s -1.2 -1,6 -2.0 -2.4 Government Sector (% of GDP) Total resources (including grants) 0.0 I.2 2,1 3.1 3.1 3.1 3.0 3.0 3.0 Total revenues 0,o 12 2.2 3.1 3.1 3.1 3.0 3.0 3.0 Direct taxes 0,o 1.1 2,l 3.0 3,O 3.0 3,O 3.0 3.0 Indirect taxes 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0.0 User fees 0,o 0.0 0,o 0,o 0,l 0.1 0.1 0.2 0.2 Foreign aid (grants) 0.0 0.0 0.0 0.0 0.0 0,o 0.0 0.0 0.0 Total expenditure 0,o 12 2.1 3.1 3.1 3,1 3.0 3.0 3.0 Spending on goods and services (total) 0.0 0.0 0,o 0.0 oxo -0.1 -0.1 -0,l -0,l Spending on maintenance 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0'0 0.1 Wages and salaries 0,o 0,o 0.0 0.0 0.0 -0.1 -0.1 -0.1 -02 Security 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Other 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Investment 0.0 12 2.2 3.1 3.1 3,1 3.0 3,O 3,O Interest payments 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Overall fiscal balance includinggrants (cash basis) 0.0 0,o 0,o 0.0 0.0 0,o 0.0 0,o 0.0 Memorandum items Private investment (%of GDP) 0.0 0.0 0.2 0.6 1.2 1.8 2.4 3.0 3,7 Public investment (% of total public expenditure) 0.0 5,O 8,l 10,7 10,7 103 10.8 10.9 10,9 Health (% of public investment) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,o 0,o Infrastructure(% of public Investment) 0.0 0,o 0,o 0.0 0,o 0.0 0.0 0,o 0.0 Education( % of public investment) 0.0 0.0 0.0 0,o 0,o 0.0 0.0 0.0 0.0 Other (% of public investment) 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,o Aid (% of total revenue) 0.0 -5.1 -9.3 -13,7 -13.8 -13,9 -14.1 -14,l -14,2 Total Aid (% of public investment) 0.0 -17.1 -25.8 -32.0 -32.2 32,4 -32.6 -32.8 -33.0 Total debt ( O h of GDP) 0.0 . 0,o -0.1 -0.2 -0,4 -0,6 -0,s -1,0 -12 Educatedlabor (in % of populatlon) 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,o 218 Haiti: Public Expenditure Management and Financial Accountability Review Table A.8.7 Higher security spending, 2007-15 (Deviation from the Baseline scenario) Years 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real GDP per capita at market prices (% change) 0,000 -0,001 0,000 0.001 0,002 0,002 0,003 0.003 0.003 Poverty rate-Ravallion's adjusted elasticity (Gini = 66.0) 0.000 0.000 0,000 0,000 0,000 -0,001 -0,001 0.002 -0.003 Poverty rate-GrOWlhelasticity of -1.0 0,000 0,001 0,000 0,000 -0,001 -0,002 -0,004 0,005 -0,007 Government Sector (% of GDP) Total resources (including grants) 0.0 3.0 3.0 3.0 3.0 2.5 2.0 1.5 1S Total revenues 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Direct taxes 0.0 0.0 0.0 0.0 0.0 0,o 0.0 0.0 0.0 Indirect taxes 0,o 0.0 0,o 0,o 0.0 0.0 0,o 0.0 0.0 User fees 0.0 0.0 0,o 0,o 0,o 0.0 0,o 0.0 0.0 Foreign aid (grants) 0,o 3.0 3.0 3.0 3.0 2.5 2.0 1.5 1.5 Total expenditure 0.0 3.0 3,O 3.0 3,O 2.5 2.0 1.5 1.5 Spending on goods and services (total) 0,o 3.0 39 3,O 3.0 2,s 2,o 13 13 Spending on maintenance 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Wages and salaries 0.0 0.0 0.0 0.0 0.0 0-0 0.0 0.0 0.0 Security 0.0 3,O 3.0 3.0 3.0 2.5 2,o 13 13 Other 0.0 0.0 0.0 0.0 0,o 0.0 0.0 0,o 0.0 Investment 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Interest payments 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Overall fscal balance including grants (cash basis) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Memorandum I t e m Private investment ( O h of GDP) 0.0 0.0 0.0 0.0 0,l 0.1 0.1 0.1 0.1 Public investment ( % of total public expenditure) 0.0 -5.8 -6.0 -6.1 -6.2 -5.3 -4.3 -3.3 3.3 Health (% of public investment) 0.0 0.0 09 0,o 0,o 0,o 0.0 0.0 0.0 Infrastructure(% of public investment) 0.0 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0,0 Education(% of public investment) 0.0 0.0 0,o 0.0 0,o 0.0 0.0 0.0 0.0 Other (% of public investment) 0,o 0.0 0.0 0.0 0.0 0,o 0.0 0,o 0,0 Aid (% of total revenue) 0.0 31.4 31,6 31.8 31.9 26.6 21.3 15.9 15.9 Total Aid ( O h of public investment) 0.0 60.7 56.1 52.2 52,2 43,6 34,9 26,2 26,2 Total debt (% of GDP) 0.0 0.0 0,o 0.0 0.0 0,o 0,o 0.0 0.0 Educatedlabor (in % of population) 0.0 0.0 0.0 0.0 Os@ 0,o 0.0 0.0 0,0 219 Haiti: Public Expenditure Management and Financial Accountability Review Table A.8.8 Higher security spending and Elasticity of Security Spending, 2007-15 (Deviation from the Baseline scenario) Years 2007 2008 2009 2010 2011 2012 2019 2014 2015 Real GDP per capita at market prices (% change) 0.000 -0,013 0,003 0,016 0.026 0.037 0,046 0,054 0,059 Poverty rate-Ravailion’sadjusted elasticity (Gini = 66.0) 0,000 0,003 0,002 -0.001 -0,006 -0,014 -0,023 -0,034 -0,046 Poverty rate-GrMh elasticity of -1.0 0,000 0,007 0,006 -0,003 6.017 -0.037 -0,061 6.090 -0,121 Government Sector (% of GDP) Total resources (including grants) 0.0 3.0 3.0 3.0 3.0 2,s 2.0 13 1.5 Totat revenues 0,o 0.0 0.0 0.0 0.0 -0.1 -0,l -0.1 -0.1 Direct taxes 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Indirect taxes 0,o 0.0 0.0 0,o 0.0 -0.1 -0.1 -0.1 -0,l User fees 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 Foreign aid (grants) 0.0 3.0 3.0 3.0 3.0 2.5 2.1 1,6 1.6 Total expenditure 0.0 3.0 3,O 3.0 3.0 2.5 2.0 1.5 1.5 Spending on goods and services (total) 0.0 39 3,o 3.0 3.0 2.5 2,o 1.5 1.5 Spending on maintenance 0.0 0.0 0.0 0,o 0,o 0.0 0,o 0.0 0.0 wages and salaries 0.0 0.0 0,o 0.0 0.0 0.0 0,o 0,o 0.0 Security 0.0 3.0 3,O 3.0 3.0 23 2,o 13 1,5 Other 0.0 0.0 0.0 0.0 0-0 0,o 0.0 0,o 0.0 Investment 0.0 0.0 0,o 0,o 0.0 0.0 0.0 0.0 0.0 Interest payments 0,o 0.0 0.0 0.0 0.0 0,o 0.0 0,o 0,o Overallfiscal balance inclLXling grants (cash basis) 0.0 0.0 0,o 0,o 0,o 0.0 0.0 0.0 0.0 Memorandum items Pnvate investment (% of GDP) 0.0 0.2 0.4 0.6 0.8 1.o 1.2 1.4 1.5 Public investment ( % of total public expenditure) 0,O -5.8 4,O 4 1 -6.1 -5.2 -4.3 -3.3 -3.2 Health (% of public investment) 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,o Infrastructure(% of public investment) 0.0 0.0 0.0 0,o 0,o 0.0 0.0 0.0 0.0 Education (% of public investment) 0.0 0.0 0.0 0.0 0.0 0.0 0,o 0.0 0.0 Other (% of public investment) 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Aid (% of total revenue) 0.0 31.2 31.5 31.7 313 26.7 21.5 16.4 16,3 Total Aid (%of public investment) 0,O 60,9 56,5 52.7 52,9 44.4 35.8 27.2 27,2 Total debt ( % of GDP) 0.0 0.0 0.0 . 0.0 0,o 0.0 0.0 0.0 -0,l Educatedlabor (in % of population) 0.0 0.0 0.0 0.0 0,o 0,o 0.0 “0.1 -0,l 220 Haiti: Public Expenditure Manasement and Financial Accountability Review Table A.8.9 Combined shock with lower collection cost, higher direct tax and security spending, 2007-15 (Deviation from the Baseline scenario) Years 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real GDP per capita at market prices (O h change) 0.0 0.1 03 0.6 0,6 0.7 0.7 0.6 0.5 Poverty rate-Ravallion'sadjusted elasticity (Gini = 66 0) 0.0 0.0 -0.1 -0.2 -0.4 -0.5 -0.6 -0.8 -0.9 Poverty rate-Growth elasticity of-1.0 0,o 0.0 -0.3 -0.6 -0,9 -1.3 -1,7 -2.0 -2.2 Government Sector (% of GDP) Total resources (including grants) 0.0 8,1 82 8.3 8.4 7.0 5.6 42 3.2 Total revenues 0.0 1.6 2.6 3.6 3.7 3.8 4.0 4.1 4.2 Direct taxes 0.0 1,l 2.1 3.0 3.0 3.0 3.0 3.0 3.0 Indirect taxes 0.0 03 0.3 0.3 0.2 0.2 0.2 0.1 0.1 User fees 0.0 0,l 0.2 0,s 0.5 06 08 0.9 1,o Foreign aid (grants) 0.0 63 5.6 4.6 4.6 3.1 1.6 0,l -0,9 Total expenditure 0.0 8.1 82 8.3 8.4 7.0 5.6 4,2 3.2 Spending on goods and sewices (total) 0.0 2.6 2.7 2.8 2.9 2S 2,l 1.7 13 Spending on maintenance 0.0 0.1 0.2 0.3 0.4 0,6 0.7 0.8 03 Wages and salaries 0.0 0,o 0.0 0.0 -0.1 -0.1 -0.1 -0,l -0.2 Security 0.0 3,O 3.0 3.0 3,O 23 2,o 1.5 1.5 Other 0.0 4 5 -0.5 -0.5 -05 -0,5 -05 -0,5 -0,5 Investment 0.0 5.5 5.5 5.5 5.5 4.5 3,5 23 1.5 Interest payments 0.0 0.0 0.0 0.0 0.0 0,o 0,o 0.0 0,o Overall fiscal balance including grants (cash basis) 0.0 0.0 0.0 0.0 0.0 0,o 0,o 0.0 0.0 Memorandum items Private investment ( O h of GDP) 0.0 -1,2 -18 -2.3 -18 -12 -0.6 -0,l 0.4 Public investment ( O h of total public expenditure) 0.0 12.1 11.1 10.2 10,o 8.5 6.8 4.8 1.5 Health ( O h of public investment) 0.0 5.0 5,O 5,O 5.0 5,O 5.0 5.0 5.0 Infrastructure( Oh of public investment) 0.0 0.0 0.0 0.0 0.0 0,o 0.0 0,o 0,o Education (YOof public investment) 0,o 0.0 0,o 0.0 0,o 0,o 0.0 0.0 0.0 Other (% of public investment) 0.0 -5.0 -5,O -5.0 -50 -5,O -5,O -5.0 -5.0 Aid (% of total revenue) 0.0 52.6 35.4 19.9 19,Z 7.3 -4.5 -16.0 -23.6 Total Aid (YOof public investment) 0.0 16.5 62 -3.3 -33 -9,7 -17,2 -26,6 -31,6 Total debt (% of GDP) 0.0 0.0 -0.1 -0.3 -0.5 -0.7 -0.8 -1.0 -1.1 Educated labor (In 46 of population) 0,000 0,000 0,000 0,002 0.004 0,007 0,010 0.013 0,016 22 1 Haiti: Public Expenditure Management and Financial Accountability Review Table A.8.10 Combined shock with higher security expenditure, direct taxes, and elasticity of security expenditure, 2007-15 (Deviation from the Baseline scenario) Years 2007 2008 2009 2010 201t 2012 2013 2014 2015 Real GDP per capita at market prices (% change) 0.0 0.0 0.5 0.6 0.6 0.7 0.7 0.7 0.6 Poverty rate-Ravallion's adjusted elasticity (Gini = 66.0) 0.0 0,o 6.1 -0,2 -0.4 -0.5 -0.6 -0.8 -0,9 Poverty rate-Growth elasticity of -1 0 0.0 0.0 -0.3 -0,6 -1.0 -1,3 -1.7 -2.0 -2,3 Government Sector (x of GDP) Total resources (includinggrants) 0.0 8.1 8.2 8.3 8.4 6.9 5.5 4.1 3.2 Total revenues 0.0 1.6 2,6 33 3.7 3.8 3.9 4.0 4,1 Direct taxes 0.0 1,l 2.1 3.0 3.0 3,O 3.0 3.0 3.0 Indirect taxes 0.0 03 03 0.2 0.2 0.2 0.1 0,1 0.0 User fees 0,o 0.1 0.2 0,s 0.5 OB 0.8 0.9 1.1 Foreign aid (grants) 0,o 6.5 5.6 4.6 4.6 3.1 1.6 0.1 -0,9 Total expenditure 0.0 8,l 8.2 8,3 8,4 6,9 5.5 4.1 3.2 Spending on goods and sewices (total) 0.0 2.6 2.7 2,8 2.9 23 2,l 1,7 1.8 Spending on maintenance 0.0 0.1 0.2 0.3 0.4 0.6 0.7 0.8 0.9 Wages and salaries 0.0 0.0 0.0 0,o -0.1 -0.1 -0.1 -0.1 -02 Security 0.0 3,O 3.0 3.0 3.0 2,s 2.0 1.5 1.5 Other 0.0 -0,5 -0.5 -0.5 -0.5 -0.5 -0.5 -0.5 -0,5 Investment 0.0 5,5 5.5 5.5 53 4.5 3.5 2.5 1.5 interest payments oxo 0,o 0.0 0.0 0.0 0.0 0.0 0,o 0.0 Overall fiscal balance including grants (cash basis) 0.0 0.0 0.0 0,o 0.0 0,o 0,o 0.0 0.0 Memorandum items Private investment (% of GDP) 0,o -1.1 -1.4 -1.7 -1.0 -0.1 0.7 1.4 2,l Public investment (% of total public expenditure) 0.0 12.1 11.1 10.2 10,o 8,5 6,8 4,9 1.6 Health (% of public investment) 0.0 5,O 5.0 5.0 5.0 5.0 5.0 5,O 5,O Infrastructure(% of public investment) 0.0 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 Education(% of public investment) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Other (% of public investment) 0.0 -5.0 -5.0 -5.0 -5.0 -5.0 5.0 -5.0 -5.0 k d (% of total revenue) 0.0 52.3 35.6 20.6 202 8,7 -2,6 -13.6 -20,7 Total Aid (% of public Investment) 0.0 17.7 7,6 -1.6 -1.5 -7.6 -15.0 -24.4 -29.7 Total debt (% of GDP) 0.0 0.0 -0.1 -0.3 -0.5 -0,7 -0.6 -1.0 -1,l Educated labor (in % of population) 0,000 0,000 -0.003 -0,007 6.013 -0,021 -0,031 -0.043 -0,057 222 Haiti: Public ExDenditure Manaaement and Financial Accountabilitv Review TabEe A.8.11 Combined shock with lower collection cost, higher security spending, and direct tax incresing later, 2007-15 (Deviation from the Baseline scenario) 2007 2008 2009 2010 2011 2012 2013 2014 2015 Real GDP per capita at market prices (% Change) 0,o 0.0 0.6 0.8 0.8 0.8 0,7 0.6 0.6 Poverty rateliavaliion's adjusted elastiuty (Gin! = 66 0) 0.0 0.0 -0.1 -0.3 -0.4 -0,6 -0,8 -0.9 -1,0 Poverty rateGrowth elasticity of -1.O 0,o 0.0 -0.3 -0.7 -1.2 -1,6 -2,0 -2.3 -2,6 Government Sector (X of GDP) Total resources (includinggrants) 0.0 8.1 8.2 8.2 8.3 6.9 5.5 4.1 3.2 Total revenues 0.0 0,6 0,6 1.2 1.8 2.4 2.9 3.1 3.2 Direct taxes 0.0 0.1 0.1 0,6 1.1 1.6 2.1 2.1 2.1 Indirect taxes 0.0 0.3 0.3 0.2 02 0.2 0.1 0.1 0.1 User fees 0,o 0,l 0.2 0.3 03 0.6 0.8 03 1.o Foreign aid (grants) 0.0 7s 7.5 7,O 6.6 4.6 2.6 I,? 0.0 Total expenditure 0,o 8,l 8.2 8.2 8,3 6.9 5.5 41 3.2 Spending on goods and sewices (total) 0.0 2.6 2.7 2.8 2.8 2.4 2.0 1.6 1.7 Spending on maintenance 0,o 0,l 0.2 0.3 0.4 0.5 0.7 0.8 0.9 Wages and salaries 0.0 0,o 0,o 0.0 -0.1 -0,l -0.1 -0.2 -0.2 Security 0,o 3,O 3.0 3.0 3.0 2.5 2.0 1.5 1.5 Other 0,O -0,5 -0.5 -0.5 -0.5 -0.5 -0.5 -0.5 -0.5 investment 0.0 53 53 53 5s 43 3,5 2.5 1.5 Interest payments 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Overall fiscal balance including grants (cash basis) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Memorandumitems Private investment ( O h of GDP) 0.0 -0.2 0.3 0,5 0,7 0.8 0.9 1.4 1.9 Public investment (% of total public expenditure) 0.0 12,l 11.1 10.2 10,o 8,6 6.9 4.9 1.6 Health ( % of public investment) 0.0 5.0 5,o 5.0 5.0 5.0 5.0 5.0 5.0 Infrastructure ( O h of public investment) 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Education (% of public investment) 0.0 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 Other (% of public investment) 0,O -50 -5.0 -5.0 -5,O -5.0 5.0 -5,O -5.0 Aid (%oftotal revenue) 0.0 71.5 71.0 59.9 49,7 27.7 7.6 -5.0 -13.4 Total Aid ( O h of public investment) 0.0 25.8 24.1 18.4 13,9 4.6 -6.6 -14,7 -16.2 Total debt ( % of GDP) 0.0 0,O -0,2 -0.4 -0.6 -0.8 -1.0 -1.1 -1.3 Educated labor (in % of population) 0.000 0,000 0,000 0,000 0,001 0,002 0,004 0,008 0,009 223 Haiti: Public Expenditure Management and Financial Accountability Review Table A.8.12 Combined shock with lower collection cost, higher security spending, elasticity of security spending, and direct tax incresing later, 2007-15 (Deviation from the Baseline scenario) Years 2007 2008 2009 2010 2011 2012 2013 2014 2015 * Real GDP per capita at market prices (% change) 0.0 0.0 0.6 03 0.8 0.9 0.8 0.7 0.6 Poverty rate-Paallion's adjusted elasticity (Gini = 66.0) 0.0 0.0 -0,l -0,3 -0,4 -0.6 -0.8 -0.9 -l,o Poverty rate-Grovuth elasticity of -1.0 0.0 0.0 -0.3 -0,7 -1.2 -1,6 -2.0 -2,4 -2.7 Government Sector (% of GDP) Total resources (includinggrants) 0,o 8.1 8.2 8.2 8.3 6.9 5.5 4.1 3.2 Total ievenues 0.0 0,6 0.6 1.2 1.7 23 2,s 3,O 3.1 Direct taxes 0.0 0.1 0.1 0.6 1.1 1.6 2.1 2,1 2.1 Indirect taxes 0.0 03 0,s 0.2 0.2 0.1 0.1 0,o 0.0 User fees 0.0 0.1 0.2 0.3 03 0.6 08 03 1.0 Foreign aid (grants) 0.0 73 7.5 7.1 6.6 4.6 2.6 1.1 0.1 Total expenditure 0.0 8.1 8,2 8.2 8.3 69 53 4.1 3.2 Spending on goods and services (total) 0.0 2.6 2.7 2.8 2.8 2.4 2.0 1.6 1.7 Spending on maintenance 0.0 0.1 0.2 0.3 0.4 0.5 0.7 0.8 0.9 Wages afld salaries 0.0 0.0 0.0 0,o -0.1 -0.1 % I - -0,2 -0,2 Secunty 0,o 3.0 3.0 3.0 3.0 23 2.0 1.5 1.5 Other 0.0 -0.5 -0.5 -03 -0.5 -0.5 -0.5 -0.5 -0,5 Investment 0.0 5.5 5.5 5.5 5.5 4.5 3.5 2.5 1.5 Interest payments 0.0 0,o 0.0 0.0 0.0 0.0 0.0 0,o 0.0 Overall fiscal balance including grants (cash basis) 0.0 0.0 0.0 0,o 0.0 0.0 0.0 0.0 0,o Memorandumitems Private investment (% of GDP) 0.0 0,o 0.7 1.1 1.5 1.9 2.2 3,O 3.7 Public investment (% of total public expenatwe) 0.0 12.1 11.1 10,2 10,o 8.6 6.9 5,O 1,7 Health (% of public investment) 0,o 5, O 5.0 5,O 5.0 5.0 5.0 5,O 5,O Infrastructure(% of public investment) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0,o 0.0 Education(% of public investment) 0,o 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Other (% of public investment) 0.0 -5.0 -5.0 -5.0 -5.0 -5.0 -5.0 -5.0 -5,O Aid (% of total revenue) 0,o 70.9 70.4 59,5 493 28.4 9.0 -3,l -11,o Total Aid (%of public investment) 0,o 27.0 25.6 20,o 15.9 6.7 -4.5 -12.6 -16,3 Total debt (% of GDP) 0.0 0.0 -0.2 -0,4 -0,6 -0,8 -1 .o -1,2 -1.3 Educated labor (in % of population) 0.000 0,000 -0.003 -0,009 0.016 -0.026 -0.037 -0.049 -0.063 224