SOWING THE SEEDS OF A SUSTAINABLE FUTURE NORTH MACEDONIA | PUBLIC FINANCE REVIEW Report No. 135704 – MK Republic of North Macedonia Public Finance Review Sowing the Seeds of a Sustainable Future March 2019 © 2018 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: “World Bank. 2018. North Macedonia Public Finance Review: Sowing the Seeds of a Sustainable Future. Washington, D.C.: The World Bank.” All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. SOWING THE SEEDS OF A SUSTAINABLE FUTURE Contents Acknowledgementsxvi Acronymsxvii Overview1 Chapter 1: The Need for Fiscal Reforms 10 1.1  Country Context and Fiscal Dynamics 10 1.2  Government Revenues 20 1.3  The Level and Composition of Public Spending 21 1.3.1  Spending by Economic Classification  22 1.3.2  Spending by Functional Classification 24 1.3.3  Budget Rigidity 26 1.4  Efficiency of Public Expenditure 27 1.4.1  Economic Efficiency 27 Chapter 2: The State of Public Financial Management 30 2.1  Sector Overview 30 2.2  Institutional Shortcomings Related to PFM: Selected Issues 32 2.2.1  Policy-based Fiscal Strategy Development and Budgeting (Formulation, Management and Monitoring) 32 2.2.2  Budget Executon and Control Process 39 2.2.3  Evaluation, Reporting and Monitoring of Government Liabilities 41 2.2.4  Public Sector Internal Financial Control and Audit  44 2.3  Reform Options 45 Chapter 3: Fiscal Decentralization and Service Delivery 46 3.1  Sector Context 46 3.2  Institutional Issues and the Vertical Structure of Government 48 3.3  Expenditure Assignments: Implications for Efficiency and Equity  50 3.4  Reinforcing the Local Revenue Base: Tax Assignments  54 3.5  The Intergovernmental Transfer System 56 3.6  Municipal Indebtedness and Borrowing Controls 62 3.7  Reform Options 64 Chapter 4: Tax Policy Options for Increasing Revenue 66 4.1  North Macedonia’s Tax System in International Context 66 4.1.1  Effective Tax Rates on Factor Incomes and Consumption 69 4.2  Personal Income Tax 71 4.3  Indirect Taxes: Vat and Excise Duties 75 4.3.1 VAT 76 4.3.2  Excise Duties 78 4.4  Corporate Income Tax 81 4.4.1  Technological Industrial Development Zones 82 Contents v REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 4.5  Taxation of Immovable Property 84 4.6  Tax Compliance and Administration 86 4.7  Reform Options 86 Chapter 5: Ensuring the Sustainability of the Pension System 89 5.1 Introduction 89 5.2 Recent Developments and Features of North Macedonia’s Pension System 91 5.3  Sustainability of the Pension System 99 5.3.1  Baseline Scenario 99 5.3.2  Alternative Policy Simulations 102 5.4  Conclusions and Reform Options 106 5.4.1  Reform Options 106 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable 108 6.1  The Social Protection System 108 6.2  Social Assistance Spending 109 6.3  Performance of Social Assistance Programs 113 6.3.1 Coverage 113 6.3.2  Distribution of Benefits and Beneficiaries 115 6.3.3  Contribution to Consumption 116 6.3.4  Impact of Social Transfers on Poverty 117 6.4 Conclusions 119 6.5  Reform Options 120 Chapter 7: Making the Grade on Education 122 7.1  Sector Overview 122 7.2  North Macedonia’s Spending on Education in International Context 129 7.2.1  Public Spending Trends and Composition 129 7.3  Efficiency of Spending 133 7.3.1  Efficiency of Resource Use 134 7.4  Financing Mechanisms 137 7.5 Intra-national Assessment of North Macedonia's Education Spending and Outcomes 139 7.6  Reform Options 140 Chapter 8: Strengthening Public Spending to Improve Health Outcomes 143 8.1 Introduction 143 8.2  Health Outcomes and Services 144 8.2.1  Life Expectancy 144 8.2.2  Maternal and Child Health  145 8.2.3  Noncommunicable Diseases  145 8.2.4  Quality of Care 148 8.2.5  Financial Risk Protection 150 8.3  Total Health Spending: Trends and Relative Health System Efficiency 151 8.3.1 Trends 151 8.3.2  Relative Health System Efficiency 153 8.4  Public Health Spending 153 vi Contents SOWING THE SEEDS OF A SUSTAINABLE FUTURE 8.4.1 Overview 153 8.4.2  Allocative Efficiency  155 8.4.3  Cost-Control Reforms 157 8.5  Reform Options 161 Chapter 9: Rebalancing Government Support to Improve Agriculture 163 9.1 Introduction 163 9.1.1  Rationale and Objectives  163 9.1.2  Methodological Approach 164 9.2  Agricultural Structure, Challenges, and Strategic Priorities 164 9.2.1  Structural Characteristics 164 9.2.2  Policy Priorities and Institutions 168 9.3  Public Spending on Agriculture and Rural Development 169 9.3.1  Total Public Spending 169 9.3.2  Spending by Economic Category 170 9.3.3  Spending by Function 171 9.3.4  Spending by Category of Support 171 9.4  Analysis of the Efficiency, Effectiveness, and Equity of Spending 173 9.4.1  Efficiency of Spending 173 9.4.2  Spending Effectiveness 179 9.4.3  Spending Equity 183 9.5  Reform Options 185 References187 Annex 1. The Government Program, 2017–20 193 Annex 2. The Fiscal Impact of Reform Options 195 Annex 3. Investment Tax Incentives in Selected Countries 198 Annex 4. Education Regressions 204 Annex 5. Funding Formulas for Education 206 Annex 6. Average Net Wages 208 Annex 7. Air Pollution in North Macedonia 209 Annex 8. Health Spending in North Macedonia and Comparators 210 Annex 9. Structure of Agricultural Output, North Macedonia 211 Annex 10. Guaranteed Minimum Income Comparisons 212 Contents vii REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 List of Figures Chapter 1: The Need for Fiscal Reforms Figure 1.1. GDP per Capita Growth Rates, 2001–08 and 2008–16 Averages, 11 Figure 1.2. Income Convergence with the European Union in Different Growth Scenarios 11 Figure 1.3. Fiscal Policy Trends of the General Government, 2001–17 12 Figure 1.4. Central Government Budget Execution, 2006–17 12 Figure 1.5. North Macedonia’s PPG Debt, 2006–17 14 Figure 1.6. PPG Debt-to-GDP Ratio, North Macedonia and Comparator Countries, 2008 and 2017 14 Figure 1.7. North Macedonia’s PPG Debt, Contribution to Change as Percent of GDP, 2008–17 14 Figure 1.8. General Government Gross Financing Needs, North Macedonia and Comparators, 2017 14 Figure 1.9. Maturity Profile, North Macedonia’s General Government Debt, 2017–35 15 Figure 1.10. Average Time to Maturity, General Government Debt, 2012–17), Months 15 Figure 1.11. North Macedonia’s General Government Debt Composition, 2012–17 16 Figure 1.12. North Macedonia’s General Government Debt Composition by Currency, 2017 16 Figure 1.13. North Macedonia PPG Debt, 2015–21 17 Figure 1.14. North Macedonia Gross Financing Needs, 2015–21 17 Figure 1.15. Debt Sustainability Scenarios: North Macedonia’s PPG Debt, 2015–21 17 Figure 1.16. General Government Revenues, 2003 and 2017 20 Figure 1.17. Composition of Public Revenues in North Macedonia, 2009–17 20 Figure 1.18. General Government Spending, 2014–17 Average 21 Figure 1.19. Cumulative change in Expenditures, 2009–17 Average 21 Figure 1.20. General Government Expenditure by Economic Classification, 2017 21 Figure 1.21. Central Government Expenditure by Functional Classification, 2017 21 Figure 1.22. Budgeted and Realized Capital Expenditures, 2009–17 24 Figure 1.23. Structure of Completed Public Construction Works, 2006–17 24 Figure 1.24. Budget Rigidity, 2011–17 27 Figure 1.25. Public Sector Performance Index, 2006 28 Figure 1.26 Public Sector Performance Index, 2016 28 Figure 1.27. Public Sector Efficiency Index by Sector Spending, 2016 28 Figure 1.28. Public Spending by GDP per Capita 2016 29 Figure 1.29. Public Sector Performance Score by Sector Spending, DEA Analysis, 2016 29 Chapter 2: The State of Public Financial Management Figure 2.1. PEFA Comparison 31 Figure 2.2: North Macedonia: Structure of the Budget 32 Figure 2.3. Original and Revised Budgets and Actual Revenues, 2014–16 35 Figure 2.4. Original and Revised Budgets and Actual Expenditures, 2014–16 35 Figure 2.5. Original Budget and Actual Revenues, 2014–16 35 Figure 2.6. Original Budget and Actual Expenditures, 2014–16 35 Figure 2.7. Ranking on the Open Budget Index 36 Figure 2.8. Elements of Public Participation 36 Figure 2.9. Self-Reported Unpaid Expenditure Obligations, May 2017 41 viii Contents SOWING THE SEEDS OF A SUSTAINABLE FUTURE Chapter 3: Fiscal Decentralization and Service Delivery Figure 3.1. Total Revenues Per Capita, MKD Municipalities, 2016 47 Figure 3.2. Total Transfers Per Capita, MKD Municipalities, 2016  47 Figure 3.3. Teachers and Students in Primary and Secondary Education, 2000/2001–2016/2017 53 Figure 3.4. Structure of LG Revenues, 2016 55 Figure 3.5. Local Government Revenues, WB6, 2015 55 Figure 3.6. Own Revenues as a Share of Total LG Revenues, 2016 55 Figure 3.7. Local Property Taxation, 2016 55 Figure 3.8. Structure of Block Transfers, 2006–16 59 Figure 3.9. Block Transfer, 2008–16 59 Figure 3.10. Structure of Capital Spending by Local Governments, 2016 61 Figure 3.11. Breakdown of Arrears, May 2017 63 Figure 3.12. Local Government Budget Arrears, May 2017 63 Chapter 4: Tax Policy Options for Increasing Revenue Figure 4.1. General Government Revenues, 2016 67 Figure 4.2. Changes in General Government Revenues, 2006–16 67 Figure 4.3. Personal Income Tax: Number of Taxpayers and Effective Tax Rate, 2017 72 Figure 4.4. Taxable Income by Source Along the Income Distribution 72 Figure 4.5. Simulated Tax Revenue Gains. Reform Scenario with Uniform Rate for Income Sources 73 Figure 4.6. Simulated Tax Revenue Gains from Reform Scenario  74 Figure 4.7. Simulated Tax Revenue Gains from Reform Scenario with Rate and Base Changes 75 Figure 4.8. VAT Revenue Ratio Compared, 2016 76 Figure 4.9. Excise Revenue by Source, 2016 79 Figure 4.10. Tobacco Excise Revenue, 2007–16 79 Figure 4.11. Effective Average CIT Rates Compared, 2016 81 Figure 4.12. Profit Tax and Net Operating Surplus, 2006–16 81 Figure 4.13. Top Business Environment Obstacles for Firms Operating in North Macedonia 82 Chapter 5: Ensuring the Sustainability of the Pension System Figure 5.1. Pension System Finances, 2001–16 91 Figure 5.2. Total Pension Spending as a Share of GDP and Old Age Dependency Ratio, 2015 91 North Macedonia – Insured Individuals, Pensioners, and Pension System Figure 5.3.  Dependency Ratio, 2001–16 92 Figure 5.4. System Dependency Ratios Compared, 2013 (2016 for North Macedonia) 92 Figure 5.5. Remaining Life Expectancy at Retirement 93 Figure 5.6. North Macedonia: Categories of Pension Beneficiaries, 1991–2016 93 Figure 5.7. Expected and Actual Indexation of Pensions in North Macedonia, 2008–16 98 Figure 5.8. North Macedonia: Average Pension Replacement Rate and Pension Spending, 2001–16 98 Figure 5.9. Replacement Rates in North Macedonia and Selected Countries, 2015 99 Figure 5.10. Public Pension Insurance Contribution Rates, Selected Developed and CEE Economies, 2016–17 99 Figure 5.11. North Macedonia - Demographic Projection, 2017–80 100 Figure 5.12. North Macedonia: Pension Contributors and Beneficiaries, Baseline Simulation, 2017–80101 Figure 5.13. North Macedonia: Baseline New Pension Replacement Rates, 2018–80 101 Contents ix REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 5.14. North Macedonia: Baseline Average Pension Replacement Rates, 2018–80 101 Figure 5.15. Baseline Pension Expenditures, Revenues and Deficit, 2017–80 102 Figure 5.16. Option Simulations: Average PAYG Replacement Rates, Male Nonswitchers, 2017–77 103 Option Simulations: Average PAYG Replacement Rates, Male Voluntary Switchers, Figure 5.17.  2018–75103 Figure 5.18. Option Simulations: Average PAYG Replacement Rates, Male Mandatory Switchers, 2017–77103 Figure 5.19. Option Simulations: PAYG Total Spending, 2017–80 103 Figure 5.20. Option Simulations: PAYG Deficit, 2017–80 104 Figure 5.21. Policy Package MIX1, Average Replacement Rates, 2017–77 105 Figure 5.22. Policy Package MIX1, PAYG Spending, 2017–80 105 Figure 5.23. Policy Package MIX1, PAYG Deficit, 2017–80 105 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable Figure 6.1. Social Protection Spending, 2014 or Latest Data 110 Figure 6.2. Spending on Social Assistance Programs, 2005–17 110 Figure 6.3. Number of Recipients for Selected Benefits. 2005–17 110 Figure 6.4. Means-Tested and. Non-Means-Tested Benefit Spending, 2005–17 111 Figure 6.5. Spending on Social Assistance by Targeting Method, Local Currency Units 112 Figure 6.6. Social Assistance Coverage of the Poorest Quintile, 2016 or Latest Data 114 Figure 6.7. Last Resort Social Assistance: Poorest Quintile Coverage, 2016 or Latest Data 114 Figure 6.8. Distribution of Social Assistance Beneficiaries by Quintile 115 Figure 6.9. Distribution of Social Assistance Benefits by Quintile 115 Figure 6.10. Last Resort Social Assistance, Percent of Benefits Received by the Poorest Quintile a 115 Figure 6.11. Distribution of Benefit Amounts per Household, MKD 116 Figure 6.12. Benefit-Cost Ratios 118 Chapter 7: Making the Grade on Education Figure 7.1. Net Attendance Rates by Economic Quintile, 2011 123 Figure 7.2. Net Attendance Rates by Location, 2011 123 Figure 7.3. Completion Rates by Level of Education, 2011 124 Figure 7.4. Rate of Students Not Attending by Level of Education, 2011 124 Figure 7.5. Enrollees in Higher Education by Previous Secondary Program, 2011–16 125 Figure 7.6. Student PISA Performance in Math, 2015 125 Figure 7.7. Change in PISA Reading Scores, 2000–15 126 Difference in PISA 2015 Science Scores Between Students with less than One Year of Figure 7.8.  Preschool and Students with more than One Year 126 Simulation of North Macedonia’s PISA Score Gains from Increasing Preschool Figure 7.9.  Attendance126 Figure 7.10. Teachers with Tertiary or Higher Education across Municipalities, 2016/17 127 Figure 7.11. Average Score in State Matura Examination by Municipality, 2016/17 127 Figure 7.12. Math Scores by Student Economic Status 127 Figure 7.13. Socioeconomic Index for Students and Schools Correlated (weighted) 127 Figure 7.14. Difference in Science Scores, Top and Bottom Quintile Students, 2015 128 Figure 7.15. Difference in Science Scores, Urban and Rural Students, 2015 128 Figure 7.16. PISA Math Scores and Youth Unemployment by Country, 2015 128 x Contents SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 7.17. Public Spending on Education, 2016 129 Figure 7.18. Spending on Education, 2011–16 131 Figure 7.19. Spending on Education Compared, 2013–15 131 Figure 7.20. Spending on Public Education Spending by Level Compared, 2016 132 Figure 7.21. Public Spending on Education by Economic Classification, 2016 132 Figure 7.22. Efficiency Scores Using Net Lower Secondary Education Enrollment as Output, 2015 133 Figure 7.23. Efficiency Scores Using PISA 2015 Math Scores as Output 134 Figure 7.24. PISA Performance and GDP per Capita, 2000 and 2015 134 Figure 7.25. Evolution of the Primary and Lower Secondary School Network, 2007–16 135 Figure 7.26. Evolution of the Upper Secondary School Network, 2007–16 135 Figure 7.27. Hours of Instruction Time in Compulsory General Education Compared 136 Figure 7.28. Annual Basic Gross Teacher Salaries Compared 137 Figure 7.29. Government Education Spending by Level and Source of Financing, 2016 137 Figure 7.30. Per-student Spending and Class Size by Education Level and Municipality, 2016–17 137 Figure 7.31. Matura Results and Poverty by Level of Education and Municipality, 2016–17 139 Figure 7.32. Poverty and Proportion of Block Grant Funds by Municipality, 2016–17 139 Figure 7.33. Efficiency Scores for Primary Completion Rates, 2016 140 Figure 7.34. Efficiency Scores for Secondary Completion Rates, 2016 140 Figure 7.35. Efficiency Scores for Matura Examinations 140 Chapter 8: Strengthening Public Spending to Improve Health Outcomes Figure 8.1. Population Age Distribution, 2015 and 2050 144 Figure 8.2. Life Expectancy at Birth Compared, 1995–2015 144 Figure 8.3. Life Expectancy at Birth Compared, 2015 144 Figure 8.4. Child Mortality Rates Compared, 1995–2016 145 Figure 8.5. Measles Immunization Rate, 2005 and 2016 145 Figure 8.6. Lung Cancer Incidence and Mortality Compared, 2012, per 100,000 Men 146 Figure 8.7. Cervical Cancer Screening Compared, Latest Data 146 Figure 8.8. Excise Tax, 2016 146 Figure 8.9. Wait Time for Non-acute CT Scan Compared, 2016 149 Figure 8.10. Wait Time in Emergency Departments Compared, 2016 149 Figure 8.11. Deliveries and C-Section Rate by Facility, 2016 149 Figure 8.12. Appendectomy and Abdominal and Other Hernia Procedures by Facility, 2016 149 Figure 8.13. Self-Reported Unmet Needs for Medical Care due to Cost by Income Quintile, 2015 150 Figure 8.14. OOP and Public Health Expenditure as a Share of Total Health Spending, 2002–14 150 Figure 8.15. OOP Health Expenditures as a Share of Total Health Spending, 2014 150 Figure 8.16. Financial Risks from OOP Expenditures for Surgical Care Compared, 2014 151 Figure 8.17. Total Health Spending as a Share of GDP Compared, 1995–2015 152 Figure 8.18. Health Spending per Capita, PPP, Compared, 1995–2015 152 Figure 8.19. Nurses per 100,000 Inhabitants Compared, Latest Available Data 152 Figure 8.20. Physicians per 100,000 Inhabitants Compared, Latest Available Data 152 Figure 8.21. Hospital beds per 100,000, Compared, 1995–2015 152 Figure 8.22. Healthy Life Expectancy and Health Spending per Capita, 2014–15 153 Figure 8.23. Public Health Spending, 2014 154 Figure 8.24. General Government Health Spending, 2014 154 Figure 8.25. Public Health Spending, 2014 154 Contents xi REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 8.26. Capital Spending as a Share of Public Health Spending, 2012–16 156 Figure 8.27. Allocated, Adjusted, and Executed Budget for Capital, 2012–16 156 Figure 8.28. Bed Occupancy Rate Compared, Latest Available Year 158 Figure 8.29. Acute Respiratory Admissions by Facility, 2016 158 Figure 8.30. Diabetes and Hypertension Admissions by Facility, 2016 159 Figure 8.31. Number of Prescriptions and Total Outpatient Drug Reimbursements, 2008–16 161 Chapter 9: Rebalancing Government Support to Improve Agriculture Figure 9.1. Agricultural Value Added, Percent of GDP, 2016, and Percent of Employment, 2015 164 Figure 9.2. Agricultural Value–Added and GDP Growth, 2010–16 165 Figure 9.3. Agricultural Value-Added and GDP Growth, 2007–16 Average 165 Figure 9.4. Contribution of Agriculture to the Economy, by Region, 2014 165 Food, Beverages, and Tobacco, 2011, Percent of Value Added in Manufacturing Figure 9.5.  Compared165 Figure 9.6. Food Trade, 2011–16 166 Figure 9.7. Food Trade Compared, 2015 166 Figure 9.8. Agricultural Trade Balance, 2016 166 Figure 9.9. Distribution of Farms by Value in Euros Compared, 2013 167 Figure 9.10. Annual Work Units per Agricultural Holding Compared, 2013 168 Figure 9.11. Farm Household Members by Age Compared, 2013 168 Figure 9.12. Total Public and Agriculture Spending, 2011–16 169 Figure 9.13. Agricultural Value Added and Spending, 2014 169 Figure 9.14. Total Budgetary Transfers to Agriculture Compared, 2010–15 Average 170 Figure 9.15. Budget Planning, Execution and Absorption, 2011–16 170 Figure 9.16. Wages & Salaries and Capital Spending in Total Public and Agriculture Spending, 2011–16170 Figure 9.17. Agriculture Spending by Category, 2011–16 170 Figure 9.18. Agriculture Spending by Function, 2011–16 171 Figure 9.19. MSDP: Budget Planning, Execution and Absorption, 2011–16 171 Figure 9.20. MSDP: Types, 2012–16 171 Figure 9.21. MSDP by Sector, 2013–16 172 Figure 9.22. Rural Development: Budget Planning, Execution and Absorption, 2011–16 173 Figure 9.23 Rural Development Support by Priority Axis, 2012–16 173 Figure 9.24. Rural Development Support by Measure, 2013–16 173 Figure 9.25. Difference Between Capital Output Elasticity and Subsidy intensity 181 xii Contents SOWING THE SEEDS OF A SUSTAINABLE FUTURE List of Tables Chapter 1: The Need for Fiscal Reforms Table 1.1. General Government Fiscal Accounts 12 Table 1.2. North Macedonia’s Central Government Debt: Cost and Risk Indicators as of 2017 15 Table 1.3. Budget Rigidity Variance-Covariance Matrix (2006–16) 27 Chapter 2: The State of Public Financial Management Table 2.1. North Macedonia - Budget Calendar, 2014–16 33 Table 2.2. Comprehensiveness of Budget Documentation 36 Table 2.3. Public Access to Fiscal Information 38 Chapter 3: Fiscal Decentralization and Service Delivery Table 3.1. Distribution of LGs by Population 49 Table 3.2. Subnational Social Spending Responsibilities, Selected ECA Countries 51 Table 3.3. Local Government Spending, 2009–16 51 Chapter 4: Tax Policy Options for Increasing Revenue Table 4.1. North Macedonia: Government Budgetary Operations, 2009–17 67 Table 4.2. General Government Tax Revenues, 2016 or Latest Year Available 68 Table 4.3. VAT, Corporate and Individual Income Tax, and Social Security Contribution Rates, 2017 68 Table 4.4. Effective Tax Rates Compared, North Macedonia and EU Peers 69 Table 4.5. Additional Revenue from Removing Preferential Treatment 77 Table 4.6. Excise Tax Rates, North Macedonia and the EU Minimum 79 Table 4.7. Cigarette Prices and Taxes in North Macedonia and Neighboring Countries, 2016 80 Table 4.8. PIT Treatment of Capital Gains from the Sale of Real Estate 85 Chapter 5: Ensuring the Sustainability of the Pension System Table 5.1. Pension Parameters in North Macedonia 90 Table 5.2. Regular, Early, and Late Retirement in OECD, EU, and FYR Countries 94 Table 5.3. Effective Gross Accrual Rates, EU and North Macedonia, 2013–60 95 Table 5.4. Pension Valorization and Indexation Rules, OECD and Eastern Europe Countries 96 Table 5.5. Macroeconomic Assumptions for Baseline Simulation, 2018–80 100 103 Table 5.6. Individual Pension Policy Scenarios Simulated with the North Macedonia Pension Model Table 5.7. Pension Policy Package Simulated with the North Macedonia Pension Model 105 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable Table 6.1. PA Recipients and Possible Spending Scenario, 2018–28 112 Table 6.2. Social Protection Coverage by Quintile 113 Table 6.3. Benefit Incidence (All Households) 116 Table 6.4. Adequacy of Benefits, Direct and Indirect Beneficiaries 117 mpact of Various Programs on Poverty Measures, Simulating the Absence of the Table 6.5. I Program118 Table 6.6. Budget Estimates and Poverty Rates for Simulated SFA Transfers 119 Contents xiii REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Chapter 7: Making the Grade on Education Table 7.1. Gross Enrollment by Level of Education, 2015 123 Table 7.2. Educational Attainment of Macedonians Aged 25–64 and 25–34, 2016 124 Table 7.3. Entrance Age and Duration of Compulsory Education, 2016 130 Table 7.4. Enrollment in Public Institutions by Level of Education, 2016 130 Table 7.5. Per-Student Public Spending Compared, 2013 or Latest, Percent of GDP per Capita  132 Table 7.6. Students-Teacher Ratios by Country, 2014–15 135 Table 7.7. Evolution of Government Per-Student Allocations, 2010–15 138 Chapter 8: Strengthening Public Spending to Improve Health Outcomes Table 8.1. HIF Revenues by Category, 2016 155 Table 8.2. HIF Spending by Category, 2016 156 Chapter 9: Rebalancing Government Support to Improve Agriculture Table 9.1. Farmland Allocation and Land Productivity, 2015 167 Table 9.2. Share of Holdings by Size of Farmed Area, 2013 167 Table 9.3. Per Hectare and Per Head Direct Payments 172 Table 9.4. Technical Efficiency for Different Farm Categories 175 Table 9.5. Within-Group Analysis for Farm Groups with the Highest Average Efficiency 176 Table 9.6. Technical Efficiency of Subsidized Farms, by Farm Size and Type 177 Table 9.7. Effects of Subsidies and Net Investment on Farm Technical Efficiency 178 Table 9.8. Technical Efficiency of Farms, by Region 178  utput Elasticities of Factors of Production and Subsidy Intensity by Subsidy Status Table 9.9. O and Type 179  utput Elasticities of Factors of Production and Subsidy Intensity by Alternative Size Table 9.10. O Classifications180 Table 9.11. Tests for the Equality of Capital Output Elasticities Across Farm Groups 180 Table 9.12. Output Elasticities of Factors of Production and Subsidy Intensity, by Region 181 Table 9.13. Relationship between MSDP and SFA payments: OLS Regression Results 182 Table 9.14. Public Spending on Agriculture, Average Annual Values, 2011–16 183 Table 9.15. Policy Impacts, 2011–16: Average Annual Impacts Compared to 2013 183 Table 9.16. Distribution of Direct Payments by Farm Type, 2014 184 Table 9.17. Distribution of Direct Payments by Farm Size, 2014 184 Table 9.18. Distribution of Direct Payments by Farm Area, 2014 185 xiv Contents SOWING THE SEEDS OF A SUSTAINABLE FUTURE List of Boxes Chapter 1: The Need for Fiscal Reforms Box 1.1. The Public Enterprise for State Roads (PESR) 12 Box 1.2. North Macedonia Fiscal Strategy, 2019–21 18 Table B1.2.1. Government Macroeconomic Projections, 2018–21, Percent of GDP 18 Table B1.2.2. Government Fiscal Projections, 2018–21, Percent of GDP 19 Box 1.3. General Government Spending Compared 25 A. Wages Spending 25 B. Social Protection Spending 25 C. Pensions and Pension Related Spending 25 D. Social Assistance Spending 25 E. Subsidies 26 F. Agriculture Transfers Spending 26 G. Health Spending 26 H. Education Spending 26 Chapter 4: Tax Policy Options for Increasing Revenue Box 4.1. Effective Tax Rates on Factor Income and Consumption 70 Chapter 8: Strengthening Public Spending to Improve Health Outcomes Box 8.1. North Macedonia's Tobacco Production and Taxation 147 Figure B8.1.1. Price of a 20-cigarette Pack, Most Popular Brand Compared, 2016 147 Table B8.1.1. Share of GDP per Capita to Buy 100 Packs, Most Popular Brand Compared, 2008–16147 Contents xv REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Acknowledgements This report was prepared by a multi-sector World Bank team led by Edith Kikoni, Senior Economist. The thematic chapters were authored by the following team: Bojan Shimbov, Barbara Cunha, David Landry, Edith Kikoni and Sanja Madzarevic-Sujster (Fiscal Policy); Dinar Prihardini and Pierre Bachas (Revenue Mobilization); Andrew Mackie, Bojan Shimbov, and Hilda Shijaku (Public Financial Management); Jorge Martinez-Vazquez and Lazar Sestovic (Intergovernmental Fiscal Relations); Zoran Anusic and Miglena Abels (Pensions); Marina Petrovic and Renata Gukovas (Social Assistance Programs); Katia Marina Herrera Sosa, Bojana Naceva, Karina Acevedo, Micheline Frias. and Miguel Ruiz (Education); Dorothee Chen, Alessia Thiebaud, Olena Doroshenko and Luka Voncina (Health); and Svetlana Edmeades, Demetris Psaltopoulos and Alexandra Kontolaimou (Agriculture). Special thanks go to David Landry for his technical support and review of the report and to Irina Capita for her inputs. Special thanks also go to Ashley Taylor, Marc Schiffbauer, Anita Schwarz and Pia Schneider who served as peer reviewers of the report. The team is grateful for their rich comments and feedback. The team is also grateful for comments, at different stages of the report preparation, from Sergiy Zorya, Igor Kheyfets, Evgenij Najdov, Volkan Cetinkaya, Aylin Isik-Dikmelik, Sebastian James, Lewis Hawke, Mediha Agar, Sanja Madzarevic-Sujster, Cem Mete, Cristian Aedo, Julian Lampietti, Enis Baris, Marco Hernandez, Timothy Johnston and Simon David Ellis. Editing by Anne Grant and Patricia Carley and administrative support from Jasminka Sopova, Anita Bozinovska. and Leah Laboy are appreciated. The team was guided by Gallina Vincelette (Practice Manager), Marco Mantovanelli (Country Manager for North Macedonia), Linda Van Gelder (Country Director for the Western Balkans), Lada Strelkova (Country Operations Advisor, Western Balkans), and John Panzer (Global Practice Director). The World Bank team would like to thank the Government of the Republic of North Macedonia and representatives of all state agencies for their outstanding collaboration, and for providing knowledge, data, and advice during preparation of this report. xvi Acknowledgements SOWING THE SEEDS OF A SUSTAINABLE FUTURE Acronyms AE Aggregate expenditures NSARD National Strategy for Agriculture and AFSARD Agency for Financial Support in Rural Development Agriculture and Rural Development OBL Organic Budget Law ALMP Active labor market program OOP Out-of-pocket ATM Average time to maturity PA Parental Allowance ATR Average time to refixing PAYG Pay-as-you-go CA Child allowance PEFA Public Expenditure and Financial CAP Common Agricultural Policy Accountability CBMIS Cash Benefits Management Information PESR Public Enterprise for State Roads System PFA Permanent financial assistance CCT Conditional cash transfer PFM Public finance management CG Central government PFR Public Finance Review CGT Capital gains tax PHC Primary health care CIT Corporate income tax PIFC Public Internal Financial Control CPI Consumer Price Index PIOM Pension and Disability Insurance Fund CT Computerized tomography PISA Program for International Student DALYs Disability-adjusted life tears Assessment DEA Data envelopment analysis PIT Personal income tax DMU Decision-making unit PM Particulate matter DRG Diagnosis-related groups pp Percentage points ECA Europe and Central Asia Region PPG Public and publicly guaranteed ETR Effective tax rate PPP Public-private partnership EU European Union PRO Public Revenues Office FADN Farm Accounting Data Network PROST Pension Reform Options Simulations FDI Foreign direct investment Toolkit GP General practitioner PSE Public Sector Efficiency GMI Guaranteed Minimum Income PSP Public Sector Performance HBS Household Budget Survey SFA Social Financial Assistance HIF Health Insurance Fund SHI Social Health Insurance IFMIS Integrated Financial Management SILC Survey of Income and Living Conditions Information System SOE State-owned enterprise IFRS International Financial Reporting SRA Special Revenue Account Standards SSC Social Security Contributions INN International Non-proprietary Name STEE7 Seven Small Transition Economies of IO Input-output Europe IPARD Instrument for Pre-Accession Assistance STR Student-teacher ratio for Rural Development TADAT Tax Administration Diagnostic LG Local government Assessment Tool LGU Local government unit TE Technical efficiency MAFWE Ministry of Agriculture Forestry and TFP Total factor productivity Water Economy TIDZ Technological Industrial Development MALMED National Drug Agency Zones MLSP Ministry of Labor and Social Protection TVET Technical and vocational education and MOF Ministry of Finance training MOH Ministry of Health UIS UNESCO Institute for Statistics MOES Ministry of Education and Science UMI Upper-middle-income MSDP Market Support and Direct Payments VAT Value-added tax MTBF Medium Term Budget Framework WB6 Western Balkans 6 (countries) MTFS Medium-Term Fiscal Strategy ZELS Association of Local Self-Government NCD Non-communicable disease Units Acronyms xvii SOWING THE SEEDS OF A SUSTAINABLE FUTURE Overview The Republic of North Macedonia needs comprehensive fiscal reformto promote economic growth and improve public services while also safeguarding fiscal sustainability and keeping debt manageable. The Republic of North Macedonia is undergoing fundamental changes. The country has overcome its most severe political crisis since 2001 and has also moved decisively to resolve the long-standing dispute with Greece about its official name—which has been inflicting economic and political damage since 1991. The return of political stability has brought with it a renewed focus on economic reforms. The current government, in office since June 2017, has embarked on a new reform program and adopted an action plan to bring North Macedonia back to its European integration path (Annex 1). Reform momentum is building and progress has already been made, especially an increase in the transparency of public finances. A growth recovery has begun and sentiment has improved, reflected in part in the country’s successful return to international bond markets. After the recent prolonged period of uncertainty and limited progress on structural reforms, North Macedonia’s changed political landscape and the recent resolution of the country name issue give policymakers the opportunity to move the country beyond short-term recovery to tackle crucial longer-term development priorities. North Macedonia must now revitalize and sustain growth without jeopardizing its fiscal sustainability. After the 2009 financial crisis, a fiscal stimulus involving significant infrastructure spending and measures to stimulate investment provided several economic benefits. This was accompanied by consumption-driven initiatives that included sizable pension increases and costly employment schemes. Despite these efforts, growth has slowed, with consequences for public finances. While still lower than in peer countries, public debt in North Macedonia has expanded considerably since 2008. Increased borrowing by successive governments to cover rising and sustained primary deficits, and by state-owned enterprises (SOEs) that are implementing the public investment agenda, is primarily responsible for the rapid build-up of debt. Low tax rates, inefficient tax collection, inefficiencies in social spending, and generous government subsidies have added to the widened deficits. Without the reforms planned, in the medium term financing requirements are projected to exceed 15 percent of GDP, limiting the fiscal space for crucial public investment and public service delivery. The aging population will intensify future financial pressures. Beyond sustainability, however, is the corresponding challenge of developing higher-quality and more effective public finances. North Macedonia’s human development index is one of the lowest in the region; improvements in health and education have proved elusive. Learning outcomes are particularly deficient and contribute to high unemployment, which, despite a steep drop since 2008, is still a high 22 percent and is mostly structural, reflecting a shortage of skilled labor. North Macedonia thus finds itself at a proverbial crossroads: the policies it adopts today will determine whether it can return to a sustainable and productive fiscal path. Comprehensive fiscal and structural reforms are therefore vital. First, a credible strategy must be adopted to reduce the fiscal deficit and stabilize the public debt-to-GDP ratio. This strategy should be based on high-quality measures that produce durable results and enhance confidence. Second, the country’s growth strategy Overview 1 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 has to move beyond the past heavy reliance on fiscal stimulus toward structural reforms that produce long-term employment gains (through improved human capital) and higher productivity growth—in other words, genuinely sustainable economic development. Third, to mitigate medium- and long-term demographic pressures, pension reform is unavoidable, especially given the structural distortions that undermine the medium-term sustainability of the pension system. Finally, public spending must be made more efficient so that better public services can be delivered more effectively. This Public Finance Review (PFR) was carried out to assist the Government of the Republic of North Macedonia in finding ways to make its public finances more sustainable while also improving critical public services. Chapter 1 summarizes the country’s current fiscal condition, focusing on the roots of its fiscal challenges. Chapters 2 and 3 outline the bottlenecks to efficient public financial management (PFM) and the problems confronting social service delivery and financing at the local government level. Chapters 4 and 5 explore options to address and lessen fiscal risks and work toward fiscal consolidation. These chapters give special attention to improving the pension system—the single- largest public finance expenditure and a source of acute fiscal vulnerability—and to expanding revenue collection, both of which will be pivotal to North Macedonia’s long-term fiscal sustainability, and to any successful strategy to reduce its fiscal deficit. Chapters 6, 7, and 8 suggest alternative ways that the government can deliver higher quality and more efficient social assistance, education, and health services. Finally, Chapter 9 reviews the state of agriculture, particularly the need to reallocate government expenditures to improve efficiency and output. Topics covering the infrastructure sectors, while important for further strengthening public finances, were treated in the previous volume and are thus outside the scope of this report (World Bank 2015). Key Messages Public Financial Management. Though some aspects are well-designed, North Macedonia’s current PFM system has a number of shortcomings. Among them are an inability to produce credible macrofiscal forecasts, inadequate commitment control and monitoring, and a limited capacity to monitor and mitigate fiscal risk, all of which hinder fiscal performance. North Macedonia’s aggregate government revenues consistently fall short of budget estimates, and its expenditures are continually under-realized. Most of the revenue under-realization is because of the value-added tax (VAT), the largest source of revenue. Non-tax funds, such as capital revenues or foreign donations, are significantly overestimated. As for expenditures, spending on such relatively inflexible items as wages and transfers (pensions and other social benefits) tends to be realized as budgeted, but the recurring revenue shortfalls require the government to cut goods and services as well as capital spending, which then must be severely under-realized. There is no centralized institution for monitoring and managing fiscal risks. Risk oversight and management responsibilities are currently spread across several ministries and agencies, which means there is no consolidated perspective on, or mitigation strategy for, overall fiscal risks like arrears. Although there is reporting on both budgetary and extra-budgetary institutions, there is neither a broad consolidated view nor adequate monitoring of their financial operations—and their inherent risks. Of particular concern is the dearth of monitoring and reporting of expenditures on 2 Overview SOWING THE SEEDS OF A SUSTAINABLE FUTURE roads and other capital investments, which amounts to 5.8 percent of GDP; the PESR (the country’s former Roads Agency) was moved off-budget in 2013. The PESR’s spending and borrowing operations, as well as those of other SOEs, must be assessed in identifying the government’s fiscal risks and debt sustainability, especially in view of the sizable investment agenda for the coming years and the need to harmonize reporting of statistics with international standards. Limited capacity to produce credible, high-quality macrofiscal forecasting has led to deviations in budget execution. Budget accuracy and reliability, which form the cornerstone of any good fiscal management process, are essential to long-term fiscal sustainability. Revenue outcomes that deviate significantly from the original budget undermine not only fiscal discipline but ultimately the government’s ability to control its resources and predictably allocate them to policy priorities. Overly optimistic revenue forecasts lead to unjustifiably large spending allocations, which then entail potentially disruptive in-year spending reductions, unplanned borrowing, and perhaps the accumulation of arrears. The weaknesses in North Macedonia’s PFM system have led to the build-up of arrears, though that can be remedied. Unrealistic budgets that are constantly revised, inadequate fiscal reporting and control systems, discrepancies between revenue flows and local government (LG) responsibilities, and defects in SOE management have all contributed to the proliferation of arrears. The treasury’s IT system requires effective controls to register both initial obligations and the due dates of invoices. As the new integrated financial management information system (IFMIS) will take time to become operational, an arrears prevention and clearance strategy is a crucial first step in containing further accumulation of debt. The strategy will only succeed, however, if budgetary institutions universally respect the procedures put in place. Medium-term planning, as part of the medium-term fiscal strategy (MTFS), could be enhanced so as to better guide government policy. The MTFS stipulates the objectives of fiscal policy and outlines basic macroeconomic projections and indicators. However, beyond the budget year, aggregates of the current strategy are not viewed as binding targets or ceilings but as indicative figures reflecting current projections. Multiyear budgeted spending ceilings are therefore not binding, and there is no effective control over obligations accruing to subsequent budget cycles. Since there are no multiyear targets, multiyear commitments are not measured against objectives and can crowd out future fiscal decisions—or potentially lead to arrears. Since 2010 North Macedonia’s budgets have become markedly less transparent and fail to provide taxpayers with information on how their money is being spent. This is critical because it has a direct impact on social trust and citizen attitudes to tax compliance. In the 2017 Open Budget Survey, a report prepared by the International Budget Partnership, North Macedonia scored lower than the global average on the usefulness of the information publicly available throughout the budget cycle. Though its budget documentation is reasonably comprehensive, among remaining gaps are harmonization of statistical reporting with international standards. The country’s PFM strategy identifies fiscal transparency as a priority for reform and the new Government has moved in this direction. However, specific actions need to be defined to ensure that there are genuine and comprehensive improvements that meet international benchmarks. Overview 3 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Fiscal Decentralization and Local Governments. Too little revenue autonomy, delegated functions that are vaguely defined, unclear expenditure assignments, and uneven institutional capacity all make it difficult for municipalities to deliver public services. North Macedonia’s municipalities are unable to finance some of their basic functions and face several challenges in providing public services. Critical social services, education for example, are underfunded. Own revenues from taxes and fees represent about a third of total LG revenue, and the fiscal imbalance is met with transfers. Block grants account for about 80 percent of total transfers to LGs and for about 49 percent of their total revenue. The value of block grants as a percentage of GDP has declined, and their funding is inadequate. Moreover, there are numerous problems in the grant distribution formulas, notably the transparency of the calculations and the process for selecting which municipalities get what transfers. The current transfers system also does not address the huge economic disparities among municipalities. The imbalances stem from the disparate expenditure needs across local governments as well as their differing economic bases—and their differing fiscal capacity to generate own revenues. In total revenue per capita, the municipality with the most funding has six times the revenue of that with the least funding. Given such sizable horizontal imbalances, the transfer system has an essential role in the intergovernmental financing system but currently does little to resolve the disparities. Often, the funds transferred as block grants are insufficient, leading municipal governments to supplement them with their own resources or to accumulate arrears. Nor can the disparities between municipalities be addressed solely by transfers and enhanced revenue autonomy. Balanced actions across several dimensions are necessary, such as building capacity for absorption and use of funds and cooperation between municipalities for virtual consolidation. Because of these disparities, increasing local revenue autonomy in North Macedonia will help improve local spending efficiency. The country’s fiscal decentralization has been asymmetric, with relatively little devolution of taxing powers. Greater revenue autonomy, which can be achieved by expanding the number of local tax instruments and using current local taxes and fees more effectively, is generally associated with greater local accountability and more responsible use of fiscal resources by local officials. A careful assessment of the entire North Macedonia intergovernmental financing system, including spending assignments, will be needed to determine the proper level of transfers. Current LG expenditure figures hide the fact that there is no clear definition of spending assignments, which affects the behavior of LGs and central government ministries alike, as well as the efficient operation of the system. Simple and transparent methods are required to calculate the spending needs associated with functional responsibilities; more broadly, the logic and practice of the functions as they are currently delegated must be fully assessed. Adding to the problem is the fact that the efficiency of local spending is still affected by the suboptimal size of some municipalities. LGs were consolidated in parallel with early reform of the intergovernmental finance system, but some municipalities are still too small to be optimally efficient. Of the 71 municipalities, 31 have fewer than 10,000 people, which is the approximate minimum size necessary to exploit economies of scale in the delivery of most local public services. Furthermore, the 4 Overview SOWING THE SEEDS OF A SUSTAINABLE FUTURE system has not adequately internalized the vast heterogeneity of the country’s cities and towns, which also undermines efficiency. Tax Collection. Tax reforms are critical to North Macedonia’s much-needed fiscal consolidation. The reductions in tax rates, proliferation of tax breaks, and inefficiencies in tax collection have seriously impeded revenue mobilization. Government revenue collection as a share of GDP has declined over the past 10 years in North Macedonia and is now among the lowest in the region. Tax revenue and social security contributions (SSC)—the largest sources of revenue—have been falling due primarily to dwindling VAT collections and SSC. The decline in the country’s revenues is also partly driven by the cuts in the personal and corporate income tax rates. There is considerable room to increase tax revenues. For example, a higher marginal personal income tax (PIT) rate would not only bring in more revenue but would also create a more progressive system. The resulting distortionary effects on the economy are not likely to be substantial, given the country’s currently low PIT and low effective tax rate on capital income compared to peer countries. Indirect taxes like VAT and excise taxes are likely to become an increasingly important source of revenue as the population ages and the base for income taxes shrinks. The government could expand the VAT base to increase VAT revenue by removing preferential treatment for goods and services that are not in line with international standards or not well-targeted to help the poor. The excise tax on certain goods—especially tobacco and some energy commodities, which are below the minimums set by the European Union (EU)—should also go up. The government’s ambitious policy to attract foreign direct investment (FDI) has contributed to the decline of the country’s revenue-to-GDP ratio. Extensive tax breaks and exemptions that have reduced the CIT base are not sustainable in the long run. First, the EU accession process restricts the size and type of incentives that can be granted by candidate countries. Second, competition between Western Balkan countries to attract FDI has intensified, which has pushed up the fiscal costs of using this strategy to secure foreign investment. Revenues in North Macedonia are falling not only because of investment tax incentives and lower tax rates but also inefficient tax collection, particularly from VAT. Even after controlling for lower rates, the collection of certain taxes like the VAT is significantly below the level of peer countries, which implies that tax administration is not efficient. In fact, the recent decline in the VAT revenue-to-GDP ratio was not the result of a change in the standard VAT rate but can be attributed in part to worsening tax efficiency. Indeed, recent analysis has exposed the existence of substantial tax arrears as well as weaknesses in North Macedonia’s management of compliance risk, its formal administrative dispute- review processes, and the coverage of its tax audits. Government Spending and the Pension System. Together with improved revenue collection, more efficient spending is a pressing policy priority to extend North Macedonia’s limited fiscal resources. Although moderate by European standards, government spending suffers from some inefficiencies and in some areas, notably pensions, is unsustainably high. Overview 5 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 North Macedonia could “spend better.” It scores lower than peer countries on the Public Sector Performance (PSP) index, which measures seven performance areas. Its similarly low score on the Public Sector Efficiency (PSE) index suggests that its spending does not maximize public sector performance. Moreover, its PSE scores vary significantly across public service areas; in 2016, for example, it scored relatively well on administration, economic stability, and income distribution but relatively poorly on economic performance and infrastructure. Government spending in North Macedonia is marked by high and still-growing spending on social protection, particularly pensions. Despite relatively moderate government spending overall, social protection spending is among the highest in the WB6 region, averaging 14.5 percent of GDP for the past nine years.1 At 44 percent of total spending, the escalating social protection expenditures, mostly comprising pensions, severely restrict the country’s remaining fiscal space and prevent it from moving forward with much-needed investment increases in infrastructure and public services. The pension system must be sustainable through the medium and long term. Despite earlier parametric reforms and introduction of a multipillar system, the pay-as-you-go (PAYG) system faces significant financial imbalances due to the aging of the population, ad hoc indexations, and simultaneous cuts in the contribution rate. Pensions account for 10.3 percent of GDP, higher than in many peer countries with larger retiree populations. Without further reforms, the sizable pension deficit of 4.5 percent of GDP is projected to reach 6 percent by 2030 as pension spending rises to 12 percent of GDP. This is unaffordable. And the state of affairs will only become more burdensome over the next decades as the country’s demographic pressures rise. The system is also inequitable: second-pillar participants are awarded low PAYG accruals, far below their share of PAYG contributions. A series of incremental pension reforms could both ensure the sustainability and adequacy of pension incomes and address the inequities. Because individual policy measures are only part of the solution, however, they could only partly stabilize the pension system. A well-sequenced package of measures, such as revising indexation to link it primarily to the consumer price index, gradually raising the statutory retirement age, adopting policies to increase labor force participation, and raising contribution rates would help to make the pension system more sustainable. Social Assistance. Despite the costs of its whole social protection system, North Macedonia’s spending on social assistance is low and does little to reduce poverty. Between 2014 and 2017 the country’s social assistance spending of 1.2 percent of GDP was among the lowest in the region and could be far more equitable. Despite North Macedonia’s relatively high poverty rate and the limited income-generation opportunities available to the poor and vulnerable, resources have increasingly been directed away from means-tested programs, such as social financial assistance (SFA) and the Child Allowance program, toward non-means-tested plans. Although social assistance is well-targeted generally, coverage of the poorest quintile is inadequate, leaving many mired in poverty. There is scope for improving the cost-efficiency of North Macedonia’s current social assistance spending. Other countries with similar or even lower social assistance spending have been able to achieve higher coverage. 1 The WB6 are Albania, Bosnia and Herzegovina, Kosovo, North Macedonia, Montenegro, and Serbia. 6 Overview SOWING THE SEEDS OF A SUSTAINABLE FUTURE Among the social transfers analyzed in this report, the SFA program appears to be the most cost- efficient. However, it currently has virtually no impact on poverty in North Macedonia because of the low poverty threshold and the small share of the population that receives the benefit. The SFA benefit levels are very low. At the same time, the financial incentives for social assistance beneficiaries to seek gainful employment are minimal. A frequent justification for offering a low amount of assistance is to promote self-reliance and encourage people to seek work, but since there is a shortage of job opportunities in North Macedonia, SFA recipients simply fall into the poverty trap. A new guaranteed minimum income scheme that could integrate most of the current SFA and PFA beneficiaries should be based on reformed income and asset tests. Education. North Macedonia’s spending on education is both low and inefficient even though enrollment is comparatively low, as are the secondary completion rates. Increasing enrollment is central to improving the country’s education outcomes. Although there is universal access to primary and secondary education, gross enrollment rates have not substantially improved since 2000 and at all levels are below those of peer countries. Furthermore, the quality of education is not only low but has been declining over time. North Macedonia may boast of equal learning outcomes among students from different socioeconomic and regional backgrounds, but this may actually be due to the low quality of education across the board. Increasing the country’s completion rates for secondary education—and the skills taught there— would substantially improve employment outcomes. North Macedonia’s completion rates are high for primary school but are much lower for secondary education, resulting in an under-educated working population. Moreover, youth unemployment is very high, which points to a mismatch between the skills taught in the education system and private sector needs. Though labor demand is relatively subdued in North Macedonia, firms that do wish to expand cite skills as the main obstacle—they cannot fill vacant positions because they cannot find workers with the right skill profiles. North Macedonia’s steady reduction in its spending on education has been exacerbated by the inefficiency of that spending. The country currently spends less than peer countries on both pre- university and tertiary education, yet a decline in the numbers of students has been accompanied by an increase in the number of teachers. Though the nine-year length of schooling resembles that in peer countries, accumulated instruction time is much lower—about 1,000 hours less than the EU average, in fact, which is equivalent to almost one full year of schooling. As in many other countries, LGs administer most of what the country spends on education, leading to substantial disparities that the transfer system must address. Many municipalities spend far less than the national per student average, and those that have more students benefit from economies of scale and are the most efficient. Capital spending for schools is managed at the central level, however. Though one-third of school buildings require major renovation, only 5 percent of total education spending is spent on capital developments, compared to 8 percent for peer countries. Because of the scarcity of capital resources, existing facilities are used very intensively, resulting in double and even triple shifts in some schools. Overview 7 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Health. North Macedonia’s aging population, along with the associated lifestyle changes, has led to a shift in the burden of disease that the health system is not addressing effectively. At 4.1 percent of GDP, North Macedonia currently allocates fewer resources to health care than peer countries. Over the past 15 years, its cost-control policies have led to a consistent decrease in public health spending as a share of GDP. However, some of the resulting financial burden has been absorbed by health care recipients, resulting in high out-of-pocket spending. Population aging, and the associated lifestyle changes, have serious implications for North Macedonia’s health system. The health burden is moving toward noncommunicable diseases (NCDs), where the risk factors are high and have worsened over time. Over half of the adult population is estimated to be overweight, a figure that has been rising recently. The Government of North Macedonia therefore needs to optimize health care resources and shift the emphasis to primary care. The share of public health spending on primary care has been relatively constant, however, which suggests that health services have not yet adapted to the rising challenges and that prevention services in particular is still underdeveloped. Though limited, the data available on the quality of care in North Macedonia suggest that while waiting times are satisfactory, whether the care delivered is appropriate may be questionable. The low birth numbers and disproportionately high rates of cesarean sections, for example, demonstrate this point, with consequences for patient safety as well as the cost of health care. In addition, the fact that people are routinely admitted to the hospital for acute respiratory conditions, diabetes, and hypertension suggests numerous unnecessary hospital admissions and a lack of uniformity in admission procedures. Agriculture. Despite its productive potential and the generous government support, agriculture is unproductive and structurally weak. Spending on agriculture has been scaled up ineffectively in pursuit of social policy objectives, but it is inefficient and requires serious review focused on reorientation and distinction from social assistance measures. The Government of the Republic of North Macedonia intervenes heavily in agriculture sector but could achieve better outcomes by supporting higher-value farming. Between 2010 and 2015, its budgetary transfers to farmers were double those of other WB6 countries. Market support and direct payments (MSDP) represent the most important spending category by far. Direct support largely favors low-value production; In 2015, for instance, the output value of tobacco was only 5.1 percent but between 2013 and 2016 it received 27.7 percent of MSDP support. What is more, MSDP has had little if any impact on poverty reduction and does not appear to be effective way to pursue that social policy objective. Despite its substantial public support and productive potential, structural weaknesses hinder the development of agriculture. Like government support, agricultural land use in North Macedonia is not driven by output value. Furthermore, farms are structured in a way that discourages technological investment and productivity. Agricultural holdings are mostly small and fragmented; farms are therefore labor intensive, technologically outdated, and saddled with outmoded production management techniques. And farmers in North Macedonia lack financial liquidity and access to capital for investments. 8 Overview SOWING THE SEEDS OF A SUSTAINABLE FUTURE Shifting from direct subsidies to support for farm competitiveness (and broader rural development measures) would be a first step toward enhancing agricultural productivity. The current sectoral allocation of farm subsidies in North Macedonia has been highly distorting and ultimately favored unproductive and inefficient farms. Government subsidies therefore actually have a negative impact on technical efficiency and also fail to address agriculture’s structural problems. North Macedonia therefore needs to move more conditional and carefully targeted support. Reform Options. This report identifies reform options that could create fiscal space and tangibly improve public service delivery and labor market outcomes. The PFR’s proposed reform options include fiscal consolidation measures for the short and medium term, depending on their priority and technical and political feasibility ( Annex 2 Table A2.1). These measures could yield additional annual savings of up to 3.5 percent of GDP for 2019–21, primarily from pension and tax revenue reforms. Replacing the current Swiss-indexation of pensions benefits with indexation to the consumer price index (CPI), a gradual increase and equalization of the retirement age for men and women, putting in place a temporary pension indexation freeze; and a gradual increase of the PAYG contribution rate could yield up to 3.2 percent of GDP over the medium term. In parallel, the following short-term tax reforms, all of which are feasible, could yield savings of up to 2.5 percent of GDP: removing tax exemptions on the sales of securities, introducing a higher tax bracket for the PIT, broadening the capital income tax base, increasing the permitted standard deduction in the PIT, and expanding the VAT base by removing preferential treatment for items not in line with international standards. While these reforms promise potentially significant efficiency gains and budget savings, other reforms in the areas of agricultural subsidy and social assistance policy, health and education reform, and fiscal decentralization and PFM are equally important for making public spending more effective. Overview 9 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Chapter 1: The Need for Fiscal Reforms Strong fiscal consolidation through more efficient taxation and spending is needed to stabilize debt and create space for fiscal policy, the country’s main countercyclical stabilization policy tool. North Macedonia’s government revenues have declined, partly because of lower tax rates in major categories but also because of generous tax exemptions and inefficient tax administration. Government spending is comparable to that of peer countries, but there is scope for raising efficiency.2 While from a global perspective public debt is moderate, the pace at which it has been building for the past decade could raise questions about sustainability. Gross financing needs are high, and by 2021 adverse shocks, particularly in GDP growth, could push public and publicly guaranteed debt above 60 percent of GDP. Revenue, expenditure, and institutional reforms are needed to safeguard the sustainability of public finances, while increasing the quality of public services. 1.1  Country Context and Fiscal Dynamics North Macedonia must now revitalize growth without jeopardizing fiscal sustainability. Its good record of macroeconomic stability and its structural reforms to improve the investment climate and reinforce public sector governance were helping to propel growth up to the 2008–09 global financial crisis. Conservative fiscal policies provided a buffer against spillovers from the crisis and enabled a solid recovery. However, growth has since moderated. Between 2010 and 2017 North Macedonia’s real GDP growth averaged about 2.4 percent, down from 4 percent for 2003 through 2009. North Macedonia’s current lackluster growth hamstrings its potential to converge economically with European Union (EU) economies. Its GDP per capita is just over one-third of the EU-28 average and has not changed much since the global financial crisis. Between 2000 and 2008, North Macedonia grew by 3.7 percent annually in real per capita PPP terms, which enabled it to increase its income as a share of the EU28 per capita average from 30.7 percent to 36.6 percent. However, between 2009 and 2017, convergence slowed significantly, so that its PPP income as a share of the EU28 average only grew from 34 to 37 percent. Within the next two to three decades North Macedonia’s real GDP per capita growth would need to accelerate to about 4.5 percent a year to converge to EU living standards.3 2 Because North Macedonia is following the path previously taken when New Member States were striving to become EU members, the analysis here relies heavily on comparisons within this subregion. It focuses on the STEE7 (Seven Small Transition Economies of Europe) group—Bulgaria (BGR), Croatia (HRV), Estonia (EST), Latvia (LVA), Lithuania (LTU), the Slovak Republic (SVK), and Slovenia (SVN)—which are similar in size, transition and legacy issues, and the reforms required for continuing convergence and making progress toward EU accession. 3 In the model presented in Figure 1.2, the EU is projected to grow annually by 1 percent, while North Macedonia is projected to grow at 3 to 4.5 percent annually. 10 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 1.1. GDP per Capita Growth Rates, 2001–08 and Figure 1.2. Income Convergence with the European 2008–16 Averages, Union in Different Growth Scenarios Percent GDP per capita, thousands US$ (2011 PPP) 7 70 6 60 5 50 4 40 3 30 2 20 1 10 0 0 0 6 12 2018 2024 2030 2036 2042 048 2054 060 066 2072 2001–2008 2009–2016 200 200 20 2 2 2 JJMKD JJWB6 JJSTEE7 JJUMI ÂÂEU (1%) ÂÂMKD (3%) ---- MKD (4.5%) Source: Data from the Statistical Office. After 2009 growth in North Macedonia benefitted from fiscal stimulus that included substantial infrastructure spending, measures to spur investment, and support for consumption through pension hikes and fiscally costly employment schemes. In response to the global crisis, the government first lowered the tax burden to support firms and to increase households’ disposable incomes through a series of interventions on the revenue side, it (1) reduced social contribution rates; (2) temporarily abolished the profit tax on re-invested earnings for 2009 through 2014; (3) broadened the list of goods with preferential tax rates; and (4) gave tax exemptions for foreign direct investment in technological industrial development zones. Second, the government increased spending to support aggregate demand by (1) permanent and ad hoc increases in pensions (above the increases mandated by law); (2) increased social assistance; and (3) heightened public investment considerably, though a significant share went to non-growth-enhancing structures (mainly administrative buildings) as part the urban- renewal project “Skopje 2014” and, in the past two years, has invested more in highways. However, as these initiatives wound down, the pace of economic growth slowed. The combination of slower growth and expansionary fiscal policy has had fiscal repercussions. The crisis-response measures prompted a sustained decline in revenues and spending rose, which has led to ten years of persistent actual deficits (Figure 1.3). From a surplus position between 2003 and 2007, since 2008 the overall fiscal balance has had an average deficit of 3.0 percent of GDP and the primary balance has had an average deficit of 2.0 percent.4 The cyclically-adjusted overall deficit has also widened considerably, illuminating the structural nature of the fiscal deterioration—since 2008 the structural deficit has averaged 3.1 percent of GDP (Figure 1.3). In recent years, fiscal balances have improved, but mainly because of under-execution of capital spending linked to the prolonged electoral cycle. For the past decade weakened fiscal discipline has been characterized by the accumulation of payment arrears and frequent issue of supplementary budgets that increased the expected deficit. 4 Unless stated otherwise, the coverage of general government used in this report follows the definition applied by the authorities of North Macedonia. It includes spending by the central government and municipalities and excludes road capital spending (by the Public Enterprise for State Road–PESR), which was transferred off-budget in 2013 (see Box 1.1). Including PESR finances in general government data, as is done in many neighboring countries, would bring the overall deficit for 2008–17 to 3.3 percent and the primary deficit to 2.4 percent. Chapter 1: The Need for Fiscal Reforms 11 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 1.3. Fiscal Policy Trends of the General Figure 1.4. Central Government Budget Execution, Government, 2001–17 2006–17 Percent of GDP Percent of GDP Percent of GDP 42 3 6 40 2 4 38 1 0 2 36 -1 34 -2 0 32 -3 30 -2 -4 28 -5 -4 26 -6 24 -7 -6 1 3 5 7 9 1 3 5 7 1 3 5 7 9 1 3 5 7 200 200 200 200 200 201 201 201 201 200 200 200 200 200 201 201 201 201 JJFiscal balance, rhs JJPrimary balance ÂÂRevenues, lhs ÂÂExpenditures, lhs JJFiscal position ÂÂOutput gap ---- Structural fiscal balance Source: Ministry of Finance and World Bank staff calculations. Note: Expenditure by the Road Agency is included in 2001–12 data and excluded from 2013–17 data. Box 1.1. The Public Enterprise for State Roads (PESR) The Public Enterprise for State Roads (PESR) is responsible for planning, construction, reconstruction, rehabilitation, maintenance, and protection of public roads and managing state roads. Previously part of the budgetary central government as the Road Fund, this entity can now borrow on its own behalf, although most of its debt carries an explicit sovereign guarantee. Since its creation in 2013, its spending has average 1.7 percent of GDP and revenue 1 percent. Shares of the oil tax are its main source of revenues, followed by income from public toll roads and vehicles registrations. While the PESR implements a central government function, it is not counted as part of the central government agenda. Similarly, toll fees collected by the PESR (making up roughly 0.5 percent of GDP) are no longer counted towards central government non- tax revenues. Ambitious borrowing plans of the PESR and an increased share of financing from central government tax revenues require careful monitoring. To further strengthen transparency and align with international statistical reporting systems, it would be advisable to consolidate the PESR within the central government for deficit and debt accounting purposes since it is a non-market non-profit institution which performs government functions. See Government Finance Statistics Manual (GFSM) 2014 and European System of National and Regional Accounts (ESA) 2010. Table 1.1. General Government Fiscal Accounts* Percent of GDP Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 Total Revenues 32.8 32.4 31.7 32.1 30.1 29.7 31.0 30.3 30.9 Taxes and contributions 27.6 27.1 27.0 26.6 25.4 25.8 26.6 26.5 26.7 Tax revenues (SRA) 0.1 0.2 0.2 0.2 0.2 0.2 0.3 0.3 0.2 Personal income tax 2.1 2.1 2.1 2.1 2.1 2.4 2.4 2.4 2.5 Profit tax 1.1 0.8 0.8 0.8 0.9 1.0 2.2 1.8 1.8 12 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 1.1. General Government Fiscal Accounts (continued) Year 2009 2010 2011 2012 2013 2014 2015 2016 2017 Property tax 0.1 0.2 0.2 0.1 0.2 0.2 0.2 0.2 0.2 VAT 8.5 8.6 9.1 8.2 7.9 8.3 7.5 7.7 7.7 Excises 3.5 3.4 3.3 3.6 3.2 3.3 3.5 3.7 3.7 Import duties 1.3 1.1 0.8 0.9 0.8 0.8 0.8 0.8 0.8 Other taxes 1.6 1.9 2.0 2.0 1.6 1.2 1.2 1.2 1.2 Contributions 9.4 8.8 8.6 8.7 8.5 8.4 8.6 8.4 8.5 Non-tax revenues 4.2 4.1 3.1 3.0 2.6 2.2 2.5 2.3 2.3 Capital revenues 0.4 0.6 1.1 1.6 1.1 0.7 0.8 0.7 0.4 Foreign donations 0.4 0.4 0.3 0.7 0.9 0.9 1.0 0.9 1.4 Revenues from repayment of loans 0.1 0.1 0.2 0.1 0.1 0.1 0.1 0.1 0.0 Total Expenditure 35.5 34.8 34.2 36.0 34.1 33.9 34.4 33.0 33.7 Current expenditures 31.4 30.3 29.4 30.8 29.7 29.7 30.2 29.2 29.5 Wages and allowances 8.0 7.9 7.7 7.7 7.2 7.0 7.0 6.8 6.6 Goods and services 5.3 4.8 4.3 4.7 4.4 4.3 4.6 4.0 3.6 Transfers 17.5 17.0 16.6 17.4 17.2 17.4 17.5 17.3 18.0 Pensions 8.8 8.6 8.5 8.8 9.0 9.1 9.0 9.1 9.4 Unemployment benefits 0.5 0.5 0.5 0.5 0.4 0.3 0.3 0.2 0.2 Social benefits 1.0 1.1 1.1 1.2 1.2 1.2 1.3 1.3 1.3 Health 4.5 4.4 4.4 4.5 4.3 4.2 4.2 4.3 4.4 Subsidies 2.1 2.0 1.8 2.0 2.0 2.2 2.1 2.0 2.4 Other transfers (SRA) 0.5 0.5 0.4 0.5 0.5 0.4 0.5 0.5 0.3 Interest 0.6 0.7 0.7 0.9 0.9 1.0 1.2 1.2 1.4 Capital expenditure 4.1 4.5 4.8 5.2 4.3 4.2 4.2 3.8 4.1 Overall Balance -2.7 -2.4 -2.6 -3.9 -4.0 -4.2 -3.4 -2.7 -2.8 Memo Items PESR expenditures 0.8 2.0 1.7 2.2 1.7 PESR revenue 0.9 0.9 1.0 1.1 1.0 Capital expenditure including PESR 4.1 4.5 4.8 5.2 4.9 5.9 5.6 5.6 5.5 Total expenditure including PESR 35.5 34.8 34.2 36.0 34.9 35.9 36.1 35.2 35.4 Overall balance including PESR -2.7 -2.4 -2.6 -3.9 -4.0 -5.3 -4.1 -3.8 -3.4 Source: Ministry of Finance and World Bank staff calculations. Note: *Consolidated Budget of the Republic of North Macedonia and budgets of municipalities. North Macedonia’s public debt is comparatively moderate and has even declined slightly over the past year. However, it has doubled in the past 10 years, because of repeated primary deficits and public investment by state-owned enterprises. In 2007, North Macedonia pre-paid part of its debt to the Paris and London clubs, which reduced debt as a share of GDP from 33.2 percent in 2006 to 23 percent in 2008—the lowest level during that decade. Since then, its debt dynamics have worsened, due to growing primary deficits and more borrowing by the government and SOEs, so that between 2008 and 2016 debt as a share of GDP shot up from 23 to 48.5 percent. But by the end of 2017, public debt had gone down slightly, to 47.6 percent, as the government withdrew its cash reserves—the first decline since 2008. This year debt has risen slightly as the country issued a €500 million Eurobond at historically favorable terms. As of June 2018, North Macedonia’s public and publicly guaranteed (PPG) Chapter 1: The Need for Fiscal Reforms 13 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 debt-to-GDP ratio was 47.7 percent, less than the 52 percent average for the Western Balkans (the WB6), but higher than the STEE7 average of 45 percent). While this remains moderate from a global standpoint, the margin for further debt accumulation has narrowed quickly. The speed at which North Macedonia has accumulated debt since 2008 calls for decisive action to stabilize public debt. At about 40 percent of GDP, central government (CG) debt is the largest component of public debt, but the debt of SOEs has been growing fast  (Figure 1.5). North Macedonia’s total public debt comprises the debt of the CG, municipalities, and guaranteed debt of SOEs. As 2017 ended, CG debt accounted for about 82.4 percent of the total PPG debt, but its share has been steadily declining as SOEs have taken on more pf the debt-financed public investment. Part of the government’s post-2009 fiscal stimulus was implemented through SOEs, which borrowed heavily with government guarantees. At the end of 2017 guaranteed debt of public enterprises and state-owned joint stock companies amounted to 8.2 percent of GDP (of which the PESR accounts for 58 percent—4.8 percent of GDP). At that point the CG had guaranteed the debt of all SOEs except for loans to MEPSO, the high-voltage electricity transmission company, and the Civil Aviation Agency, which jointly account for 0.2 percent of GDP. Figure 1.5. North Macedonia’s PPG Debt, 2006–17 Figure 1.6. PPG Debt-to-GDP Ratio, North Macedonia and Comparator Countries, 2008 and 2017 Percent of GDP Percent of GDP 50 90 80 40 70 60 30 50 40 20 30 10 20 10 0 0 6 7 8 9 0 1 2 3 4 5 6 e V N E B B K D H U A R S T E 6 8 200 200 200 200 201 201 201 201 201 201 201 2017 HR SV MN AL SR SV MK BI LT LV BG KO ES 7STE WB EU2 JJCentral government JJFunds JJMunicipalities JJGuaranteed debt of SOEs JJ2017 QQ2008 Source: WB6 authorities and World Bank staff calculations. Figure 1.7. North Macedonia’s PPG Debt, Contribution Figure 1.8. General Government Gross Financing to Change as Percent of GDP, 2008–17 Needs, North Macedonia and Comparators, 2017 Percent of GDP Percent of GDP 7 30 6 5 25 4 3 20 2 15 1 0 10 -1 -2 5 -3 -4 0 8 9 0 1 2 3 4 5 6 7 ALB MNE MKD HRV SRB S LVA BIH BGR 200 200 201 201 201 201 201 201 201 201 SVK SVN KO LTU EST JJPrimary deficit JJReal interest rate JJReal GDP growth ---- WB6 average ÂÂSTEE7 average JJExchange rate depreciation JJOther debt-creating flows JJResidual including asset changes ÂÂChange in gross public sector debt Source: Ministry of Finance and World Bank staff calculations. Source: IMF and World Bank staff calculations. 14 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 1.2. North Macedonia’s Central Government Debt: Cost and Risk Indicators as of 2017 Foreign Domestic Risk Indicators Total Debt Currency Debt Currency Debt Amount, MKD million 190,668 51,865 242,535 Amount, € million 3,100.8 843.5 3,944.3 Nominal debt, percent of GDP 30.8 8.4 39.2 Cost of debt Interest payments, % of GDP 1.3 0.2 1.3 Weighted av. IR (%) 4.1 2.9 3.9 Refinancing risk Average time to maturity, years 5.4 1.5 4.6 Debt maturing in 1 year, % of total 9.2 79.2 24.2 Debt maturing in 1 year, % of GDP 2.8 6.7 9.5 Interest rate risk Average time to refixing, years 4.6 1.5 4.0 Debt refixing in 1 year, % of total 22.7 79.2 34.9 Foreign currency risk Foreign currency debt, % of total 78.6 Source: Ministry of Finance and World Bank staff calculations. The structure of CG debt has improved noticeably in recent years, but the country remains exposed to refinancing and interest rate risk. North Macedonia has been successful in lengthening the maturity of its debt, particularly for domestic securities. In a shift away from short-term debt, 3-month T-bills, which accounted for 65 percent of domestic debt at the end of 2012, were phased out during 2013 as the share of 12-month T-bills rose from 6.5 to 38 percent. In 2014 the Ministry of Finance (MOF) introduced 10-year T-bonds and in 2015 15-year T-bonds, and in April 2018 it launched a 30-year bond. As a result, in March 2018 the average time to maturity (ATM) of North Macedonia’s domestic securities portfolio was 5.1 years, up from 1.1 year in 2012; the ATM of total debt rose from 3.8 years in 2012 to 4.6 years as of 2017 (Figure 1.10). Figure 1.9. Maturity Profile, North Macedonia’s Figure 1.10. Average Time to Maturity, General General Government Debt, 2017–35 Government Debt, 2012–17), Months MKD billion Months 60 70 50 60 50 40 40 30 30 20 20 10 10 0 0 7 9 1 3 5 7 9 1 3 5 201 201 202 202 202 202 202 203 203 203 2012 2013 2014 2015 2016 2017 JJExternal JJDomestic ÂÂDomestic debt ÂÂExternal debt ---- Total Source: Ministry of Finance and World Bank staff estimates. The aggregate numbers in the domestic securities portfolio hide two opposite realities: While the government was able to issue medium and long-term securities in the domestic market linked to the euro (foreign exchange clause), the securities issued in domestic currency are still mainly short- term. As of 2017, 55 percent of the domestic securities were in domestic currency and 68 percent of which were T-bills maturing in less than one year. The very concentrated investor base—dominated Chapter 1: The Need for Fiscal Reforms 15 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 by four big banks and two pension funds—adds to the refinancing risk, although the current excess liquidity in the market mitigates the risk. The maturity profile of CG debt illustrates the short-term concentration due to the T-bills and the peaks during 2020–2023, which are the maturity dates of the three outstanding Eurobonds5 (Figure 1.9). Reflecting high roll-over needs, the gross financing requirements of the public sector are a sizable 14 percent of GDP (Figure 1.8). This is very close to the 15 percent of GDP threshold for middle-income economies.6 Because so much of its domestic currency debt is short-term, North Macedonia’s exposure to interest rate risk is also not negligible. The government has been successful in significantly reducing the interest rate risk, as evidenced by the reduction in the share of variable rate debt from 48.7 percent in 2012 to about 20 percent in 2017. However, about 27 percent of the total public debt still re-fixes its interest rates in one year, mainly due to the amount of MKD T-bills, which are fixed-rate securities but the rates re-fix in the short term, but also due to the variable-rate loans from multilaterals.7 Average time to refixing (ATR) is 4.0 years for the whole portfolio, reflecting a mix of much higher risk in the domestic currency debt (ATR of 1.5 years) and lower risk on the external currency debt (ATR of 4.6 years). Figure 1.11. North Macedonia’s General Government Figure 1.12. North Macedonia’s General Government Debt Composition, 2012–17 Debt Composition by Currency, 2017 Percent of GDP Percent 50 45 40 2 35 11 3 30 25 72 20 15 22 10 5 0 2012 2013 2014 2015 2016 2017 JJExternal debt JJDomestic MKD debt JJDomestic FX debt JJMKD JJEUR JJJPY JJRMB JJGBP JJUSD Source: Ministry of Finance and World Bank staff calculations. North Macedonia’s debt portfolio is exposed to exchange rate risk, which has grown significantly with the expansion of foreign currency debt, but that risk is mitigated by the peg. Since 2008, external public debt has increased by 15.5 percent of GDP to reach 31.7 percent of GDP in 2017; it constitutes 66.6 percent of total general government debt. It is mainly composed of multilateral and bilateral loans, four outstanding Eurobonds, and a few commercial loans. While North Macedonia’s main creditors are multilateral development banks (the International Bank for Reconstruction and Development, the International Development Agency, and the European Investment Bank) the four Eurobonds and the commercial loans together account for 65 percent of the external debt portfolio. Given that 45 percent of North Macedonia’s domestically-issued debt is linked to the euro, FX-denominated debt accounts for 82.2 percent of its public debt. Having so much FX-denominated public debt should make North 5 In January 2018 the MOF conducted a liability management operation in the external markets, buying back about one-third of the Eurobonds maturing in 2020, improving the debt profile and reducing the refinancing risk. 6 See MAC DSA. 7 If state-guaranteed debt is included, the share of variable interest rate debt goes up to 44 percent. 16 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE Macedonia’s portfolio vulnerable to exchange rate risks, but since most of its foreign currency risk is tied to the euro—only 6.5 percent has no euro exposure—it is mitigated by the exchange-rate peg. Figure 1.13. North Macedonia PPG Debt, 2015–21 Figure 1.14. North Macedonia Gross Financing Needs, 2015–21 Percent of GDP Percent of GDP 60 16 15.3 56.3 56.6 15 55 54.7 54.8 14.2 53.8 14.1 14.0 54.2 14 13.9 13.7 13.7 50 50.6 13 13.1 48.4 12.9 47.5 46.6 12.4 45 12 2015 2016 2017 2018 2019 2020 2021 2015 2016 2017 2018 2019 2020 2021 ÂÂNo consolidation ---- Government plan ÂÂNo consolidation ---- Government plan Source: World Bank staff calculations. Widening primary deficits and continued Figure 1.15. Debt Sustainability Scenarios: North Macedonia’s PPG Debt, 2015–21 borrowing by SOEs to finance infrastructure Percent of GDP investments are expected to push up public debt 70 further in the medium term.8 However, policy choices will determine how much will be fiscally 65 sustainable. In the baseline scenario, which assumes gradual fiscal consolidation in the 60 medium term, as envisaged in the Government’s 55 Fiscal Strategy for 2019–21, public debt would grow by about 7.3 percentage points (pp)of GDP 50 between 2017 and 2020 before declining. This scenario assumes sustained political stability 45 2015 2016 2017 2018 2019 2020 2021 due to timely resolution of the name dispute and ÂÂInterest rate shock ÂÂGrowth shock ÂÂCombined shock ---- Government plans expedited accession to EU and NATO, which would Source: National authorities and World Bank staff estimates. Note: Baseline scenario reflects the government current fiscal and macro scenario. The DSA bolster confidence and accelerate growth toward scenario assumes a one-time real devaluation of 15 percent in 2018;. The growth shock uses the public sector deficit trajectory for the forecast years. A 15-percent depreciation scenario assumes an average growth rate of 2.1 percent from 2018 to 2020, compared with 5 percent by 2021. This would also allow for timely 4 percent in the government macro framework. The interest rate scenario assumes the interest rate goes up by 200 basis points. The combined shock scenario assumes that consolidation measures to bring the primary no fiscal consolidation takes place, lower growth, a one-off depreciation, and a higher interest rate. deficit of the public sector down to 1.4 percent of GDP in 2020 (0.7 percent of GDP for the general government excluding roads coverage). In this scenario, baseline public gross financing needs would remain high and peak in 2020–21 largely because of the general government’s amortization of large Eurobonds. In this high-growth baseline scenario, public debt would peak at 54.8 percent in 2020 and decline thereafter. However, continuing on the historical trajectory of the fiscal deficit would cause public debt to accelerate along with gross financing needs. 8 The DSA framework for Market-Access Countries developed by the IMF is applied. See IMF “Guidance Note on the Application of the Joint Bank-Fund Debt Sustainability Framework for Low-Income Countries”. The DSA uses the public-sector deficit (including spending by the PESR). Chapter 1: The Need for Fiscal Reforms 17 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 The high-growth baseline scenario incorporating gradual consolidation is associated with downside  mong them (1) further tightening of U.S. monetary policy and euro depreciation; (2) potential risks, a deterioration in the external environment, such as a weaker EU outlook and a heightening of regional geopolitical tensions; and (3) potential delays in completing consolidation measures, or accumulation of new arrears and activation of contingent liabilities. The debt sustainability analysis is carried out on the extended trajectory for the general government sector (general government including roads. A 15-percent currency depreciation would push PPG debt to 59 percent of GDP by 2020, while with interest rates higher by 200 bps, public debt would be close to 56 percent of GDP by 2020. With lower GDP growth (on average 2 pp lower than in the government medium-term macro and fiscal scenario9), PPG debt would rise to almost 63 percent by 2021, while combined risks, although with negligible probability of materializing, would push public debt above 66 percent of GDP by 2021. Without fiscal consolidation, both public debt and gross financing needs would continue to climb in the medium term. This would not only increase financing costs, diverting resources from growth-promoting investments, but undermine the country’s ability to respond to future economic or demographic shocks. Box 1.2. North Macedonia Fiscal Strategy, 2019–21 On May 26, 2018, the Government adopted its 2019–21 Fiscal Strategy, which targets a gradual fiscal consolidation underpinned by projections for strong economic growth. The average real growth rate for 2018–21 is projected at 4 percent, reaching 5 percent by 2021. It assumes continued strong private consumption, a boost in investments (average growth of 5.5 percent in real terms 2019–21), a positive and increasing contribution of net exports, and a small positive contribution of public consumption. The rise in private consumption is partially explained by the projected continued rise in real wages bt over 2 percent during the projection period and a continued rise in employment. Despite the rise in domestic demand, imports growth is expected to be moderate, leading to a lower current account deficit. Table B1.2.1. Government Macroeconomic Projections, 2018–21, Percent of GDP 2017 2018f 2019f 2020f 2021f GDP growth rate (%) 0.0 3.2 3.5 4.0 5.0 Inflation (%) 1.4 1.7 2.0 2.0 2.2 Exports of goods (nominal growth, %) 15.7 10.1 9.0 9.6 10.9 Imports of goods (nominal growth, %) 11.8 7.4 8.1 8.5 9.1 Trade deficit (% of GDP) -18.1 -16.9 -16.8 -16.6 -15.9 CAD (% of GDP) -1.3 -1.9 -1.7 -1.6 -1.4 Net-wage (real growth, %) 1.2 2.3 2.4 2.4 1.5 Unemployment rate (%) 22.4 21.5 20.5 19.6 18.5 Employment rate (%) 44.1 44.9 45.8 46.8 47.9 (continued on next page) 9 North Macedonia Government Fiscal Strategy 2018–21. 18 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE (B0x 1.2. continued) The medium-term fiscal strategy (MTFS) is prepared for the central and local government; it does not include the PESR. It foresees a decline of the primary deficit from 1.4 percent in 2018 to 0.2 percent in 2021, leading to a moderation of public debt by 2021. While the MTFS projects that public and publicly guaranteed (PPG) debt will reach 54.8 percent of GDP in 2020 and decline thereafter, the MTFS sets a limit of 60 percent of GDP for the PPG debt and a maximum of 13 percent of GDP for the guaranteed debt. Table B1.2.2. Government Fiscal Projections, 2018–21, Percent of GDP 2017 2018f 2019f 2020f 2021f Central Government Revenues 26.2 26.6 26.4 26.0 Expenditures 29.0 29.1 28.7 28.0 Overall deficit -2.7 -2.5 -2.3 -2.0 Local Government Revenues 4.9 4.8 4.6 4.4 Expenditures 4.9 4.8 4.6 4.4 Overall deficit 0.0 0.0 0.0 0.0 Consolidated General Government Revenues 31.2 31.4 31.0 30.4 Expenditures 33.9 33.9 33.3 32.4 Overall deficit -2.7 -2.7 -2.5 -2.3 -2.0 Public debt 39.3 41.1 43.6 43.8 42.8 Guaranteed debt 8.3 9.5 10.6 11.0 11.0 Public and publicly guaranteed debt 47.6 50.6 54.2 54.8 53.8 Note: The CG numbers differ from the ones presented on the CG because of the netting out process due to transfers from CG to LG. Although the MTFS is announced as the consolidation plan, consolidation is largely expected to occur through growth. The following reforms contributing to the consolidation process have been announced: •• On the revenue side, the new progressive personal income taxation (savings of €30 mill); •• On the expenditure side, (1) strict control of employment and spending on wages, although with a disclaimer that if fiscal conditions are favorable, wage increases in some sectors is possible; and (2) timely payment of social transfers, such pensions, and reform of social benefits to improve efficiency. A number of measures to boost public spending were also outlined: (  1) considerable amounts of agricultural subsidies; and; (2) a rise in capital spending for rehabilitation of regional and local roads, some new regional roads, railway corridor VIII to Bulgaria, gasification, and local government investments). Chapter 1: The Need for Fiscal Reforms 19 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 1.2  Government Revenues North Macedonia’s public revenues have been trending downward trend for the past decade and in terms of GDP are now among the lowest in Europe. After peaking at 36 percent of GDP in 2003, the country’s total general government revenues steadily declined to an all-time low of 29.7 percent in 2014. By 2017, they had risen to 30.9 percent of GDP, making them the third lowest among the WB6 countries behind Kosovo (26.6 percent) and Albania (27.6 percent). Including the revenues of the PESR would bring total revenues in North Macedonia to about 31.9 percent of GDP, which does not compare well to the 40 percent observed in some WB6 countries and which is the average for EU member states. Figure 1.16. General Government Revenues, 2003 and Figure 1.17. Composition of Public Revenues in North 2017 Macedonia, 2009–17 Percent of GDP Percent of GDP 50 35 45 30 40 35 25 30 20 25 20 15 15 10 10 5 5 0 0 V H N B E T K V R U D B S 6 7 8 9 0 201 1 2 3 4 5 6 201 7 HR BI SV SR MN ES SV LT BG LT MK AL KO WB STEE EU2 200 201 201 201 201 201 201 JJ2017 QQ2003 JJPIT JJCIT JJProperty JJVAT JJExcise JJImport duties JJContributions JJOther revenues Source: Eurostat, Ministry of Finance, World Bank staff calculations. Source: Ministry of Finance. An ambitious FDI policy has contributed to the decline of North Macedonia’s revenue-to-GDP ratio. The government has pursued an effective policy of using tax breaks to attract FDI. Investors in technological and industrial development zones are exempted from the corporate income tax (CIT) for 10 years and may qualify for a personal income tax (PIT) reduction of up to 100 percent PIT. Investors are also exempt from paying the VAT and customs duties for goods, raw materials, equipment, and machines, and the government also significantly lowered social contributions to reduce the labor tax wedge. Social contributions averaged roughly 10.5 percent of GDP between 2001 and 2006—in line with regional averages at the time—with contribution rates totaling 32 percent of gross salaries. However, with North Macedonia’s 2006 labor tax wedge of 40.2 percent of labor costs and social contributions accounting for more than two-thirds of this wedge, the government began lowering contribution rates in 2009 in an effort to improve cost competitiveness and boost employment, formal sector activity, and FDI. Falling revenues are the result of inefficiencies in revenue collection as well as tax incentives and lower tax rates. Tax efficiency for both personal and corporate income taxes in North Macedonia has improved in recent years and is now in line with regional peers, even controlling for lower rates the collection of certain taxes, such as the VAT, is far below that of peer countries, which suggests tax administration problems. In fact, the recent decline in the VAT revenue-to-GDP ratio was not the result of a change in the standard VAT rate. Recent analysis has revealed weaknesses in North Macedonia’s compliance-risk management framework, its formal administrative dispute-review processes, and 20 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE the coverage of its tax audits, in addition to substantial tax arrears. All point to inefficiency in tax administration (IMF 2016). These issues are addressed in depth in Chapter 4. 1.3  The Level and Composition of Public Spending The Government of the Republic of North Macedonia is of relatively moderate size and has remained stable over the past decade. Its general government expenditure-to-GDP ratio has averaged 34.4 percent since 2009, which tracks the average worldwide for upper-middle-income countries (UMICs) and is below the WB6 average of 38 percent of GDP and the STEE7 average of 40.8 percent. Adding in PESR spending, the general government expenditure-to-GDP ratio has averaged 35.3 percent since 2009. North Macedonia’s current spending (84 percent of total spending) and capital spending (16 percent) shares also compare favorably with peers. Still, close examination of the economic and functional composition of public spending suggests, as for its neighbors, scope for improvement in its efficiency and quality. Figure 1.18. General Government Spending, 2014–17 Figure 1.19. Cumulative change in Expenditures, Average 2009–17 Average Percent of GDP 50 Current expenditures 45 Wages and allowances 40 Goods and services Pensions 35 Unemployment benefits 30 Social benefits 25 Health 20 Subsidies 15 Other transfers (SRA) 10 Interest 5 Capital expenditures Total expenditures 0 V E B H N K T A U R D B S 6 7 HR MN SR BI SV SV ES LV LT BG MK AL KO WB STEE -2.0 -1.5 -1.0 -0.5 0 0.5 1.0 1.5 JJCapital expenditures JJCurrent expenditures Percent of spending Figure 1.20. General Government Expenditure by Figure 1.21. Central Government Expenditure by Economic Classification, 2017 Functional Classification, 2017 Percent Percent 5.5 36 1.4 0.3 6.7 2.4 10 1.5 12 3 4.4 3.8 8 2 16 11 9.4 11 JJWages JJGoods and services JJPensions JJHealth JJGeneral public services JJDefense JJPublic order and safety JJEconomic affairs JJUnemployment and social benefits JJSubsidies JJOther transfers JJEnvironmental protection JJHouse building and public utilities JJHealth JJInterest JJCapital expenditure JJRecreation, culture and religion JJEducation JJSocial protection Source: MOF data and World Bank staff calculations. Chapter 1: The Need for Fiscal Reforms 21 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 1.3.1  Spending by Economic Classification Current spending in North Macedonia is marked by high and growing social protection spending, which warrants monitoring. The country’s social protection spending is among the largest in the WB6 region, averaging 15 percent of GDP for the past decade.10 It accounts for 45 percent of current general government spending and, except for interest spending and subsidies, it is the only recurrent spending budget category to have gone up. The rise is largely due to heavier spending on pensions since 2012. Pensions represent the general government’s single biggest spending item, accounting for 28 percent of the total. They have increased over time because of generous indexation, frequent ad hoc increases, cuts in contribution rates, and transition costs associated with introduction of a second pillar. In fact, pension and pension-related spending as a share of GDP grew from 8.8 percent in 2007 to 10.3 percent in 2016—high given current demographics. To meet the burden, the central government has had to make regular transfers from its budget to the pension fund., which at the end of 2016 accounted for over 40 percent of pension fund’s revenues, or 4.4 percent of GDP.11 Though pensions accounted for half of all social transfers in 2016, only 7.1 percent of pensioners are poor, compared to 28.6 percent of youth and 21.9 percent of the general population. Clearly, a reallocation of the country’s social spending could be beneficial in terms of reaching the most vulnerable segments of its population. The pensions system, and the risks it poses to the country’s fiscal situation, are addressed in Chapter 5. Although in contrast, social assistance spending in North Macedonia is low, there is clear scope to reduce overlaps and raise its efficiency. Though social assistance spending as a share of GDP went up from 1.0 to 1.3 percent between 2009 and 2017, that is still far below the WB6 average. North Macedonia has a wide array of cash benefits programs, among them a last-resort social financial assistance program, child allowances (CAs), and parental and disability benefits. There is a need to eliminate potential overlaps, to clarify the relation between categorical benefits, and means-test the benefits. Moreover, while the country’s targeting of social assistance is good, coverage is poor and much lower than spending levels would suggest . Spending on unemployment benefits is also remarkably low, despite a 2017 unemployment rate of 22.4 percent: it has averaged 0.4 percent of GDP since 2009 and is declining. North Macedonia’s social assistance system, spending by program, and program efficiency levels, are addressed in Chapter 6. Transfers to the Health Insurance Fund account for about 25 percent of all social transfers and has been declining. In contrast with regional peers and EU countries, North Macedonia’s total health expenditures as a share of GDP gradually declined between 2008 and 2016 from 4.6 to 4.3 percent of GDP, and out-of-pocket spending is very high. Meanwhile, arrears have been accumulating, which reached 0.6 percent of GDP at year-end 2016, which implies structural inefficiencies.12 Finally, because of its current adjustment structure, the system tends to favor the least efficient over the most efficient facilities. Chapter 8 addresses these issues in depth. Subsidies have largely been constant but are high by regional standards. In 2017 they constituted 2.4 percent of GDP, not counting state aid provided through tax relief and other tax expenditures. The 10 Social protection spending in North Macedonia includes pension and disability insurance (52 percent of social transfers), transfers to the Health Insurance Fund (25 percent), social assistance (7 percent), and unemployment benefits (1 percent). 11 This included both standard transfers and transfers to cover the pension fund deficit and the transition costs. 12 About two-thirds of these are arrears between public health providers, which can be cleared by netting-out. 22 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE lion’s share of subsidies goes to agriculture, as support to farmers and rural development programs. Yet agriculture’s contribution to GDP has been falling for the past decade, with other sectors generating greater value added—a characteristic of structural transformation. Compared to regional standards and to countries with similar economic structures and development challenges, the relative share of North Macedonia’s support for agriculture. Moreover, a significant part of it is directed to uncompetitive products like tobacco, which depresses growth and delays economic transition. World Bank analysis has found that given the available technology an average farm in North Macedonia could produce the same output using 55 percent less inputs. North Macedonia’s generous—and inefficient—system of agriculture support is explored in chapter 9. All other major spending categories—wages, goods and services, and interest spending—are relatively low or on par with regional peers. North Macedonia’s general government wage bill is the second lowest in the region, averaging 7.3 percent of GDP since 2009, against the regional average of 9.6 percent. In line with regional trends, it has declined in recent years because public-sector wages were largely frozen between 2009 and 2016. Spending on goods and non-labor services is also lower than most regional peers. From 2009 through 2017, general government spending on goods and services dropped from 5.3 to 3.6 percent of GDP, which is significantly less than the WB6 average of 5.6 percent. As that category includes maintenance and repairs, the country’s low spending has contributed to under- provision of buildings and infrastructure maintenance, and a deterioration of public assets, such as roads, schools, and hospitals.13 If unaddressed, this could eventually pose significant additional costs for the government and service users. While interest spending remains relatively low, between 2009 and 2017 it more than doubled as a share of GDP, from 0.6 to 1.4 percent as direct central government debt shot up from 23.6 to 39.2 percent of GDP. And interest costs are bound to increase further, because of rising debt and the longer maturity and repayment schedules of the remaining concessional loans with lower interest rates. Capital spending has increased, but maintenance funds are not efficiently allocated. Since 2009 North Macedonia’s capital spending, including that of the Roads Agency, has averaged 14.4 percent of total spending , or 5.1 percent of GDP. This is one of the highest ratios among WB6 countries and new EU Member States. Furthermore, it has been rising steadily risen for the past decade. In part this trend can be explained by the government’s initiative to scale up investments in upgrading and extending the road network. The initiative saw the Road Fund transformed into the PESR and moved off budget in 2013. Recent investments in new construction have come at the expense of maintenance spending, which at €7.029 /km is less than what is needed to keep he network in acceptable condition and lower than regional comparators, though they also do not spend enough on maintenance. The budget for maintenance is set as a residual—what is left after subtracting the estimated costs of construction, upgrading, and rehabilitation, to be carried out by the PESR, and other large expenditures, such as loan repayments, from PESR’s income. Moreover, the budget limitations mean that the quality of maintenance is not satisfactory. Maintenance needs are projected to remain high and should now be given priority over new construction in order to avoid worsening of long-term maintenance costs from persistent under-spending. Capital spending has consistently been under-realized. While between 2009 and 2017 central government budgeted capital spending averaged 14 percent of total spending (4.9 percent of GDP) 13 See Public Expenditure Review: Fiscal Policy for Growth (http://documents.worldbank.org/curated/en/895641468269982851/FYR-Macedonia-Public- expenditure-review-fiscal-policy-for-growth). Chapter 1: The Need for Fiscal Reforms 23 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 27 percent of the budgeted funds were not used. Under-realization is likely to stem from over-budgeting to begin with, low absorption capacity among budget users, and the fact that capital expenditures were treated as a residual to achieve fiscal targets. However, most capital spending in North Macedonia is directly related to the performance of non-tax revenues and foreign grants—their main source of financing—which have also been under-performing. Hence, a significant part of the under-execution of capital spending is in fact neutral to the budget deficit. Figure 1.22. Budgeted and Realized Capital Figure 1.23. Structure of Completed Public Expenditures, 2009–17 Construction Works, 2006–17 Percent of GDP Percent of total construction 8 70 7 60 6 50 5 40 4 30 3 20 2 1 10 0 0 9 0 1 2 3 4 5 6 7 6 7 8 9 0 1 2 3 4 5 6 7 200 201 201 201 201 201 201 201 201 200 200 200 200 201 201 201 201 201 201 201 201 JJRealized capital expenditures JJOriginally budgeted capital expenditures ÂÂReconstruction and maintenance ÂÂInvestment in infrastructure ---- Investment in residential and nonresidential buildings Source: MOF data and World Bank staff calculations. Source: State Statistics office and World Bank staff calculations. The quality of capital spending, in terms of its potential to support economic growth, also seems to have fallen in recent years. It is not clear how public investments are prioritized in North Macedonia, and there are no criteria for ranking alternatives because the Organic Budget Law has no specific requirements for appraising, selecting, monitoring, or evaluating projects. Between 2009 and 2014, as “Skopje 2014” was carried forward, spending on administrative buildings increased at the expense of investments in infrastructure and maintenance. In 2009, 81 percent of public investment was spent on either infrastructure (63 percent) or maintenance (18 percent); only 19 percent for administrative buildings. By 2012, under “Skopje 2014,” investment in administrative building had reached 33 percent. This likely had negative consequences for economic growth—spending on administrative buildings does not have the same economic impact as infrastructure investment. However, the trend has now reversed: the share of investment in infrastructure is again 63 percent, largely because of construction of the two Corridor VIII highways. 1.3.2  Spending by Functional Classification By function, social protection, economic affairs, health ,and education account for most of the CG’s spending between 2010 and 2016. Because of frequent and ad-hoc increases in pensions and social benefits, social protection has risen consistently, reaching 36.4 percent of CG spending in 2016—10.7 percent of GDP. Health spending is the second biggest item, averaging 16 percent of CG spending—4.4 percent of GDP—though for the past three years spending on health has somewhat declined. Economic affairs accounted for an average of 12 percent of spending, 4 percent of GDP, but 24 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE declined steadily throughout that period, largely as a result of decline in goods and services spending, to constitute 3.4 percent of GDP in 2016.14 North Macedonia currently spends less on public education than WB6 peers, and its outcomes are among the lowest. Its education spending averaged 4.1 percent of GDP between 2010 and 2016, close to the regional average, but it has been going down steadily reduced and is now lower than that of the EU and peer countries. A deterioration in the quality of primary education and the relevance of secondary education has accompanied the spending drop. For example, the 2015 Program for International Student Assessment (PISA) test showed a substantial decline in students’ education outcomes since the 2001 test, and found that about two-thirds of North Macedonia’s students have less than basic proficiency in all three subjects (math, science and reading). Chapter 7 delves into North Macedonia’s education system. Box 1.3. General Government Spending Compared A. Wages Spending B. Social Protection Spending Percent of GDP, 2017 Percent of total expenditure, 2016 or latest year available 14 45 12 40 35 10 30 8 25 6 20 15 4 10 2 5 0 0 MN E V H N U K S A HR BI SV LT SV KO LV EST SRB MKD BGR ALB MK D BG R SRB LTU MN E HR V LVA EST ALB BIH KO S ÂÂWB6 average ÂÂSTEE7 average ÂÂWB6 average C. Pensions and Pension Related Spending D. Social Assistance Spending Percent of GDP, 2015 Percent of GDP, 2016 or latest year available 14 50 12 40 10 8 30 6 20 4 2 10 0 SRB SVN MNE HRV MKD IH-RS -FBIH BGR SVK ALB EST LVA LTU KOS 0 D B BIH BIH HR V KO S EST LTU SRB ALB MN E MK BG R LVA ÂÂWB6 average ÂÂSTEE7 average ---- EU28 average ÂÂWB6 average ÂÂSTEE7 average (continued on next page) 14 The main components of the category are: general economic, commercial and labor activities; agriculture, forestry, hunting and fishing; transport; and other economic activities (which account for most of the spending in this category and are largely comprised of goods and services spending). Chapter 1: The Need for Fiscal Reforms 25 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 (Box 1.3. continued) E. Subsidies F. Agriculture Transfers Spending Percent of GDP, 2017 or latest year available Percent of GDP, 2010–2015 average 3.0 1.2 2.5 1.0 2.0 0.8 1.5 0.6 1.0 0.4 0.5 0.2 0 0 D SRB RV GR BIH LVA VN NE EST VK ALB LTU OS D SRB BIH S E ALB MK H B S M S K MK KO MN ÂÂWB6 average ÂÂSTEE7 average ---- EU28 average ÂÂWB6 average ---- EU28 average G. Health Spending H. Education Spending % of GDP, 2016 or latest year available % of GDP, 2016 or latest year available 8 7 7 6 6 5 5 4 4 3 3 2 2 1 1 0 0 B R D A B SVK BIH HRV SVN SR BG LTU EST MK LV MNE AL KO S EST SVN LVA LTU BIH HRV KOS MNE SRB SVK MKD BGR ALB ÂÂWB6 average ÂÂSTEE7 average ---- EU28 average ÂÂWB6 average ÂÂSTEE7 average ---- EU28 average Source: World Bank staff calculations. 1.3.3  Budget Rigidity Rigid budget items have taken an increasing share of North Macedonia’s government expenditures over time. In other words, the government has less flexibility to reallocate spending to items it deems to be a priority, or from less efficient sectors to those that have greater economic impacts. Between 2009 and 2017 the share of government expenditure occupied by high-rigidity items increased from 66 to 69 percent. For example, interest payments on the national debt, which are highly rigid, more than doubled from 1.7 to 4.0 percent of total government spending. It is important to look into the possibility of enhancing the flexibility of some rigid expenditures. The variance-covariance matrix (Table 1.3) shows that pensions occupy a disproportionate share of CG expenditure variance. Furthermore, their’ share of CG spending went up from 25 to 28 percent between 2009 and 2017. Future budget exercises should particularly scrutinize pension. 26 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 1.24. Budget Rigidity, 2011–17 MKD billion Percentage of total spending 250 100 200 80 150 60 100 40 50 20 0 0 9 0 1 2 3 4 5 6 7 9 0 1 2 3 4 5 6 7 200 201 201 201 201 201 201 201 201 200 201 201 201 201 201 201 201 201 JJHigh JJMedium JJLow JJHigh JJMedium JJLow Source: World Bank staff calculations; data from North Macedonia’s BOOST database for 2011 to 2016. Note: High-rigidity items are wages and allowances, transfers to LGs, interest payments, and social benefits. Medium-rigidity items are goods and services and subsidies and transfers. Finally, low-rigidity items include reserves, undefined expenditures, and capital expenditures. Table 1.3. Budget Rigidity Variance-Covariance Matrix (2006–16)15 Wages and Goods and Pensions Unemployment Social Subsidies Other transfers Capital Health Interest allowances services benefits benefits (SRA) expenditures Wages and 1.62 allowances Goods and 0.62 0.57 services Pensions 4.47 1.77 13.23 Unemployment -0.24 -0.11 -0.71 0.04 benefits Social benefits 0.81 0.34 2.32 -0.12 0.42 Health 1.66 0.59 4.72 -0.25 0.84 1.74 Subsidies 0.98 0.46 2.99 -0.16 0.53 1.04 0.72 Other transfers 0.26 0.12 0.76 -0.04 0.13 0.28 0.18 0.06 (SRA) Interest 1.15 0.44 3.32 -0.18 0.60 1.21 0.75 0.20 0.86 Capital 1.19 0.28 3.09 -0.14 0.61 1.20 0.68 0.17 0.86 1.45 expenditures Source: World Bank staff calculations; data from North Macedonia’s BOOST database. 1.4  Efficiency of Public Expenditure 1.4.1  Economic Efficiency According to the Public Sector Performance (PSP) Index, which measures seven areas of government performance, North Macedonia registered a relative improvement between 2006 and 2016. Still, its public services score low compared to peers. In 2006, North Macedonia’s PSP score was 0.90, placing it 14th out of the 16 countries analyzed.16 That year, North Macedonia’s score was 12 percent lower than 15 The government budget execution for the years 2006 to 2016 was used to compile this variance-covariance matrix, whereby each component is divided by the variance of total government expenditures. For a detailed description of the methodology used, see Barrot and Battaile (2013). 16 The countries contained in this analysis are Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, North Macedonia, Hungary, Latvia, Lithuania, Montenegro, Poland, Romania, Serbia, Slovakia, and Slovenia. Chapter 1: The Need for Fiscal Reforms 27 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 the sample mean of 1.02. By 2016, North Macedonia’s PSP score had risen to 0.97, while the average had remained relatively consistent at 1.03. Therefore, in 2016, the gap between North Macedonia and the sample average was reduced to 5 percent, and it was ranked 10th. North Macedonia’s PSP index score suggests that the performance of its public sector lags behind that of regional peers, but that it has made some progress towards bridging that gap in the past decade. Figure 1.25. Public Sector Performance Index, 2006 Figure 1.26 Public Sector Performance Index, 2016 PSP index PSP index 1.3 1.3 1.2 1.2 ALB 1.1 1.1 ALB 1.0 1.0 MNE MKD MNE 0.9 MKD 0.9 BIH SRB SRB 0.8 BIH 0.8 0.7 0.7 0.6 0.6 20 25 30 35 40 45 50 55 20 25 30 35 40 45 50 55 Public sector expenditure, percent of GDP Public sector expenditure, percent of GDP Source: World Bank staff calculations. Note: The Public Sector Performance Index measures outcomes in seven government-related sectors: administrative quality, educational achievements, health outcomes, public infrastructure quality, income distribution, economic stability, and economic performance. For more details, see Afonso, Schuknecht, and Tanzi (2003). The PSE Index suggests that North Macedonia’s Figure 1.27. Public Sector Efficiency Index by Sector Spending, 2016 spending does not maximize public sector PSE index performance. The PSE index combines the PSP 2.0 index with government spending on each of the 1.8 ALB measure’s seven categories as a share of GDP. While North Macedonia scores relatively better 1.6 on the PSE than on the PSP index, its PSE index 1.4 score is below the regional trend line (though only 1.2 MKD marginally), which suggests that the efficiency 1.0 of its public spending could be improved. MNE SRB BIH 0.8 Furthermore, that North Macedonia can score higher on the PSE than on the PSP may well be 0.6 20 25 30 35 40 45 50 55 because it spends less on these areas than its Public sector expenditure, percent of GDP peers (see Figure 1.28). In fact, on average North Source: World Bank staff calculations. Note: The Public Sector Efficiency index is calculated by combining Public Sector Macedonia’s government spending between 2006 Performance scores and the government spending incurred to achieve these outcomes. See Afonso, Schuknecht, and Tanzi (2003). The performance scores used are for 2016; the and 2013 was the second lowest of the countries these data because there can be a lag effect between spending in a specific sector and government spending figures are averages for 2006–2013. The decision was made to use sampled as a share of its GDP. Similarly, North measurement of the impacts of that spending. Macedonia’s PSE score would almost certainly suffer if it were to increase its spending without a comparatively greater heightening of efficiency, because some of the PSE outcome measures exhibit diminishing returns to inputs. North Macedonia’s PSE scores vary significantly by public service area. In 2016, it scored 2nd on administration, 3rd on economic stability, and 4th on income distribution, However, it scored 11th on economic performance and 12th on infrastructure. North Macedonia’s low score on infrastructure 28 Chapter 1: The Need for Fiscal Reforms SOWING THE SEEDS OF A SUSTAINABLE FUTURE suggests that, while its spending on infrastructure is in line with that of its peers, as noted earlier, that spending fails to produce the expected results. In other words, North Macedonia could have better infrastructure outcomes using the same funds or use fewer resources to achieve its current results. Figure 1.28. Public Spending by GDP per Capita 2016 Figure 1.29. Public Sector Performance Score by Sector Spending, DEA Analysis, 2016 Public expenditure, percent of GDP DEA Theta 60 1.2 50 1.0 ALB BIH 40 0.8 SRB MKD BIH 30 ALB MKD MNE 0.6 MNE SRB 20 0.4 10 0.2 0 0 10,000 20,000 30,000 40,000 20 25 30 35 40 45 50 55 GDP per capita PSP score Source: World Bank staff calculations. Source: World Bank staff calculations. A Data Envelopment Analysis (DEA) also suggests that North Macedonia could achieve better public performance using the same resources. The DEA measures the efficiency of specific inputs in generating a given outcome compared to a production possibilities frontier estimated using the most efficient case sampled in the data. This DEA analysis of North Macedonia and 15 regional peers gives equal weights to per capita GDP and per capita public expenditures as inputs, and measures PSP scores as the outcome variable. Using 2016 data, North Macedonia ranked 6th among the 16 countries sampled in terms of the PSP scores achieved with its per capita GDP and public spending levels. Again, the fact that North Macedonia appears below the trend line in Figure 1.29 suggests that its PSP level could be higher using the same inputs and at its current income level. In conclusion, whether North Macedonia’s government can sustain its current spending, tax, and other policies through the medium to long term without jeopardizing solvency or defaulting on liabilities or promised spending hinges on the success of fiscal consolidation measures. The next chapters explore opportunities for delivering the needed fiscal consolidation through enhanced revenue collection and potential efficiency gains. Chapters 2 and 3 examine institutional aspects by exploring PFM bottlenecks and the challenges facing LG financing and delivery of social services. Chapters 4 and 5 identify options for addressing fiscal risks and working toward fiscal consolidation. They focus on reform options to raise revenue and improve the fiscal sustainability of the pensions system, which are central to delivering the needed fiscal consolidation. Section 3 (chapters 6, 7 and 8) analyze how the government of the Republic of North Macedonia can deliver higher quality and more efficient public social assistance, education, and health services. Finally, Chapter 9 reviews the scope for efficiency-enhancing reallocations of government expenditures on agriculture. Chapter 1: The Need for Fiscal Reforms 29 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Chapter 2: The State of Public Financial Management The Government of North Macedonia has adopted an ambitious and comprehensive PFM Reform Program which, could address many of the PFM weaknesses identified in recent diagnostic reports. This program addresses seven core pillars of a well functioning PFM system and sets out a plan of action and a generalized schedule of deliverables. While the schedule allocates a relatively short timeframe for achieving some key reforms and there are questions about the capacity of some institutions to deliver, the government appears to be committed to the program objectives and plans to launch it through a series of action plans. . This chapter argues that the priority for the short term should be to address the pervasive challenge of accumulated expenditure arrears by adopting new modules for registering commitments and creating a published register of budget users who do not report expenditure arrears promptly. In the medium term, institutions need to build capacity to produce credible macrofiscal forecasts as a start on addressing weaknesses in translating fiscal strategy into a credible budget that agencies can execute. The chapter recommends that the strategy be extended to include measures to monitor and report on off-budget operations and control fiscal risk. This could be done using comprehensive fiscal risk management reporting and an institutional framework for analyzing fiscal risk. Finally, further work is needed in improving internal audit mechanisms so that they provide specific actionable reform options for budget institutions. 2.1  Sector Overview Improving North Macedonia’s PFM system is still a major institutional and governance challenge. Despite improvements in the past few years in its PFM practices, such as introduction of the medium- term fiscal framework and a single investment pipeline, several reports in recent years point out that North Macedonia could benefit from further action to strengthen performance across all PFM elements. Finally, comparing North Macedonia’s relative PFM performance to that of peer countries using the most recent available Public Expenditure and Financial Accountability (PEFA) reports17 clearly demonstrates that North Macedonia—like most WB6 countries—has considerable room for PFM improvement.18 While North Macedonia has relatively orderly budgeting and payment arrangements, actual aggregate revenues have consistently fallen short of budget estimates. With expenditures rising, this has operated to expand the fiscal deficit. In addition, the reporting of significant payment arrears in mid- 2017 shed light on systematic shortcomings in the PFM system. Some of the most prominent of these: •• There is a lack of institutional capacity to produce high quality and credible macrofiscal forecasts to support the MTFS. 17 These reports are prepared at different times but use the PEFA 2011 Methodology. 18 Because this chapter does not cover external scrutiny and parliamentary oversight, those indicators are excluded from the comparator set. 30 Chapter 2: The State of Public Financial Management SOWING THE SEEDS OF A SUSTAINABLE FUTURE •• The MTFS has no binding aggregate spending Figure 2.1. PEFA Comparison ceilings and budget allocations are not well- Credibility of the budget 8 linked to government policy priorities. This 7 6 results in a short-term focus for fiscal policy, 5 4 inability to control line ministry spending, and Predictability and 3 Comprehensiveness control in budget and transparency sub-optimal delivery of essential services. 2 execution 1 Since there are no multi-year targets, multi- 0 year commitments are not measured against targets, and crowd out future fiscal decisions— another potential cause of expenditure Accounting, recording Policy budgeting arrears. and reporting ÂÂALB 2017 ÂÂBIH 2014 ÂÂKOS 2016 ÂÂMKD 2015 ÂÂMNE 2013 ÂÂSRB 2015 Source: World Bank Staff calculations. •• Investment decisions are not well-linked to Note: Ordinal scale converted to numerical scale (A = 8; B+ = 7; etc.). government priorities and sectoral strategies. Projects that enter the budget are neither prioritized to reflect policy objectives nor properly costed, and their social and economic returns are not adequately costed for the project life cycle. •• Commitment control mechanisms (such as multiyear commitments) are deficient and have failed to halt the accumulation of arrears. Though commitment control measures are in place, they are not universally respected by budget institutions. •• The treasury system has not been very effective in monitoring and controlling commitments. The present system prevents MOF from maintaining a comprehensive centralized record of financial liabilities, such as arrears. •• There are weaknesses in accounting and government financial reporting of government assets and financial liabilities. Furthermore, there is a lack of transparency in the reporting of off-budget expenditures. •• Finally, institutional structures to oversee fiscal risks are underdeveloped. There is no clear line of accountability for assessing and reporting fiscal risks, either in the MF or elsewhere in the government. The ambitious and comprehensive PFM reform strategy to which the Government has committed identifies the questionable capacity of some institutions to deliver, but nevertheless sets a relatively short schedule for accomplishing some key reforms. The Government appears committed to the objectives of the Strategy and plans to implement it through a series of annual action plans. The 2018 Action Plan adopted in February details the activities for 2018, their costs, and sources of financing, and defines targets and indicators. Chapter 2: The State of Public Financial Management 31 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 2.2: North Macedonia: Structure of the Budget PUBLIC SECTOR GENERAL GOVERNMENT (CONSOLIDATED) BUDGET OF NORTH MACEDONIA PUBLIC ENTITIES REGULATORY INSTITUTIONS Second level budget users: 290 units 9 units CENTRAL GOVERNMENT CENTRAL GOVERNMENT STATE-OWNED BUDGET ENTERPRISES First level budget users: PUBLIC ENTITIES 27 units 93 units IN HEALTH SECTOR Second level budget users: 112 units HEALTH SECURITY FUND SOCIAL SECURITY FUNDS EMPLOYMENT AGENCY OF NORTH MACEDONIA PENSION AND DISABILITY FUND MUNICIPALITIES' BUDGETS PUBLIC ENTITIES STATE-OWNED LOCAL GOVERNMENT ENTERPRISES First level budget users: Second level budget users: ? units 81 units 556 units JJBudget/financial plan approved by the Parliament JJBudget/financial plan indirectly approved by the Parliament JJBudget/financial plan not approved by the Parliament Source: World Bank 2015. 2.2  Institutional Shortcomings Related to PFM: Selected Issues 2.2.1  Policy-based Fiscal Strategy Development and Budgeting (Formulation, Management and Monitoring) Macroeconomic and Fiscal Forecasting For an effective MTFS, North Macedonia needs more capacity to produce credible macrofiscal forecasting. The MOF Macroeconomic Policy Department provides macroeconomic forecasts and some macrofiscal analysis ,such as calculation of the primary and structural deficits and sensitivity analysis of the budget deficit, but the Ministry lacks a unit dedicated to producing and overseeing the macrofiscal forecasting framework. The present process is fragmented across the Macroeconomic and Budget Departments, the Public Revenue Office, and Customs, which in turn gather information from other departments and institutions. As a result, no one owns the forecast, which dilutes accountability. Revising the current institutional structure would address concerns about accountability and help make the forecast more accurate and implementation of the MTFS more successful. The MOF has already moved toward this goal by undertaking an EU-funded twinning project to better align medium-term budget planning and fiscal reporting with EU standards by, e.g., using an improved macroeconomic model to reinforce medium-term macroeconomic forecasting and improved tax calculation models to produce more accurate medium-term tax revenue estimates.19 In addition, the 2019–2021 Fiscal 19 The twinning program brought North Macedonia together with the Finnish Ministry of Finance through the project “Strengthening the Medium-Term Budgeting for Effective Public Financial Management”. The program terminated in November 2017. 32 Chapter 2: The State of Public Financial Management SOWING THE SEEDS OF A SUSTAINABLE FUTURE Strategy adopted in May 2018 for the first time contains a chapter on macrofiscal risks and possible consequences. Budget Preparation The limited participation of stakeholders undermines the Government’s ability to translate macroeconomic, fiscal, and sector policies into the budgeting process. The Organic Budget Law (OBL) sets out a regular calendar (Table 2.1) for the dissemination and approval of key documents, which is widely followed by stakeholders. Nevertheless, the budgeting process does not integrate a top- down fiscal strategy with bottom-up development plans by spending units, which impacts expenditure outcomes vis-à-vis the MTFS and leads to constant deviations from the planned fiscal framework. Expenditure ceilings reflect the previous year’s budget rather than government-wide decisions, and they are not enforced. Table 2.1. North Macedonia - Budget Calendar, 2014–16 Budget law Budget year 2013 Budget year 2014 Budget year 2015 1 Strategic priorities determined by Government 15 April 24.4.2012 1.5.2013 9.9.2014 2 Fiscal strategy approved by Government 31 May n/a 19.9.2013 3.10.2014 In the period 2012–2014 (for Budgets 2013, 3 Ceilings (limits) approved by Government 31 May 2014, 2015) the limits were set at the end of the budget of the previous year 4 MoF issues Budget Circular 15 June 15.6.2012 15.6.2013 9.6.2014 Spending units submit Development Spending units submit Development Program Program Plans in their budget request in 5 15 July Plans a defined form as a special development subprograms Development program plans is an integral part of the Budget as a special part Development Program Plans approved by Development part of Budget of Republic of 6 15 August Government North Macedonia and they are approved by the Government when the Budget of Republic of North Macedonia is approved 7 Spending units submit budget requests 1 September 1.9.2012 1.9.2013 1.9.2014 8 Government submits budget to the Parliament 15 November 4.11.2012 19.9.2013 3.10.2014 9 Parliament approves the budget 31 December 26.12.2012 20.12.2013 16.12.2014 Source: MOF and World Bank staff elaboration. Investment decisions are not well-linked with government priorities and sectoral strategies. The Organic Budget Law (OBL) currently requires budget users and municipalities to prepare a draft development plan for capital spending programs, in accordance with the circular issued during the budget planning process. However, the OBL makes no reference to the requirements for projects’ economic appraisal, costing, selection, and monitoring and evaluation, which represents an important shortcoming. Though some ministries have prepared development strategies, they have not been reflected in the fiscal envelope of the MTFS. Investment projects that enter the budget are not prioritized to reflect Government policy objectives, and social and economic returns are not adequately costed for the entire project life cycle. The 2015 PEFA (World Bank 2015) found that North Macedonia falls short of identifying the full lifecycle costs Chapter 2: The State of Public Financial Management 33 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 of projects, and that the necessary regulations to do so are not yet in place (PEFA 2015).20 Transparent criteria to determine whether investment project proposals are in line with government priorities— and whether they are prioritized by their economic and social returns—are vital. The government needs to develop an institutional framework that will ensure that investments are planned, allocated, and carried out efficiently. Investment projects chosen need to be fully costed for their whole project life cycle to include subsequent recurrent operating costs (such as maintenance and repair) and/or the assets to be generated. Multiyear budget spending ceilings are not treated as binding, and there is no effective control over accumulated obligations for subsequent budget cycles. Medium-term planning is reflected in the Fiscal Strategy, which, in accordance with the OBL is adopted on a rolling basis each year for a three-year period. The Strategy provides guidelines and objectives of fiscal policy, and outlines basic macroeconomic projections and indicators. It also estimates revenues, appropriations, the budget deficit, and debt for that period. Nevertheless, beyond the immediate budget year, the aggregates included in the MTFS are not viewed as binding targets or ceilings but simply as indicative figures reflecting current projections. Since there are no binding multiyear targets, multiyear commitments are not measured against targets, and can crowd out future fiscal decisions, or potentially create arrears. The MTFS is revised every year, but no record is kept of the rationale for adjusting targets in subsequent MTFSs. Budget Reliability There have been budget deviations in recent years, in part because North Macedonia has limited capacity for macrofiscal forecasting  (Figures 2.3–2.6). In each of the past three years both revenues and expenditures have been overestimated. On the revenue side, most of the shortfall occurs because of the VAT—not surprisingly since it is the largest source of revenue. Non-tax revenues, such as capital revenues or foreign donations, are also significantly overestimated. However, the PIT and CIT are often underestimated, partly due to changes in government policies. On the expenditure side, spending on relatively inflexible items such as wages and transfers (pensions and other social benefits) tends to be realized as budgeted. However, due to recurring shortfalls on the revenue side, the government must cut spending on goods and services, as well as capital, which tend to be severely underrealized—by an average of 22 percent for 2014–2016. Overoptimistic budgeting decisions and unrealistic budgetary allocations have been more problematic in some sectors than others. For example, it is evident that the education budget allocation to finance delegated functions is inadequate to cover all the costs related to transport of children to schools (particularly secondary school);21 and the health budget does not adequately cover the cost of treatments offered. Addressing these issues might begin with re-assessing and reinforcing the connection between funding decisions and policy priorities, and between funding levels and the outcomes for end-users. The problem could also be addressed through program budgeting, but that is an advanced reform which should be introduced gradually once the Government is better able to manage the current budgeting process. 20 The 2015 Public Expenditure Review (PER) by the World Bank also found that operational and mainteneces costs were not considered in assessing public investments, in particular those for transport infrastructure. 21 In 2008 North Macedonia introduced mandatory secondary education and authorites had to provide transport of students to schools. 34 Chapter 2: The State of Public Financial Management SOWING THE SEEDS OF A SUSTAINABLE FUTURE The PFM Strategy outlines a number of activities to address weaknesses in the current fiscal strategy and the budget formulation procedures outlined above. For example, one of the actions would build up the capacities of the MOF Tax and Customs Policy and Macroeconomic Policy Departments through staffing increases and developing new tax revenue and macroeconomic models. The Strategy also proposes activities to put in place a comprehensive medium-term budget framework (MTBF) linked to the overarching fiscal strategy, and will become the key document establishing priorities and allocation of resources to different budget users in a multiannual expenditure planning. To do this, the PFM strategy recommends providing additional resources to improve the capacity of the MOF Budget and Funds Department and budget users, in line with the new tools, budget planning process, and medium-term budget reforms. Finally, the government plans a comprehensive diagnostic of its public investment management (PIM) system. Figure 2.3. Original and Revised Budgets and Actual Figure 2.4. Original and Revised Budgets and Actual Revenues, 2014–16 Expenditures, 2014–16 Percent of GDP Percent of GDP 31 34 30 33 29 32 28 31 27 26 30 2014 2015 2016 2014 2015 2016 JJOriginal budget JJRevised budget JJOutturn JJOriginal budget JJRevised budget JJOutturn Figure 2.5. Original Budget and Actual Revenues, Figure 2.6. Original Budget and Actual Expenditures, 2014–16 2014–16 Percent Percent 140 140 120 120 100 100 80 80 60 60 40 40 20 20 0 s 0 t PIT CIT VAT ise axe s tion s ues ues ges and fers res ital Exc er t ibu e ven e ven Wa ds ices ns Inte Captures t h t r x r er r Gooserv Tra i O Con n ta Oth end No exp JJ2014 JJ2015 JJ2016 JJ2014 JJ2015 JJ2016 Source: MOF and World Bank staff calculations, Note: The CIT outcome in 2015 reflects the re-introduction of taxation on corporate earnings. The draft OBL would also, if enacted, address some of the weaknesses already identified and outlined above. It proposes adoption of fiscal rules to provide an explicit and credible guide for future fiscal policy and establishment of a fiscal council to oversee implementation of fiscal policy. Finally, the law would amend the medium-term fiscal strategy to incorporate a MTBF and aggregate and budget user ceilings. Chapter 2: The State of Public Financial Management 35 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Budget Comprehensiveness and Fiscal Transparency North Macedonia has a mixed record of fiscal transparency, according to international rankings. In the 2017 Open Budget Survey, it scored lower than the global average on the usefulness of the budget information made available to the public throughout the budget cycle. Its overall score was unchanged from its 2015 ranking, and it performed poorly relative to regional peers (Figures 2.7 and 2.8). Of particular concern was the lack of opportunities for the public to engage in the budgetary process. Nevertheless, the Survey was carried out in the first half of the year—before the new government took office and before some significant improvements to transparency, like publication of the Citizen’s Budget—were implemented. The country is likely to score higher in the next iteration of the survey. Figure 2.7. Ranking on the Open Budget Index Figure 2.8. Elements of Public Participation BIH MKD MKD SRB SRB ALB ALB BIH HRV HRV SVN SVN Global average Global average 0 10 20 30 40 50 60 70 0 5 10 15 20 25 30 Index Index Source: Open Budget Survey 2015. Though North Macedonia’s budget documentation is reasonably comprehensive, there are gaps. The country scores relatively high for the comprehensiveness of budget documentation, with key documents made available to Parliament. Among them are budget breakdowns for the current, previous, and three forthcoming years; debt indicators and deficit financing figures; and key macroeconomic assumptions. However, the budget documentation falls shorts with regards to disclosing information on fiscal risks and contingent liabilities, financial assets, the quantification of tax expenditures, and the budget implications of new policy initiatives. These gaps impact North Macedonia’s budget reliability. Table 2.2. Comprehensiveness of Budget Documentation Element/Requirements Met (Y/N) Evidence Used/Comments Forecast of the fiscal deficit or surplus or accrual Yes Fiscal Strategy 2017–19 and Budget 2018 operating result Previous year’s budget outturn, presented in the Yes Fiscal Strategy 2017–2019 same format as the budget proposal. Current fiscal year budget presented in the same Yes Fiscal Strategy 2017–19 format as the budget proposal. Aggregated budget data for both revenue and spending, according to the main classifications used, including data for the current and previous year with Yes Fiscal Strategy 2017–19 a detailed breakdown of revenue and expenditure estimates. Deficit financing, describing its anticipated Yes Fiscal Strategy and Budget 2018 composition. 36 Chapter 2: The State of Public Financial Management SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 2.2. Comprehensiveness of Budget Documentation (continued) Element/Requirements Met (Y/N) Evidence Used/Comments Macroeconomic assumptions, including at least estimates of GDP growth, inflation, interest rates, and Yes Fiscal Strategy and Budget 2018 the exchange rate Debt stock, including details at least for the beginning of the current fiscal year presented in Yes Fiscal Strategy 2017–19 accordance with GFS or other comparable standard. Financial assets, including details at least for the No cash balances or other financial assets beginning of the current fiscal year, presented in No and projected movements are shown in accordance with GFS or other comparable standard. budget documentation. While there is information on public Summary information of fiscal risks, including guaranties included, public and publicly contingent liabilities such as guarantees, and guaranteed debt contingent obligations contingent obligations embedded in structure No embedded in structure financing financing instruments, such as public-private instruments such as public-private partnership (PPP) contracts, and so on. partnership (PPP) are not disclosed. Explanation of budget implications of new policy Major investments are not fully documented initiatives and major new public investments, with in the budget by including operational and estimates of the budgetary impact of all major No maintenance costs over the life cycle of the revenue policy changes and/or major changes to project. expenditure programs. Documentation on the medium-term fiscal forecasts. Yes Fiscal Strategy 2017–2019 Quantification of tax expenditures. No Source: MOF and World Bank staff (based on PEFA 2016 performance indicator No 5). Reporting and Monitoring Proper reporting and monitoring are critical for effective fiscal management and accountability. They help to highlight fiscal risks, allowing for an earlier and smoother fiscal policy response to changing economic conditions and reducing the incidence and severity of crises. In particular, proper reporting by, and monitoring of, units outside the central government, such as LGs, SOEs, and other extrabudgetary units are essential for containing fiscal risks. In that sense, the extent to which government financial reports reflect actual operations reduce fiscal risk. While there is reporting on budgetary and extra-budgetary institutions in North Macedonia, there is no consolidated report or adequate monitoring of the financial operations and risks of these institutions. CG operations, which are performed by budgetary institutions (such as the three Social Insurance Funds—Pensions, Health and Unemployment), are fully included in the consolidated budget. Not included are the operations of the 14 SOEs,22 a considerable number of LG utility companies owned, and 9 regulatory agencies, which are mainly financed through their own fees and charges. All these bodies provide regular—mostly quarterly—financial reports to MOF and sponsoring ministries, as well as semi-annual and annual reports to the government and national assembly, which compare their revenue and expenditure performance over time. Nevertheless, these reports have not been consolidated into an overall report on financial performance, and their fiscal risks have not been properly assessed. 22 These cover water supply, railway infrastructure, road construction and maintenance, radio and television broadcasting, the agricultural commodity exchange, the management of forest and pasture lands, and the Government Gazette. Chapter 2: The State of Public Financial Management 37 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 The monitoring and reporting of spending on roads and other infrastructure is of particular concern. The former Roads Agency was moved off-budget in 2013. Its activities are financed by a share of excise duty on road fuel, the revenue from road tolls, and government-guaranteed borrowing. PESR spending in 2016 amounted to more than 12 billion MKD, 7 percent of government spending, but it reports only to the government, and its operations and financial activities are not consolidated in any government report. Furthermore, the costs of very large building and infrastructure investments, such as the Skopje 2014 program, were not financed through the CG budget, but through bodies associated with LGUs. There is no comprehensive report on the total spending on which an assessment of the use of public resources could be based. Public Access to Fiscal Information Formally, North Macedonia’s fiscal information is publicly accessible. The only exceptions are resources provided to individual schools and health care institutions, and some deficiencies in the coverage of financial statements, in-year budget execution reports, and audit reports. This was confirmed by the 2015 PEFA Report. The new 2016 PEFA criteria23 comprise additional standards, including a prebudget statement, a comprehensive medium-term fiscal framework, and a convenient budget summary, which are available in North Macedonia, except for external audit reports. That said, there is scope to improve report content, the information provided, and their simplicity. Table 2.3. Public Access to Fiscal Information Element/Requirements Met (Y/N) Evidence used/Comments Annual executive budget proposal documentation: Information including the Budget Circular, A complete set of executive budget proposal the Fiscal Strategy, and the budget itself documents (as presented by the country in PI-5) Yes is published on the government and is available to the public within one week of their parliament websites when the budget is submission by the executive to the legislature. presented. Enacted budget: The annual budget law approved by The full text of the annual budget law the legislature is publicized within two weeks of its Yes as enacted is publicly available on the passage. websites of the MOF and the parliament. In-year budget execution reports: The reports are Summaries of monthly and quarterly routinely made available to the public within one Yes reports are published on the government month of their issuance, as assessed in PI-27. website within one month of completion. The annual report is according to the law submitted to the National Assembly by Annual budget execution report: The report is made June 30 for the previous year. The reports available to the public within six months of the end Yes and the State Audit Office audit reports of the fiscal year. are published on the National Assembly website (http://www.sobranie.mk)3. The annual audit report on the Audited annual financial reports, incorporating or government's financial statements is accompanied by the external auditor’s report. The Yes published as soon as it is presented to the reports are made available to the public within 12 house of assembly (also the treatment of months of the end of the fiscal year. year-end financial statements above). Additional elements 23 The 2015 PEFA assessment was done using 2011 PEFA criteria. The new 2016 PEFA criteria constitute the most comprehensive upgrade to the PEFA framework since it was first published in 2005 and incorporate new and updated benchmarks that reflect the changing landscape of PFM reforms and the evolution of good practices over the last decade. 38 Chapter 2: The State of Public Financial Management SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 2.3. Public Access to Fiscal Information (continued) Element/Requirements Met (Y/N) Evidence used/Comments Prebudget statement: The broad parameters for the executive budget proposal regarding expenditures, The Fiscal Strategy contains such planned revenues, and debt, are made available to Yes information. the public at least four months before the start of the fiscal year. Summary of the budget proposal. A clear, simple summary of the executive budget proposal, or the enacted budget, accessible to non-experts (often referred to as a “citizens’ budget”) and where A citizen budget has recently been Yes appropriate translated to the most commonly spoken prepared. local language, is publicly available within two weeks of the proposal’s submission to the legislature and within one month of the budget’s approval. Macroeconomic forecasts: The forecasts, as assessed Macroeconomic parameters are available in in PI-14.1, are available within one week of their Yes the Economic Reform Program and in the endorsement. Fiscal Strategy. Source: MOF and World Bank Staff (Based on PEFA 2016 Performance Indicator No 9). The PFM strategy identifies fiscal transparency as a priority reform area, but specific actions will need to be defined within the strategy to ensure comprehensive improvements against international benchmarks. Since June 2017 the government has been working to improve fiscal transparency. The MOF has published its first citizens’ budget, which accompanied the supplementary budget for 2017, and repeated the practice with the 2018 budget. It has also been making more fiscal information available on its website; in 2017, for the first time LG fiscal data and a complete set of macroeconomic indicators (including forecasts) was made available. Moreover, the MOF has significantly improved the format and accessibility of fiscal data, moving from PDF to Excel-based presentations. And all agreements on public procurement must now be published on the Public Procurement Bureau website. Finally, the MOF has published and submitted GFS yearbooks for the years 2013 to 2016 to the IMF. These are positive steps that need to be supported by setting transparency standards for PFM reporting, among them inclusion in the PFM Strategy of specific indicators to measure improvements in fiscal transparency. Doing so will set the basis for improving the transparency and effectiveness of the institutions engaged in the PFM process. In the medium to long term, this commitment will improve the country’s fiscal credibility, and enhance market perceptions of its fiscal track record. 2.2.2  Budget Executon and Control Process The OBL provides a solid foundation for commitment controls, though it has some shortcomings. In accordance with the OBL, budget users are not allowed to commit to expenditures in excess of their budget appropriation for the year. However, the OBL does allow commitments beyond the annual appropriation if they are donor-funded, or from special-revenue accounts (SRAs)—own- source revenues—of the budget users. This presents a challenge for commitment controls when supplementary budgets are passed and allocations reduced, as the control is directed to targets that are not credible, which can cause arrears to proliferate. For example, between 2013 and 2016, own- source tax and non-tax revenues from SRAs were underrealized by an average of 20 and 30 percent, respectively, while capital revenues were underrealized by 30 percent. This overcommitment could be avoided by strengthening the forecasting process and contrasting budget users’ estimates to their Chapter 2: The State of Public Financial Management 39 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 historical realization. The OBL is now being revised; and the update should address this issue, as discussed below. Previous commitment controls have failed to halt the accumulation of arrears. The present measures— such as refusing payment of inadequately documented liabilities, and absorbing the cost of multiyear commitments within the authorized budgets—should have provided incentives to ensure compliance. In reality, not all budgetary institutions respect the procedures. Invoices are not being presented to the treasury on time, especially when budget institutions know that funds are not available. In addition, while budget entities should not commit to expenditures without an approved budget, obligations can easily be moved to subsequent years if funds for the current period are inadequate. Moreover, the requirement to report larger commitments only after contract signature does not provide sufficient control over commitments, and should be re-examined. Finally, budget users are currently only obliged to report contracts exceeding €5,000, which only represents part of their commitments.24 The treasury system has not been particularly effective in controlling and monitoring commitments. It currently is unable, for instance, to record the date when a budget entity receives an invoice. Thus, there is no comprehensive, automated list that can allow the MOF to monitor payments coming due or in arrears. Furthermore, budget users have to estimate their cash requirements, broken down into monthly financial plans. However, though the cash requirement represents a ceiling on payments the budget users can make, they are not a ceiling on commitments because users can still commit to payments in future periods. Finally, while the present treasury commitment module includes contracts and goods and services, it excludes social assistance, pensions, and other acquired rights. Long-overdue upgrades to the treasury system are needed to control commitments and prevent arrears accumulation. Controls are needed within the treasury IT system to record both the initial incurrence of obligations and the invoice due date. The revised reporting system could then report on all stages of the payment cycle (expenditure commitments, invoices received and verified, accounts payable, invoices, and arrears paid). The monitoring systems should capture the size, maturity, and composition of the stock of arrears—information that can be used to identify their causes and prioritize their clearance. The system for monitoring multiyear commitments also needs to be built up because they impose additional responsibilities. Such commitments should only be incurred after obtaining formal MOF approval and being registered in the treasury system. Budget institutions should give the MOF full details of multiyear contracts, including anticipated commitments and payments for each year, which would also be incorporated into the MTBF. The measures discussed would also address weaknesses in accounting and government financial reporting. These not only have operational impact, they also limit the ability of the government to improve its financial reporting to align with international standards. Without a comprehensive centralized MOF record of financial liabilities, in-year budget reports only record expenditures at the time of payment, without giving any information on the magnitude of outstanding commitments. The Law of Accounting does require budgets and budget users to record financial liabilities and to report these in the annual accounts submitted to MOF. But though the law requires an annual statement of central government revenues and expenditures, it does not require information about financial assets 24 For example, one of the largest creditors identified during the May 2017 arrears self-reporting exercise was North Macedonia’s electricity retailer. A sizeable proporation of the debtors were small public entities with contracted amounts smaller than €5,000. 40 Chapter 2: The State of Public Financial Management SOWING THE SEEDS OF A SUSTAINABLE FUTURE and liabilities. This is an essential reporting requirement as the government moves to extend the information and data available in its audited annual financial statements. The PFM strategy outlines a comprehensive series of activities to address the weaknesses inherent in the current system, especially the design of the new integrated financial management information system (IFMIS). The current treasury cash ledger system works well, but the current data sources for fiscal reporting are fragmented. This presents the government with an opportunity to widen the scope of budget execution monitoring and fiscal reporting. It is important that the new system allows the MoF to monitor and report all stages of the expenditure cycle from procurement to payment at least monthly, to track budget users and obtain a panoramic view. The PFM reform strategy prioritizes new modules for registering commitments by modifying the existing e-commitments module in a way that allows budget beneficiaries to input data related to contracts and invoices via web-based applications. Other priorities are modules for invoices and financial plans, modules for accounts receivable and liquidity forecasting, fixed assets registration, and introduction of a general ledger. 2.2.3  Evaluation, Reporting and Monitoring of Government Liabilities Effectiveness of Controls to Manage Financial Liabilities - An Assessment of Arrears Nonpayment is a complex and recurrent problem in North Macedonia. In 2011–12, liquidity shortages allowed arrears to accumulate equivalent to 1.2 percent of GDP,25 mainly arising from overdue VAT refunds and unpaid goods and services. The government addressed these arrears through a process which involved preparation of an inventory of arrears with an agreed timeline and explicit clearance plan. This was achieved through a supplementary budget in 2013. At the time, the government introduced a commitment control module in the treasury system for recording, reporting, and validating multi-year commitments, and strengthened procedures for reporting liabilities on concluded contracts and Figure 2.9. Self-Reported Unpaid Expenditure Obligations, May 2017 their validation by internal auditors. Percent Despite these short-term measures and issuance 21 of a circular reminding budget users of the 12 statutory penalties associated with breaches of the requirement to report commitments, arrears continue to be a problem in North Macedonia. At 27 year-end 2016 arrears owed by public health care 26 institutions amounted to 0.5 percent of GDP.26 Furthermore, LGs had substantial but unreported arrears related to infrastructure projects. After 14 a self-reporting exercise conducted in early JJCentral government (CG) JJPublic health providers JJPublic enterprises (under CG) JJPublic enterprises (under the LGs) 2017, the stock of unpaid obligations has been JJLocal governments (LGs) estimated at €360 million, or 3.5 percent of GDP,27 Source: MOF. 25 The amount presented at that time covered only the central government, without the health sector, LGs, and SOEs. 26 The Health Insurance Fund is the only public institution that has consistantly publishing monthly reports of arrears. 27 The Law on Financial Discipline. Official Gazette of the Republic of North Macedonia. No. 187/13 and 201/14.defines arrears as a subset of payables/creditors unpaid for 60 days or more beyond any specified due date (regardless of any contractual grace period). If no due date is specified, arrears are defined as payables/creditors unpaid for 60 days or more after the date of the invoice or contract. Chapter 2: The State of Public Financial Management 41 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 though actual arrears are likely to be smaller due to inaccurate self-reporting and mutual debt between budget users and SOEs. Of that total, central budget users owed 0.9 percent of GDP, public health providers 0.5 percent, CG SOEs 0.9 percent, LGs 0.7 percent, and LG SOEs 0.4 percent (Figure 2.9). The build-up of unpaid obligations has been attributed to factors related to underlying PFM problems, among them formulation of unrealistic budgets and ad hoc revisions throughout the year, inadequate fiscal reporting and control systems, mismatches between revenue flows and LG responsibilities, and weaknesses in SOE management. Without an appropriate accounting for arrears, the fiscal picture presented by the reported budget deficit of 2.7 percent of GDP and debt estimated at 47.5 percent can be misleading. The partial operationalization of reforms has allowed shortcomings to persist in areas that are important for controlling and preventing the emergence of arrears. Furthermore, the MOF lacks a robust system for controlling data on expenditure arrears. Resolving the country’s arrears is central to the execution and transparency of fiscal policy, the quality of medium-term budget planning, and the functioning of public-private relations. Many of the activities outlined in the PFM Strategy could address the recording, monitoring, and control of payment obligations. Most important will be the upgrading of the treasury system already discussed which, if successful, would provide reports at all stages of the payment cycle. The web- based IFMIS system should also enable budget users to enter contracts and invoices directly into the system. In time, the system should allow the MOF to prepare a comprehensive, centralized record of financial liabilities. The PFM Strategy also pledges creation of a register of budget users who failed to promptly report arrears, which by 2019 should be published on the MOF website every quarter. In addition to commitment controls, the government needs the institutional reforms outlined above, to help improve the accuracy of budget forecasts, and the realism of budget decisions and allocations. As the new IFMIS will take time to become operational,28 an arrears clearance strategy is a crucial step for preventing future arrears accumulation. A credible strategy for orderly and transparent arrears- clearance is a vital government commitment that would offer some reassurance to suppliers that the Government is serious about addressing this issue. Contingent Liabilities and Other Fiscal Risks Fiscal risks can arise from adverse macroeconomic conditions, the financial positions of subnational governments or SOEs, and contingent liabilities from the programs and activities of the CG and extrabudgetary units. They can also arise from other implicit and external risks, such as market failures and natural disasters. Extrabudgetary units can have a significant impact on the central government’s fiscal sustainability. The extent to which the government’s financial reports reflect its operations reduces fiscal risk. Although the budgets of extrabudgetary units may be subject to approval by the legislature, because they represent independent bodies, they have discretion over the volume and composition of their spending, and the timeliness of information may not match the regular budget calendar because the calendar of audited reports of their previous activities differs from the regular budget cycle. 28 After the design phase implementation of the modular system is phased and planned to be fully operational by 2021. 42 Chapter 2: The State of Public Financial Management SOWING THE SEEDS OF A SUSTAINABLE FUTURE Fiscal Risks Arising from SOEs The fiscal risks related to SOEs have serious implications for the general government budget. The indebtedness of SOEs has risen consistently in recent years, from 2.5 percent of GDP in 2008 to an estimated 8.5 percent in 2017. SOEs are also a significant source of arrears. In May 2017, the arrears of CG SOEs represented 0.9 percent of GDP and the arrears of LG SOEs accounted for another 0.4 percent. This much SOE indebtedness and arrears constitutes a sizable fiscal risk to the central government, as many SOEs are already loss-making and others raise concerns, according to audited financial statements performed from the State Audit Office.29 Given the fiscal risk, the PFM Strategy should make improving the monitoring of the financial position of SOEs a priority. SOEs are now required to provide quarterly reports (balance sheet and financial statement) to the MoF, which is then required to inform the government about the financial position of the SOEs. However, the role of the MoF is limited to this operation; it does not undertake any oversight of the financial positions of individual SOEs. The oversight role is fragmented among the different line ministries that oversee SOEs There is currently no consolidated report or institution that oversees all of them; transparency could be improved by publishing the financial reports of all SOEs in accordance with International Financial Reporting Standards (IFRS), and a consolidated report for all SOEs addressing all the pertinent financial aspects of their operations, including capital structure, financial results, and any SOE losses that have an impact on the general government budget. Fiscal Risks Arising from Local Governments North Macedonia’s municipalities do not have enough revenue to finance all their basic functions, which leads to their accumulating arrears, which constitutes a fiscal risk. In the May 2017 survey, LGs reported arrears equivalent to 1.1 percent of GDP—about a third of all public sector arrears. Most relate to municipal budgets (0.7 percent of GDP), the rest (0.4 percent to their SOEs. There is anecdotal evidence for two sources of problems: (1) the implementation of infrastructure projects for which funding is fragmented and inadequate,30 and (2) the financing of education from CG block transfers, which it appears are often insufficient (for a detailed analysis of LG financing, see chapter 3). As a result of the financial pressures on LGs, the number with blocked accounts due to court procedures has been rising—about 20 currently have blocked accounts. The significant source of fiscal risk that the LG’s present should be addressed in the PFM strategy. Currently all LGs must submit to the MoF quarterly reports on the execution of their budgets. On a separate form they also summarize their outstanding liabilities. Furthermore, as all LGs must use the single treasury account, the MoF also has live data on their budget execution. All these measures give the MoF a good, up-to-date understanding of LG financial positions. The Law on Local Self Governance gives the MoF some authority to intervene when LGs get into financial difficulties, but this option has never been exercised. The MoF should take on a more proactive role in monitoring LGs by giving formal notice to those in financial difficulty and requiring that they take action to rectify the situation. 29 For example, the Public Enterprise for State Roads has its 2016 financial statements qualified by the auditor for its policy of capitalizing foreign exchange losses on payment to contractors, which if accounted properly would lead to the company incurring losses for 2016. 30 There are multiple sources of funding from CG agencies that provide capital grants, among them the Youth and Sports Agency, the Ministry of Environment, the Ministry of Agriculture, the Ministry of Local Self-Government, and the Protection and Rescue Directorate. According to one recent EU report, municipalities can apply for capital transfers through 18 different programs. Chapter 2: The State of Public Financial Management 43 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 A report summarizing current financial information for each LG and ranking the soundness of its finances could be considered. PPPs and Concessions Public finances in North Macedonia are exposed to numerous other fiscal risks that the PFM Strategy does not address. In addition to the build-up of SOE and LG debt and arrears, the government has entered into a number of public-private partnerships (PPPs), among them two large ones for airports, and LGs are also entering into PPPs. However, there is s no single consolidated list of the risks and obligations PPPs entail. Furthermore, every year North Macedonia is confronted by about 12,000 orders to enforce unfulfilled contractual obligations—largely payments for goods and services already provided—which may have adverse effects on the budget. Despite the significant potential burdens of these risks to the public purse, the PFM Strategy does not address them. There is currently no centralized institution to manage fiscal risks. Some of these responsibilities, such as management of state guarantees, fall under the MoF’s purview but in general they are scattered across different ministries and agencies. Thus there is no complete picture of all the fiscal risks, and no general mitigation strategies to deal with them, though both would be important for the MTFS and for policymaking. A framework for comprehensive fiscal reporting on risk management should be put in place as part of an institutional framework for managing fiscal risk analysis, with the MoF as the overseer. The government could also consider developing and publishing a comprehensive statement on fiscal risks as part of the budget declaration that would quantify risks and lay out mitigation strategies. 2.2.4  Public Sector Internal Financial Control and Audit Institutional arrangements for the effective implementation of financial management control regulations are minimal. This report addresses a number of significant control weaknesses, including as the lack of effective control of controls, the lack of effective procedures for managing public investment, and failures in the budget formulation process. The principals of distributed budgets and delegated decision-making—a central aspect of the PIFC law—have been slow to develop. There is limited delegation below the level of ministers and state secretaries, which undermines the effectiveness of the internal control function. Furthermore, full implementation of the regulations themselves is too complex for many smaller budgetary organizations. In practice, treasury controls will continue to function as an important safeguard, ensuring that spending complies with all the laws and regulations that apply. The decentralized model of internal audit is also making slow but steady progress. Internal audit units have been established in all first-level budget institutions. While the PIFC law and the related manuals comply with international standards, both the quality of auditors’ work and management responses to their recommendations need explicit attention. According to the Central Harmonization Unit (CHU) 2016 report, only 36.3 percent of all internal audit recommendations were adopted. This is explained in part by a lack of management action to address the irregularities presented by the internal audit recommendations, but also due to the fact that some recommendations were not clearly articulated or were simply too general to be effectively enacted. 44 Chapter 2: The State of Public Financial Management SOWING THE SEEDS OF A SUSTAINABLE FUTURE 2.3  Reform Options Short Term 1. Recognizing that payment arrears are a pervasive problem, draft a strategy to clear arrears that is orderly, transparent, and credible. Medium Term 2. Reinforce the treasury information system so that it can •• Record the date of receipt of an invoice and its due date). •• Monitor and report monthly on all accounts payable by age, based on due dates, amounts, and specific creditor information. •• Record commitments before the contracts are signed. 3. Gradually align accounting and financial reporting with international standards. An important first step would be to require budget users to present comprehensive and timely information about their financial assets and liabilities. 4. Centralize and strengthen macrofiscal forecast capacity by tasking the MoF Macroeconomic Department with preparation of all macroeconomic and macrofiscal forecasts. 5. Strengthen the link between government priorities and sectoral strategies and investment decisions. This should include putting in place procedures for prioritizing projects that reflect overall policy objectives, requirements for projects’ economic appraisal, costing, selection, and M&E of their social and economic returns over the project life cycle. 6. Draft a comprehensive MTBF, linked to the overarching fiscal strategy. The MTBF can become the key document establishing priorities and allocation of resources to different budget users in multiyear expenditure planning. 7. Reinforce assessment and monitoring of fiscal risks by drafting a consolidated fiscal risk management statement to be included with the fiscal strategy and budget documentation, detailing the type, size and, where possible, likelihood of specific fiscal risks. Additionally: •• Broaden revision of the OBL to include the necessary provisions to facilitate capacity development for fiscal risk management and SOE oversight and make public consolidated reports on extra-budgetary spending units. •• Consolidate oversight of fiscal risks within a single body, either in the MoF or at cabinet level. 8. Improve the transparency of SOEs by publishing the financial reports of all SOEs in accordance with IFRS. The MOF should also consider issuing a summary financial report for all SOEs that addresses all relevant financial aspects of their operations, their capital structure, and their financial results. 9. Better monitor LG financial performance. This could be done by enforcing the MoF’s role in monitoring LGs, which includes formally notifying those in financial difficulty and requiring concrete actions to rectify the situation. Chapter 2: The State of Public Financial Management 45 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Chapter 3: Fiscal Decentralization and Service Delivery Although North Macedonia has made progress in many aspects of intergovernmental finance, its 81 local governments still find it difficult to ensure adequate and efficient coverage of social services. Municipalities are responsible for many public services including basic education, but their expenditures as a share of GDP are a modest 5 percent of so. The practice of delegated functions has affected incentives for LGs to spend their own resources on services; ambiguous spending assignments have undermined accountability; and because there is not yet an effective methodology to calculate spending needs, some communities are underserved. Municipal revenues fall short, and LGs fail to fully use the little revenue autonomy they have. The lack of revenue autonomy leads to vertical imbalance and transfer dependence, which detracts from spending efficiency and the accountability of municipal governments. Disparities in the quality of basic services continue. While the VAT transfer performs some degree of equalization between municipalities, its role could be enhanced. However, the current disparities cannot be addressed solely by transfers and enhanced revenue autonomy. Balanced actions are needed across several dimensions, such as building LG capacity to absorb and use funds and cooperation for virtual consolidation of municipalities. Finally, municipal debt remains a problem, not because of excess borrowing—in fact borrowing has been limited—but because of a lack of discipline with regard to budget arrears, some of which have been carried over since decentralization began. 3.1  Sector Context Advancing fiscal decentralization to strengthen service delivery is a Government policy priority. Among the challenges for municipalities in North Macedonia in providing public services are underfunded social services, especially education, lack of clarity on expenditure assignments and the feature of delegated functions, horizontal disparities in the resources available, little tax revenue autonomy, the fragmentation of capital grants, and accumulating budget arrears. The current Government identified many such issues in its 2017–20 program31 and several proposals for reform have been put forward. However, as evidenced by the 2018 budget, most of the proposals involve increasing LG reliance on transfers from the central government. At best, more CG funding for municipalities is likely to be only part of the solution. Simply providing LGs with additional transfer funds might help with coverage of some services but will likely do little to make spending more efficient, heighten service quality, stimulate LG efforts to increase tax collection and accountability, or help reduce horizontal disparities in access to basic services. In terms of total revenues per capita, the best-funded municipality has six times as much revenue as the least funded (Figure 3.1), largely because of how intergovernmental transfers are currently designed: six of the 10 richest municipalities are among those receiving the highest transfers (Figure 3.2). 31 Source: http://vlada.mk/sites/default/files/programa/2017-2020/Programa_Vlada_2017-2020_MKD.pdf. 46 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 3.1. Total Revenues Per Capita, MKD Municipalities, 2016 In thousands 45 40 35 30 25 20 15 10 5 0 Cesinovo I Oblesevo City of Skopje Radovis Cair Kocani Debar Kumanovo Zelenikovo Staro Nagoricane Vevcani Vasilevo Aracinovo Lipkovo Lozovo Krusevo Prilep Karbinci Demir Kapija Karpos Centar Bogovinje Plasnica Bosilevo Valandovo Tearce Brvenica Saraj Suto Orizari Gorce Petrov Vrapciste Studenicani Jegunovce Butel Gazi Baba Zelino Zrnovci Novo Selo Kisela Voda Aerodrom Gostivar Rankovce Struga Vinica Delcevo Bogdanci Veles Demir Hisar Mogila Krivogastani Bitola Gradsko Cucer I Sandevo Konce Kratovo Rosoman Sveti Nikole Resen Tetovo Dolneni Kriva Palanka Caska Kavadarci Pehcevo Berovo Ohrid Mavrovo I Rostuse Novaci Strumica Debarca Negotino Probistip Stip Centar Zupa Gevgelija Sopiste Petrovec Dojran Kicevo Ilinden Makedonski Brod Makedonska Kamenica Source: Ministry of Finance and World Bank staff calculations. Figure 3.2. Total Transfers Per Capita, MKD Municipalities, 2016 In thousands 30 25 20 15 10 5 0 Cesinovo I Oblesevo City of Skopje Radovis Cair Studenicani Kocani Kumanovo Debar Zelenikovo Karpos Vevcani Staro Nagoricane Centar Vasilevo Aracinovo Lipkovo Krusevo Prilep Demir Kapija Karbinci Bogovinje Lozovo Aerodrom Gorce Petrov Kisela Voda Plasnica Tearce Bosilevo Gazi Baba Brvenica Suto Orizari Valandovo Saraj Butel Cucer I Sandevo Vrapciste Jegunovce Zelino Gostivar Bitola Ohrid Novo Selo Delcevo Struga Ilinden Veles Bogdanci Zrnovci Probistip Vinica Sveti Nikole Stip Strumica Rankovce Kavadarci Gevgelija Pehcevo Kriva Palanka Sopiste Resen Rosoman Novaci Krivogastani Demir Hisar Berovo Tetovo Mogila Kratovo Negotino Gradsko Makedonska Kamenica Dojran Petrovec Mavrovo I Rostuse Konce Caska Debarca Dolneni Centar Zupa Kicevo Makedonski Brod Source: MOF and World Bank staff calculations. The situation calls for a comprehensive review of fiscal decentralization and a reform strategy outlining sustainable solutions for the medium term. Rather than discussing what would be most effective, the current debate is monopolized by the topic of transfer sizes: LGs claim that the transfers are insufficient, and the central government asserts that it cannot afford to increase them. Successfully assessing how much to transfer depends on a careful assessment of the entire system of intergovernmental finance:(1) There needs to be a fresh look at how spending responsibilities are assigned at the local level. Is the distinction between own and delegated responsibilities still a valid and useful approach? Does calling some responsibilities “delegated” oblige the central government to use block transfers? Does designating some responsibilities “delegated” justify LG refusal to spend any of their own funds in such areas as education? (2) Beyond the longstanding complaint that transfers do not cover the historical costs of providing services, little has been done to identify methodologies for the costs associated with spending responsibilities. For example, what should be the standard expenditure per student in primary education? How much funding can North Macedonia afford, given education’s central role in the attainment of its long-term development goals primary school? (3) On the side of revenue assignments, municipalities still do not have sufficient revenue autonomy—yet they fail to fully use the revenue autonomy they have. (4) In the transfer system, horizontal disparities in access to basic services call for more equitable distribution of the general (VAT) transfer. Other transfers, including the block grants and capital grants, need to be simplified Chapter 3: Fiscal Decentralization and Service Delivery 47 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 and made more transparent and effective. (5) Finally, municipal debt continues to be a problem area because of arrears. Whether North Macedonia can afford to spend more on services that are currently decentralized is a serious issue. Although subnational public spending could be made more efficient through deep reform of the intergovernmental finance system, addressing current problems with delivering basic services will require increased funding, whether from giving LGs more revenue autonomy or increasing central government transfers. Whether the necessary fiscal space is created by cutting CG spending or raising more taxes is ultimately a political decision based on national priorities. In 2016, the share of general government spending as a percentage of GDP in North Macedonia was 33 percent—compared to 42.3 percent in Bosnia and Herzegovina, 43 percent in Serbia, 45.1 percent in Croatia, and 45.6 percent in Montenegro. North Macedonia also appears to have scope to increase LG spending as a percent of GDP (now 5 percent) and general government spending (16.1 percent), both of which are below those in the rest of the region, even though in North Macedonia, education wages have been decentralized. In 2016, LG spending as a percentage of GDP was 5.4 percent in Montenegro, 6 percent in Serbia, and 6.9 percent in Bosnia and Herzegovina. 3.2  Institutional Issues and the Vertical Structure of Government Though it decentralized late, with its early reforms between 2005 and 2007 North Macedonia made significant strides in the design of fiscal decentralization. However, the process has been stalled for the last decade. Inspired by the European Charter on Local Self Governments and the Ohrid Agreement, the country’s 2002 Law on Local Self Government defines the composition, organic structure, and devolution of competencies to municipalities, which was to occur in two phases. The Laws on Territorial Organization, Financing of the Units of Local Self Government, and Property Taxes, all from 2004, cover the resources at the disposal of municipalities. The second phase of decentralization started in 2007 with full delegation of authority to LGs for social sectors. However, with the global crisis in 2008, the reform and modernization process essentially came to a halt. Nevertheless, the current legal framework, though in need of updating, is quite sufficient to provide a basis for a new round of reform. A positive aspect of the decentralization process so far has been the continuous dialogue between the two levels of government, facilitated over the years by the Association of Local Self Government Units (ZELS), and with several CG ministries involved. The Ministry of Local Self Government (MoLSG) is responsible for coordinating and monitoring the system generally and the MoF is responsible for revenue assignment and the transfer system. Over the years, several working groups with both CG and ZELS representation have been charged with planning reforms. While municipalities report to CG ministries on budget execution, the latter could make better use of the information. A reason they do not is that the specific roles of the ministries are still not well defined, especially in terms of promoting more efficient local governments. Both the MoLSG and MoF could use the data they receive from municipal operations as a clear reference for LG performance and to promote healthy competition between them. However, today, monitoring mainly consists of inputs from local governments, such as the quarterly financial reports submitted to the MoF.32 The 32 The State Audit Office also performs annual audits of all municipal financial statements. 48 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE same applies to line ministries, such as the Ministry of Education and Science (MOES), with regard to use of block grants. As the decentralization process matures, CG ministries must find new roles in the analysis and promotion of better LG outcomes. This is a key part of an overall strategy for making subnational public spending more efficient and enhancing the quality of basic services. The fragmentation and lack of modernization of the LG budgeting process is a serious impediment to more efficient spending and the quality of public services. LGs in North Macedonia still operate through a series of separate budgetary accounts: (1) the principal budget; (2) the donations budget; (3) the block grants budget; (4) the budget for self-financing activities; and, since 2008, (5) the borrowed funds budget. The result is fragmented decision-making and difficulties in setting clear priorities. For example, because the block grant budget funds education, LGs resist spending money on it from the principal budget (their own general funds). In addition, the capital budget is currently highly fragmented, and is generally deemed to be an inefficient system for the financing of capital projects. These core PFM issues are addressed in full in the preceding chapter of this report, but they have a significant impact on how the intergovernmental finance system operates. Local public spending would also be made more efficient if budgets were planned using a medium- term framework and if there was more effective budget execution, procurement, and ex-post program evaluation. Issues such as a strict adherence to the budget calendar to inform the magnitude of revenue sharing and other transfers for next year are of critical importance to allow for the better planning of local budgets. Similarly, there is also a need to re-introduce the requirement for automatic supplementary budgets when actual revenues underperform below a certain threshold to avoid overspending or arrears because of too optimistic budgeting. The MoF could also compare quarterly data on budget execution with municipal budget plans and introduce an early warning system for revenue shortfalls and greater than planned or extraordinary expenditures: eliminating the possibility of committing financial resources for procured services that have not been budgeted would help control the accumulation of arrears. Finally, ex-post program evaluation and creation of a culture of performance could help make municipal spending significantly more efficient. Local expenditure efficiency is still impacted by the suboptimal size of some municipalities. Among the early reforms of the intergovernmental finance system, some LGs were consolidated, but some are still too small to be efficient. The number of LGs was first reduced from 124 to 85, and then to Table 3.1. Distribution of LGs by Population 80, 10 of which constitute Skopje.33 As Table 3.1 Population Number of LGs shows, 31 of 71 LGs have a population of less < 5,000 34 16 than 10,000, the approximate minimum size 5,000-10,000 15 necessary to exploit economies of scale in the 10,000-20,000 17 delivery of most local public services (Martinez- 20,000-50,000 12 Vazquez and Lago-Peñas 2013). This problem is 50,000-100,000 9 somewhat masked by the fact that municipalities over 100,000 2 in North Macedonia (excluding Skopje) average Total 71 about 22,000 citizens, which compares favorably Source: Statistics Office. internationally. Efficiency problems are most 33 In the 2002 Census—the most recent available—the City of Skopje represented a fourth of the entire North Macedonian population. Except for boundaries of Kicevo, which amalgamated five neighboring municipalities in 2014 (MFR II Report, 2017), the number of municipalities has been stable. 34 In total there are 81 local governments but this analysis excludes the 10 city-municipalities that constitute the City of Skopje. Chapter 3: Fiscal Decentralization and Service Delivery 49 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 likely concentrated in the 16 municipalities with less than 5,000 residents, where 17.5 percent of the population resides.35 North Macedonia’s intergovernmental finance system does not adequately recognize the vast heterogeneity of municipalities, which also detracts from overall efficiency. There are large differences between municipalities in terms of size, economic features, urbanization levels, living standards, and fiscal and administrative capacity—the last having the most impact on overall efficiency because it affects the ability to perform LG responsibilities.36 Assignment of the same functions to all municipalities, regardless of their capacity, carries a risk of substandard service provision. Short of another round of forced mergers, or the introduction of asymmetric assignment of functional responsibilities depending on size, the easiest solution to the inefficiency of too small municipalities is closer inter-municipal cooperation. Cooperative arrangements among smaller municipalities and service agreements between smaller and larger municipalities have helped mitigate these problems,37 but there have been too few such agreements.38 The Law on Inter-Municipal Cooperation, adopted in 2009, needs to be supported by changes in the structure of incentives in the financing system. For example, delegated functions financed with block grants could incentivize service-purchase agreements between smaller and large municipalities, or cooperation among smaller municipalities could be promoted by better conditions in the distribution of capital grants to inter-municipal projects. 3.3  Expenditure Assignments: Implications for Efficiency and Equity Expenditure assignments in North Macedonia compare well with those of other decentralized countries at the same level of economic development. Although economic principles—e.g., the presence of externalities, size of the benefit area, economies of scale in service delivery, and the degree of redistribution the service implies—inform their assignment, in reality, there is no “best” expenditure assignment. Factors such as country traditions and political preferences can also have an important role. The most important principles, once expenditure assignments are established, are that they should be transparent and unambiguous (what level of government is responsible for what), and be stable over time; in North Macedonia they may have been stable, but they have not been unambiguous. Currently, municipalities are in charge of spending on pre-school, primary, and secondary education; water and sewerage; streets and local roads; public transportation; parks maintenance; fire protection; culture; and social assistance, including care for the elderly. Aside from education and teacher wages in particular, all these are typically considered local functions in decentralized countries. Social expenditure assignments tend to vary more internationally (Table 3.2). Keeping public health services centralized—especially secondary facilities—is also common internationally. Though LGs are responsible for many public services, including wages in primary and secondary education, their spending as a share of GDP is modest, about 5 percent. Naturally, LG spending as a share of GDP for any country reflects decentralized functional responsibilities and the services actually delivered for each functional responsibility. Nevertheless, local spending as a share of GDP 35 These tend to be rural, with less land area, fewer settlements, and less revenue per capita from own taxes and fees. 36 The Law on Financing of the Units of the Local Self Government requires municipalities to have a department of budget and accounting with at least two financial officers, a chief accountant, an internal audit section, and a three-person tax department. It is not clear whether it is enforced. 37 Areas of cooperation have included urban planning, communal services, infrastructure, and financial management. 38 The actual funding of many of these services with block grants, which only municipalities with physical facilities get, has been a barrier to greater cooperation and service agreements. 50 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE is generally taken to represent the degree of a country’s decentralization. Internationally, this ratio ranges from less than 1 to over 60 percent of GDP; and the EU average in 2016 was 15.6 percent.39 In North Macedonia (Table 3.3), total LG spending was 4.9 percent of GDP in 2016, down just 0.1 percent from 5.0 percent in 2009, having peaked in 2012 at 6.2 percent. Elsewhere in the region in 2016 LG spending in Bosnia and Herzegovina was 6.9 percent, Serbia 6 percent, and Montenegro 5.4 percent.40 While North Macedonia’s degree of decentralization is low by EU standards, it is in line with its WB6 neighbors, though in those countries teachers’ salaries are not paid through municipal budgets. Table 3.2. Subnational Social Spending Responsibilities, Selected ECA Countries Preschool Education Primary Education Secondary Education Primary Health Secondary Health Upkeep Wages Upkeep Wages Upkeep Wages Upkeep Wages Upkeep Wages Albania xx xx xx xx xx Moldova xx xx xx xx xx xx Romania xx xx xx xx xx xx xx xx xx xx Bulgaria xx xx xx xx xx xx North Macedonia xx xx xx xx xx xx Kosovo xx xx xx xx xx xx xx xx Slovenia xx xx xx xx Serbia xx xx xx xx xx Croatia xx xx xx xx Ukraine xx xx xx xx xx xx xx xx xx xx Source: Network of Associations of Local Authorities of South-East Europe (NALAS). Table 3.3. Local Government Spending, 2009–16 Percent of GDP 2009 2010 2011 2012 2013 2014 2015 2016 Total spending 5.0 5.4 5.6 6.2 5.5 5.2 5.2 4.9 Wage bill 2.5 2.7 2.7 2.9 2.7 2.6 2.6 2.4 Goods and services 1.4 1.4 1.3 1.5 1.4 1.3 1.3 1.2 Other recurrent 0.2 0.2 0.2 0.3 0.3 0.2 0.2 0.2 (e.g., social assistance) Capital spending 1.0 1.1 1.3 1.6 1.2 1.1 1.1 1.1 Source: MOF. Wages and salaries dominate LG spending, but it is not clear that this means resources are allocated inefficiently. Nearly half of total LG spending goes to wages and salaries (2.6 percent of GDP on average for 2012–16). The largest share of wages goes to education sector employees (about 80 percent of the total wage bill). In fact, without education the wage bill is one of the lowest in the region, at about 0.6 percent of GDP, compared to 0.8 percent in Albania, 1.2 percent in Croatia, and 1.3 percent in Serbia.41 The second largest spending item is goods and services, which accounts for about 25 percent of LG budgets. 39 This figure includes public spending all levels of government below the central government: states, regions, provinces etc. 40 See the World Bank South East Europe Municipal Finance Review (MFR) II Report, 2017 (forthcoming). 41 The Civil Servants Agency, a central government uni, is supposed to approve the personnel structure of municipalities but reportedly does not constrain local government discretion. Chapter 3: Fiscal Decentralization and Service Delivery 51 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 LG spending on infrastructure is minimal. Capital investment by LGs represented 18 percent of total local expenditures in 2016, or about 0.9 percent of GDP, on par with neighboring Serbia and Albania (1.1 percent) but lower than any ECA countries that have joined the EU, where it accounts for 2–4 percent. In North Macedonia LG capital spending accounts for about 25 percent of total general government investment, compared to about 40 percent in other ECA countries.42 This in part relates to the fragmented capital transfers system, and the limited subnational use of loans as an alternative source of financing for local infrastructure. From a functional standpoint education is by far the most important component of LG spending. In 2016, it represented 44.9 percent, far ahead of housing and communal amenities (such as public lighting and water services) at 25.3 percent and general public services at 16.1 percent. Relatively smaller items, such as social protection (6.8 percent), recreation and culture (3.7 percent), and public order and safety and economic affairs (1.4 percent each) made up the balance.43 LG spending statistics mask the lack of clarity in expenditure assignments in North Macedonia. The Law on Local Self Government of 2002 defines the functional responsibilities of the municipalities as44 (1) general competency to perform activities of local importance—those “not under the competency of the organs of the state administration” that can be considered exclusive responsibilities of local governments; (2) specific competencies assigned to municipalities to “be performed in accordance with the standards and procedures determined by law, which have been interpreted as concurrent or shared responsibilities”45; and (3) delegated competencies, entrusted to LGs for execution, and rarely to be financed by the central government, which in turn is supposed to keep local authorities accountable for the delivery of services, thus retaining subsidiary responsibility for the outcomes. However, it is unclear what obligations municipalities have for delegated responsibilities. For example, are they supposed to spend their own revenues beyond transferred funds in areas like education? When responsibilities are delegated, what functions are to be performed by line ministries, such as the MoES—should they oversee overall performance or conduct efficiency audits? The lack of clarity in defining expenditure assignments in North Macedonia affects the behavior of LGs and central ministries, and efficient operation of the intergovernmental finance system. A clear definition needs to make explicit which level of government is responsible for the three main features when any function is assigned: (1) regulation and setting standards for the service; (2) financing; and (3) actual delivery of the service. That has not been done in North Macedonia, and the ambiguity undermines the accountability of municipal governments to their citizens and makes it difficult to assess their performance. There is a need to re-examine the logic and practice of delegated functions in North Macedonia. As in other countries, the line between delegated and devolved public services is burred. The rationale for having delegated, rather than devolved, services is that the central government still has a strong interest in ensuring that these services are provided to some particular standard, for instance because 42 The latest available year for Albania, Bulgaria, Croatia, Hungary, Poland, Romania and Serbia. Source: IMF-GFS. 43 Despite the fact that health services are formally listed as a shared responsibility, Macedonian LGs spend nothing on health services. This is a general feature of expenditure assignments for countries in the region in contrast to countries elsewhere. 44 One complication is that functional responsibilities assigned to the municipalities are further regulated by sectoral laws. The City of Skopje has special legal provisions for sharing the competencies with its constituent municipalities, but as a whole Skopje is assigned the same package of functions as other municipalities. 45 The list of concurrent responsibilities (Art.22) covers urban and rural planning; protection of the environment and nature; local economic development; communal activities; culture; sports and recreation; social welfare and child protection; education; health care; civil protection; firefighting, and other activities assigned in separate legislation. 52 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE of equity or efficiency. Categorizing a service as “delegated” obligates the central government to, among other things, provide a minimum level of financing. North Macedonia does this through block grants. Delegated functions should also mean that the CG funding comes with strings attached, but in the case of North Macedonia, there are too few strings attached. Other than monitoring that the funds are spent on the specified sector, there is no follow-up by any CG unit on how efficiently these funds are spent and on differences in how the LGs perform. The categorization as delegated services in practice has also meant—at least as interpreted by the municipalities themselves—that all spending in those sectors should be financed by CG funds. Yet there is no real reason why that should be the case.46 The rationale of delegated functions is that the central government provides financing for some minimum level of service but LGs can and should contribute funds to increase the scope of the services. However, one issue for LGs is that the central government does not provide the actual “minimum” funds needed. Therefore, an effective methodology is needed to identify and allocate responsibility for adequate minimum funding. Are delegated functional assignments to be based on the client population or simply on the historical presence of physical facilities? A complication perhaps unique to North Macedonia is that in some cases the right of a LG to receive a block grant for a delegated function depends on the existence of a physical structure for delivering the service at the time of the functional decentralization.47 As a result, two de facto classes of municipalities were created. Rural and smaller municipalities that did not possess physical facilities do not receive the block grants. Therefore, the residents of these rural municipalities must in most cases forego the service or perhaps access it in a neighboring municipality. Rethinking the concept of “delegated” functions should mean considering the source of the need for it, which is the resident population to whom the services are addressed, and not the historical presence of facilities. This is the principle most countries follow. It is also possible for a LG to open up a physical facility to residents of other LGs with explicit arrangements regarding compensation for services rendered. There is clearly a need for simple, transparent Figure 3.3. Teachers and Students in Primary and Secondary Education, 2000/2001–2016/2017 methods to calculate the costs associated with In thousands In thousands functional responsibilities. In other words, how 300 20 do we know whether education and other basic services are not adequate? The dialogue between 250 15 central and local governments on potential 200 reforms of the system of intergovernmental 150 10 fiscal relations is almost exclusively focused on financing issues, such as how much PIT revenue 100 5 is to be shared or what pool of funds determines 50 the general VAT transfer. Yet none of the financing figures discussed is as yet based on a clear 0 1 3 5 7 9 11 3 5 70 0–0 2–0 4–0 006–0 008–0 2010– 2–1 4–1 6–1 200 200 200 2 2 201 201 201 methodology aimed at providing some measure ÂÂStudents ---- Teachers, rhs of expenditure needs associated with the current Source: Statistical Office data. expenditure assignments to LGs. The best current 46 For example, although responsibility for maintaining school buildings falls on LGs, financing for the delegated functions in education is mostly used for salaries, so school buildings tend to suffer from lack of maintenance. 47 The process of delegating responsibilities focused on transfer of facilities, their operation and maintenance, and the personnel needed to operate them. These included physical facilities for culture, social protection, and fire protection units. Chapter 3: Fiscal Decentralization and Service Delivery 53 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 example of an acceptable methodology estimating expenditure needs is the formulas the MoES uses for the block grants for elementary and secondary education. These are based on a spending standard per client adjusted for geographical differences in the costs of providing the service. The methodology could be easily extended to all other local spending responsibilities.48 The choice of a funding level that is affordable and desirable for a particular service should be informed by clear and explicit sectoral policy strategies that fit in the medium-term expenditure framework. Such decisions require careful analysis of current resources and likely future needs. For example, in primary and secondary education, since 2001 the number of students has been trending down, while the number of teachers has steadily risen, reaching what seems to be very low pupils-to-teacher ratios (for more discussion, see chapter 7). 3.4  Reinforcing the Local Revenue Base: Tax Assignments As is common in other systems, fiscal decentralization in North Macedonia has been asymmetric, with relatively little taxing power devolved to LGs. Own revenues from taxes and fees represent about a third of total LG revenues. Transfers account for most of the remaining two-thirds (Figure 3.4). Revenue assignments are defined in the Laws on Financing the Units of Local Self Government (Art. 4-7) and on the Property Tax, and subsequent legislation. The main items are property taxes (annual tax, tax on transfer of real estate, tax on inheritance and gifts); local fees (communal and administrative), and local charges (construction land charges, communal activity charges, spatial planning charges). The transfers, discussed in detail in the next section, include shared taxes (about 5–6 percent of total revenues) and other transfers (about 60 percent of total LG revenues). In North Macedonia, LGs and the central government share revenues from the VAT, PIT, and agriculture land leases. LGs get 4.5 percent of VAT collected; 3 percent of the PIT, and 10 percent of land lease proceeds (chapter 4 for a fuller discussion of taxation in North Macedonia). With its high transfer dependence, the country’s intergovernmental finance system is slightly worse than that of other countries in the region in terms of the share of own revenues in total local revenues (Figure 3.5).49 Furthermore, there are huge disparities between municipalities. While, on average, own revenues account for 36.6 percent of the total revenues available to LGs, in 36 municipalities they are less than 20 percent (Figure 3.6). International evidence suggests that giving LGs more revenue autonomy will help make local spending more efficient and can be achieved through better use of existing local taxes and fees and by expanding the number of local tax instruments. Greater revenue autonomy is generally associated with greater accountability of local officials to residents in the use of fiscal resources.50 Meanwhile, low revenue autonomy (large vertical imbalances and transfer dependence) as in North Macedonia is generally associated with a lack of LG spending discipline, less effort to collect taxes, and the expectation of central government bailouts of LGs. Another rationale for more revenue autonomy is the fact that municipalities in North Macedonia often do not have enough revenue to finance all their basic functions, which has led to arrears build-up and poor-quality public services. In a government stocktaking exercise in May 2017, LGs reported arrears equivalent to 1.2 percent of GDP. Thus it could 48 The computation of expenditure needs can be done separately by category of expenditures or for the composite of several local services. For example, it may make sense to group under “general public services” some functions that are unimportant in budgetary terms, and use the local population to estimate the number of clients for this composite category. 49 Local own revenues in North Macedonia are a bit inflated because the PIT share is treated as local tax revenue source, when in fact, like any other form of revenue sharing, it is simply another transfer. 50 As to how much revenue autonomy is desirable, a good general rule is that own tax revenues should provide the richest jurisdictions—in this case, Skopje— with close to 100 percent of their spending needs. 54 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE be highly beneficial to increase local revenue autonomy while building up local PFM capacities and institutions. The important question is how best to achieve that. Figure 3.4. Structure of LG Revenues, 2016 Figure 3.5. Local Government Revenues, WB6, 2015 Percent Percent 100 61.0 90 80 70 60 50 1.4 40 1.0 30 20 10 0 a n ska atia a o bia ani rat io Srp Cro oni egr Ser Alb ede ika a ced n ten HF o 36.6 BI ubl rth M M Rep No BIH JJOwn JJTransfers JJGrants JJShared JJOwn revenues JJShared taxes JJGeneral taxes JJBlocked transfers JJCapital transfers Source: MFR II Report and NALAS, 2017. Figure 3.6. Own Revenues as a Share of Total LG Revenues, 2016 Percent 70 60 50 40 30 20 10 0 Cesinovo I Oblesevo Staro Nagoricane Radovis Vevcani Zelenikovo Debar Cair Kocani Kumanovo City of Skopje Karbinci Lipkovo Karpos Lozovo Aracinovo Demir Kapija Krusevo Vasilevo Prilep Centar Centar Zupa Bosilevo Valandovo Konce Rankovce Bogovinje Studenicani Saraj Zrnovci Gradsko Dolneni Mogila Vrapciste Caska Krivogastani Kratovo Suto Orizari Brvenica Demir Hisar Berovo Pehcevo Plasnica Jegunovce Vinica Debarca Novo Selo Mavrovo I Rostuse Zelino Bogdanci Delcevo Tearce Resen Tetovo Makedonski Brod Struga Rosoman Kicevo Veles Gostivar Butel Negotino Kriva Palanka Kavadarci Sveti Nikole Dojran Gazi Baba Petrovec Novaci Strumica Bitola Stip Gevgelija Gorce Petrov Sopiste Probistip Cucer I Sandevo Ohrid Kisela Voda Aerodrom Makedonska Kamenica Ilinden JJOwn revenues ---- Average Source: Ministry of Finance and World Bank staff calculations. Although property tax collection has been Figure 3.7. Local Property Taxation, 2016 Percent of GDP y-axis label improving, it still has significant potential for 1.4 30 adding to LG revenues. In 2016, collection of local property taxes amounted to 0.5 percent of North 1.2 25 Macedonia’s GDP, 9.3 percent of total LG revenues, 1.0 20 and for more than 25 percent of municipal taxes. 0.8 However, property taxes as a percentage of GDP 0.6 15 are well below the regional average (Figure 3.7)— 10 0.4 it is less than a third of Montenegro’s and half 5 that in Serbia. The two main sources of property 0.2 tax revenues are the annual real estate tax 0 0 and the tax on the real property transfers. The Montenegro Serbia North Macedonia JJTotal property tax ÂÂAs a share of total local revenues, rhs annual property tax could bring in more revenue Source: NALAS, 2017. by increasing the maximum permitted tax rates, Chapter 3: Fiscal Decentralization and Service Delivery 55 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 better coverage of the real estate market, or both. The Law on Property Taxes allows municipalities to set the rate at 0.1 to 0.2 percent of a property’s “market” value. More revenue could by produced by updating property valuations and tightening actual collections, since municipalities themselves administer the property tax. However, the quality of local property cadasters varies, with those of smaller LGs having less current information. Addressing the issues in the medium term may require an asymmetric approach to cadaster building and property valuation. While large LGs could continue to exercise those functions, for smaller LGs, the cadaster updating and property valuation process could be carried out by a central agency, financed by partner LG contributions. Before the global recession, the transfer tax yielded considerably more revenues than the annual tax, but because the transfer tax tends to be distortionary for the housing market, it is less desirable to try to raise more revenue from this source.51 There is also some fiscal space for raising more local revenue from charges and communal fees.52 At 20.2 percent of total local revenues in 2016, local fees and charges are sizable but could probably produce even more local revenues. In the communal tax category, the main sources have been the development fee, the electricity fixed levy (public lighting tax), and the business signage fee. The municipalities set some fees and tariffs themselves but for communal and some other fees, the central government limits them to certain ranges. Raising more revenue from fees and charges will require updating tariffs and wider application of fees and charges among municipalities—keeping in mind that if they are too high, they may discourage economic activity, as Skopje’s development fee is said to have done. Another option for expanding local tax autonomy is to introduce an annual vehicle registration tax. Many countries levy an annual tax on motor vehicles associated with annual license renewal, in addition to the tax on immovable property. The rationale for this levy is the benefits to vehicle owners from the use of roads. To adequately price the consumption of government services and environmental externalities, the tax could vary by vehicle age, engine size, or weight. In administering this tax, valuation decisions could be assisted by the use of privately published lists of vehicle values by model and year.53 3.5  The Intergovernmental Transfer System Given the sizable vertical and horizontal fiscal imbalances, the transfer system has a powerful role in North Macedonia’s intergovernmental finance system. Own revenues finance just over a third of the total budgets of LGs (Figure 3.4). PIT revenue sharing, the VAT general transfer, the block grants, and the capital grants all help address the existing vertical imbalance. Transfers are also needed because there are horizontal fiscal imbalances produced by the disparate spending needs across jurisdictions and in the differing economic bases of municipalities—and therefore in their capacity to generate own revenues. CGs generally use “conditional” transfers, such as North Macedonia’s block 51 This is fundamentally a sales tax on real estate transactions, which explains the volatility of its collections over the business cycle. The tax base is the same fiscal valuation of property as for the annual property tax. The law allows municipalities to set a rate of 2 to 4 percent of “market” value, with most municipalities choosing 3 percent according to past information from ZELS. 52 These include signage fees, charges for utilizing pavements in front of business premises, advertisement fees, music festival and other entertainment fees, parking fees, passenger vehicle fees, use of public electricity fee, etc. 53 Apparently, the Customs Administration has a catalogue with valuations for practically all cars. Administration of this levy could be piggy-backed on the vehicle registration fee. 56 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE and capital grants, to stimulate local spending in areas of national priority so as to ensure minimum spending in critical social areas like education, or to address cross-jurisdictional environmental and other externalities. As currently structured, North Macedonia’s transfer system contains the right elements for a complete modern system of transfers, which should allow the central government to pursue all the desired policy objectives. However, some transfer instruments need to be revised and improved. The sharing of PIT revenues could be increased to reduce vertical imbalances. The general transfers sharing VAT proceeds can be easily converted into an equalization grant to make a greater contribution to closing fiscal disparities between municipalities. The block grants are used to achieve CG objectives in education and several social areas, but are not always transparent or effective. Finally, capital grants already support, though minimally, the development of much needed local infrastructure. Revenue sharing via the PIT is a useful, but likely insufficient, way to help correct the vertical fiscal imbalance. Although traditionally PIT revenue sharing has been treated as another source of own local tax revenue in North Macedonia, the correct interpretation is to treat it as a form of transfer.54 Formally, 3 percent of the national PIT collection is allocated to the taxpayer’s municipality.55 In general, if revenue-sharing is to be used to close the LG vertical fiscal gap, then sharing the PIT on a residence basis would be a preferred form of revenue-sharing,56 so the country’s sharing of PIT revenues rests on a sound theoretical base. However, it has been criticized for being too small, even trivial, given that it constituted less than 1 percent of municipal budgets in 2016. The fact that national PIT collection rates are low—less than 3 percent of GDP—contributes to its low importance to local revenues. ZELS, like LG associations in neighboring countries, has long demanded an increase in the rate of PIT revenue- sharing. This should be given consideration after careful evaluation of municipal spending needs, alternative ways to increase the adequacy of local revenues, and LG capacity to effectively absorb funds and deliver public services. Although North Macedonia has no explicit equalization grant, the VAT formula-driven revenue- sharing system may be thought of as such a grant. Since 2005, a share of VAT collected in the previous year has been allocated to the municipalities as an unconditional general-purpose grant. The amount of VAT shared with municipalities was originally 3 percent, but it was recently raised to 4.5 percent, and it may be raised further. The funds are distributed by formula. For the most recent year available, the distribution is as follows: each municipality (including the city of Skopje) gets a fixed sum of MKD 3 million. Of the remaining funds, 12 percent goes to the city of Skopje and its municipalities and 88 percent to the other municipalities. Of the 12 percent, 40 percent goes to the city of Skopje itself and 60 percent to its municipalities. Distribution of the 88 percent is weighting as follows: 65 percent for population, 27 percent for geographical area, and 8 percent for the number of settlements within the municipality. Typically, equalization grants are unconditional grants that are distributed taking into account the spending needs and fiscal capacity of subnational governments. The VAT-sharing grant meets all these criteria except for the fiscal capacity condition. 54 It is a transfer in the sense that LGs have no discretion whatsoever on the revenue yield. It all depends on CG decisions. Actually, the Law on Financing the Units of Local Self-Government correctly defines PIT revenue sharing as a transfer. 55 The sharing rate for the PIT from physical persons who are performing craft activities, registered in the LG territory in accordance with the Law on Craft Activity is 100 percent. 56 Using a residence rather than a place of work makes individuals pay income tax in the jurisdiction where they live, which may differ from -where they work. The presumption is that individuals tend to consume more local public services in their place of residence. Chapter 3: Fiscal Decentralization and Service Delivery 57 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 The VAT transfer performs some degree of equalization. This is because poorer rural municipalities with less capacity to collect revenue tend to have a lower population density—or more land and settlements per capita—which drives the distribution of 40 percent of the VAT transfer funds.57 The actual degree of equalization has changed over time, in part because the distribution formula changed. For example, in recent years, the relative weight given to the number of settlements in a municipality— which tends to benefit rural municipalities—declined from 13 to 8 percent, with the balance going to the population variable.58 Using 2016 data reflecting pre-VAT transfer revenues, rural municipalities have 13 percent less per capita revenue than their urban counterparts. After the VAT is transferred, the gap is reduced to 8 percent. Also, based on pre-VAT transfer 2016 data, Skopje’s municipalities have 50 percent more revenues per capita than other urban municipalities. This figure drops to 40 percent after the VAT transfers. The role of the VAT transfer as an equalization grant could be enhanced by making fiscal capacity an additional distribution criterion.59 As discussed earlier, municipalities vary widely in their ability to collect property taxes and development fees. Per capita collections are lower in rural than in urban municipalities, and Skopje’s municipalities have the highest collection rates in the country (in 2016 collections were 2,975 MKD in rural; 4,223 MKD in urban, and 10,415 MKD in Skopje LGs). These disparities limit how much the system of municipal finance can rely on increases in tax autonomy and revenue-sharing on a derivation basis (as in the case of the PIT), unless the equalization grant formula helps address the divergent fiscal capacities. All these arguments imply that fiscal capacity per capita should also be a criterion for distributing the VAT shares.60 Taking into account fiscal capacity in the VAT distribution formula may also lead to greater equalization among municipalities that have similar populations and land area but different economic bases. Although computing differences in fiscal capacity could present some difficulties, it could be done using proxy variables, such as PIT collection per capita or the value of the LG housing stock per capita. Block grants for delegated functions are crucial to the financing of municipal governments.61 They support education, social policy, culture, and firefighting, with funds allocated by line ministries and disbursed monthly to the municipalities monthly. Block grants account for about 80 percent of total transfers to LGs, and 49 percent of total LG revenues. The largest component, by far, is education, followed by social welfare; grants for culture and firefighting are relatively small (Figure 3.8). The MoES distributes separate grants for primary education, secondary education, and the transport costs for primary students. The formula for primary education takes into account the number of pupils, adjusted for population density, and the transport grant takes into account the number of students and route lengths. The funds for secondary schools are distributed on the basis of the number of secondary students and the share of students attending vocational schools. The adjustment coefficients for 57 Another contributing factor is that while Skopje has 25 percent of the national population it only receives 12 percent of the VAT pool. 58 For example, using data for 2012, Cyan et al. (2012) report that as a result of the VAT transfer, rural municipalities having 40 percent less pre-transfer revenue per capita than urban municipalities outside Skopje, eventually enjoy 10 percent more total revenue per capita after the allocation of the VAT transfer. They also report that there is a substantial effective equalization with the comparatively richer municipalities in Skopje. The Skopje municipalities, while having three times more pre-transfer own revenue per capita than urban municipalities elsewhere, after the VAT transfer end up having only twice as much revenue per capita. 59 Another potential improvement for the new equalization transfer would be to change the pool of funds to be distributed. Identifying the pool as a percentage of VAT revenues carries two issues. It would be less politically divisive to identify the source of funds for the equalization grant as a percentage of general CG revenues (not only VAT but also other CG tax revenues). Using only the VAT revenues could generate claims for more funds from particular jurisdictions where the VAT is collected, if not generated. The general funds could also be more stable over time. For the medium term, determining the pool of funds could be more sophisticated, based on more accurate estimates of the fiscal gap for each municipality. This gap is typically computed as the difference between spending needs and fiscal capacity, and on the equalization intended to fill the gaps identified. 60 VAT-derived transfers account for about 10 percent of total transfers to municipalities in 2016. 61 During the first phase of decentralization, these were called earmarked grants and did not cover salaries of public employees, which were still paid by the central government. 58 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE density and technical and vocational education and training (VET) are based on initial cost delivery data that are now more than a decade old. Figure 3.8. Structure of Block Transfers, 2006–16 Figure 3.9. Block Transfer, 2008–16 MKD, billions Percent of GDP 16 3.5 14 3.0 12 2.5 10 2.0 8 1.5 6 1.0 4 2 0.5 0 0 8 9 0 1 2 3 4 5 6 8 9 0 1 2 3 4 5 6 200 200 201 201 201 201 201 201 201 200 200 201 201 201 201 201 201 201 JJCulture JJEducation JJSocial protection JJFirefighting Source: Ministry of Finance and World Bank staff calculations. Source: Ministry of Finance and World Bank staff calculations. There are additional, smaller block grants that finance specific activities. The Ministry of Labour and Social Policy distributes one block grant for kindergartens and one for homes for the elderly located in the municipalities. The formula for kindergarten grants takes into account the number of children in each, expenses per child, useful space per child, and length of the heating period. Grants for homes for the elderly are based on number of people accommodated, maintenance expenses, and length of the heating period. The Ministry of Culture distributes block grants to cultural institutions, based on a series of expense estimates for existing facilities. Finally, the Ministry of Interior distributes the firefighting grant based on the number of firefighting units in each municipality. An important continuing issue with block grants is their (in)adequacy. Though the pool of funds for each transfer has gone up in both nominal and real terms (the most recent increase of 6.5 percent nominally is reflected in the 2018 budget), block transfers as a percentage of GDP are declining. According to 2016 data, there was a continuous decline in the amount allocated for block transfers over the previous four years. At their peak, they accounted for about 3 percent of GDP; in 2016, 2.4 percent (Figure 3.9). The initial funding for each grant, in 2005, was supposedly based on the historic costs borne by the responsible line ministries before decentralization. However, the actual funds allocated were less. This deficit has been carried forward over time, becoming a source of acrimony between LGs and the central government. There is also a misconception among local governments that block grants are supposed to cover all the service delivery costs.62 In reality, the funds transferred as part of the block grants are often insufficient, so that LGs have to top up the transfers with their own resources or let unpaid bills accumulate as arrears. Resolving the funding inadequacy may require at least two changes: (1) There is a need to clarify that block grants for delegated responsibilities are only supposed to finance minimum required expenditures in that sector, and that they need to be complemented by the municipalities themselves. (2) For the education grant, there is a need for annual dialogue between the MOF, the MoES, and ZELS to consider budget proposals for the minimum standard per student to be used in the 62 This is in part based on having separate budgets for use of the funds. Chapter 3: Fiscal Decentralization and Service Delivery 59 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 distribution formulas for primary and secondary education.63 A similar approach could apply to the other block grants. The distribution formulas also present several problems. However, these are related not to the criteria used for distribution but to the transparency of the calculations and decisions on of which municipality gets what transfers. Criteria for the distribution of earmarked grants are set annually by the respective line ministry. The ordinance is supposed to be adopted by the central government at least two months before the budget law applies. However, LGs have complained about a lack of transparency because the ordinance does not specify calculation procedures or the values of key parameters. Although the criteria are clear, their actual application is not because LGs are not informed about them. Annual publication of the calculations on the web page of each line ministry should be enough to resolve the problem. A serious problem is that the distribution of earmarked block grants is still very uneven. Rural municipalities, which constitute half of all municipalities and a fifth of the national population, do not receive any block grants for firefighting and in prior years it was not clear whether they received funds for culture, kindergartens, or high schools. While urban municipalities outside Skopje receive earmarked grants for each of these functions, the per capita amount they receive is less than for Skopje, except for transport for elementary and kindergarten students. This anti-rural bias is somewhat offset by the fact that facilities in urban municipalities may also be serving residents of adjacent rural municipalities, though no information is available on such cross-jurisdictional services. It is doubtful that all rural municipalities are served by such arrangements and, in any case, it is unlikely that their residents could hold urban municipalities accountable for the availability and quality of the services they receive. The cause of this situation is that at the beginning of the first wave of decentralization in 2005, municipalities that did not have physical facilities were not selected to receive grants. Reversing this situation will be complex. A potential remedy would be to change the selection formula to reflect the presence of potential clients (beneficiaries of the services) rather than the presence of physical facilities, and then allow the use of vouchers for clients in grant-less LGs to pay for services in municipalities that receive grants. One question that remains is whether there is a continuing need for all the current block grants based on delegated services. If some assignments were changed from delegated to devolved functions, municipalities would become responsible for financing them out of their general budget funds. As discussed in the expenditure assignment section, there may be good reasons to continue to consider education or social welfare as delegated functions. However, local culture and fire protection services could perhaps be devolved. Regardless of those changes, more precise definitions of delegated functions and what they entail in terms of the roles of LGs (such as contributions to finance the service) and line ministries (such as oversight and feedback on LG performance) should help to improve how block grants are used. Of course, changes in the level and composition of block grants should be offset by changes in other financing instruments for municipalities. 63 Regarding the formulas themselves, it would be desirable to update every three years or so the cost adjustment coefficients used for density, transportation etc. 60 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE Capital transfers for local investment projects in North Macedonia occur through three major channels: ad hoc grants from line ministries, Road Fund grants,64 and Development Fund grants managed by the Bureau of Regional Development.65 LGs account for a smaller share of total public investment—25 percent of the total66—than in most other countries. The share is 40 percent in UMICs and close to 66 percent in advanced economies. Capital transfers account for just 4 percent of all intergovernmental transfers in North Macedonia, compared to 20 percent in other UMICs. As shown in Figure 3.10, the lion’s share of local capital spending goes to economic infrastructure, roads in particular (51 percent), followed by solid waste treatment (7.1 percent) and water supply (4.9 percent). The geographical distribution of capital transfers is uneven, with Skopje receiving more than two thirds of capital transfers and close to one-fifth of the Road Fund grant pool. The system of capital grants is highly fragmented. Figure 3.10. Structure of Capital Spending by Local Governments, 2016 There are multiple sources of funding,67 and Percent municipalities can apply to 18 different programs for capital transfers.68 Getting funding for capital 51 projects is cumbersome: LGs must submit project proposals to different line ministries, and most programs envisage relatively small grants for each municipality. Furthermore funding awards suffer from a lack of clear guidelines and transparency. 7 This fragmentation leads to ineffectiveness and 5 0 0 significant administrative costs for municipalities 2 35 and end users. In particular, the fragmentation of capital grants among different agencies JJStreets JJSolid waste JJTelcom JJCulture JJWater JJEnergy JJOther and sectors is likely to impede prioritization of Source: MFR II. projects. It is also unclear whether the funding available for capital grants is adequate. Given the magnitude of the unmet needs in infrastructure, and the continuing demands of municipalities, the answer is likely to be no. Furthermore, as discussed below, the central government has not always met the requirement of allocating 1 percent of GDP for equitable regional development. Restructuring the system of capital grants in North Macedonia will require rethinking the objectives to be pursued and the means to achieve them. Objectives should include elimination of disparities in local infrastructure endowments, subsidies for capital projects with cross-jurisdictional benefits, and facilitation of joint operations across municipalities. Achieving these objectives will require consolidating the current plethora of investment programs into fewer block grant programs with a clearer focus and formal and transparent allocation criteria. That would allow municipalities to direct resources to the infrastructure that is most needed. The funding formula for capital block grants could use a sliding rule that takes into account LG fiscal capacity and ability to borrow in credit markets. 64 The Public Roads Fund finances construction and maintenance of local roads. The criteria for the allocation are prescribed in a Rulebook with the decisions updated every year by the Ministry of Transport and Communications. The grants do not require municipal matching contributions. The amounts allocated each year to different municipalities do not appear to be published. The total amount of road grants— including those for maintenance—has declined over time and is extremely small as percent of GDP. 65 The Development Fund is analyzed separately in the next subsection. 66 Average 2006–16. 67 Among CG agencies providing capital grants are the Youth and Sports Agency; the Ministries of Environment, Agriculture, and Local Self-Government; and The Protection and Rescue Directorate. 68 European Union 2016. For example, the same local project for water supply can be submitted to several investment programs of the national government (e.g., Ministry of Agriculture, Forestry and Water Economy; Ministry of Transport and Communications; and Ministry of Environment and Physical Planning). Chapter 3: Fiscal Decentralization and Service Delivery 61 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Other desirable features of the capital block grants to be explored include matching arrangements (to increase the “additionality” of central government funds) and performance-based criteria for continued disbursements. Also helpful would be more transparency in functional assignments—with a clearer delineation of the roles of central and local governments—in the construction and maintenance of public infrastructure, such as schools. How allocation of capital transfers is budgeted also needs to be rethought. In practice, capital grants have been allocated by ring fencing a portion of public funds for investment purposes and then allocating these based on project applications. This process is questionable because it creates dual budgeting by effectively excluding a portion of public funds from policy prioritization. A preferable approach would be to integrate programs involving both capital and current outlays within the multiyear framework (MTEF) and incorporate and evaluate operation and maintenance costs side by side with investment costs. In theory the Development Fund complements the rest of the transfers system; in practice, its operation has had serious shortcomings. While all the transfers reviewed so far help finance better and more equal access to local public services, Development Fund grants should contribute to the broader goal of regional economic convergence. The fund, which provides capital development grants, is administered by the Bureau of Regional Development in the Ministry of Local Self Government. The funds are distributed to regions considered “underdeveloped” according to specified. By law, funding for a “more equal regional development” (not the Development Fund itself) is set every year at 1 percent of GDP, plus external donations. However, the goal has not always been met. Besides public infrastructure projects—which to some extent duplicate the uses of capital grants—the Development Fund grants finance other types of projects. Beneficiaries need to submit their projects annually and financing often covers only part of the project costs. The criteria used and processes followed have not been transparent and, historically, there has been low utilization of the funds made available by the Bureau. The Development Fund could be rethought as an instrument for furthering a regional policy program for North Macedonia. This should be clearly differentiated from such subnational public finance instruments as revenue- sharing and block and capital grants, which are specifically aimed at providing similar levels of access to citizens regardless of where in the country they live. The Fund could also become a vehicle for attracting and managing EU and other development assistance funds. It could also be used to promote a policy of regional incentives for private investment in the least developed regions of North Macedonia. In all instances, the disbursement of funds should be preceded by analysis of the social and economic rates of return of the different programs, subject to rigorous cost-benefit analyses and allowing for multiyear completion schedules. 3.6  Municipal Indebtedness and Borrowing Controls The current legal system of local government borrowing is mostly adequate. Municipalities may borrow, with MOF consent, from domestic and foreign creditors and from the Treasury. LG borrowing is regulated by the Law on Financing of the Units of Local Self Government and the Law on Public Debt. Borrowing is subject to procedural rules for the form of borrowing request and the documents 62 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE required.69 LGs can use their property as collateral for borrowing but the law prohibits use of property that serves a public function or interest for this purpose. Short-term loans and short-term liquidity borrowing from the Treasury should not exceed 30 percent of current operational revenues from the previous fiscal year. Also, short- term borrowing is not connected with the fiscal year, but it should be repaid within 12 months. The limit for annual repayment of long-term borrowing—long-term loan servicing plus long-term borrowings from the Treasury—should not exceed 30 percent of current operational revenues from the previous fiscal year. The stock of long-term debt plus the issued guarantees also should not exceed 100 percent of current operational revenues from the previous fiscal year. The Law allows the central government to guarantee LG loans, but other than guarantees there is no legal basis for the central government to take responsibility for municipal debt. The amount of total formal debt held by municipalities in North Macedonia is equivalent to 0.1 percent of GDP70 but, as discussed in Chapter 2 there has been a significant accumulation of local budget arrears. Based on a survey conducted in 2017, local governments reported 1.2 percent of GDP in arrears. This accounted for a third of all arrears in the public sector (Figure 3.11).71 Most of the arrears arise from municipal budgets (4.6 billion MKD) and the other 2.5 billion MKD is on the books of municipal utility companies.72 Arrears tend to be concentrated in a small number of municipalities: seven are responsible for over 70 percent of total arrears, with Kumanovo, Ohrid, and Skopje having by far the largest stocks. Figure 3.11. Breakdown of Arrears, May 2017 Figure 3.12. Local Government Budget Arrears, May 2017 Percent Percent 21 27 53 15 12 11 14 10 26 3 3 5 JJCentral government (CG) JJPublic health providers JJOhrid JJTetovo JJKarpos JJStruga JJPublic enterprises (under the CG) JJPublic enterprises (under the LGs) JJChair JJKumanovo JJOthers JJLocal governments (LGs) Source: MOF and World Bank staff calculations. There appear to be multiple causes of arrears. Anecdotal evidence from the 2017 survey flags two main problems: taking on infrastructure projects and insufficient financing for education. Also, municipal SOEs owe considerable amounts to the Public Revenue Office (PRO), mainly because of past unpaid 69 MOD approval depends on two conditions: (1) The municipality has regularly submitted positively assessed financial reports for at least 24 months; and (2) it has had no arrears to creditors in the last 2 years. As a result, a dozen LGs with arrears dating to before decentralization in 2005 with a state construction company have not been able to get credit. Municipalities in financial difficulties –which get their accounts blocked—are given a monthly ceiling on expenditures set by courts and all revenues above that threshold go to creditors. It is not clear how courts calculate these thresholds. 70 However, some municipalities, like the 12 mentioned in the previous footnote have formal debt exceeds their total annual revenues. 71 MOF data show an upper-bound estimate of about around 3.5 percent of GDP in May 2017 for all levels of government, not just municipalities and their public enterprises. This means that municipality arrears may be as much as 1.15 percent of GDP. As can be seen in Chart 3.11 current arrears are spread across the entire public sector: central government, local government, local SOEs, central government SOEs, health providers. 72 A significant amount of arrears are owed to other public institutions (SOEs to the PRO; LGs to SOEs; SOEs to SOEs, etc.). Chapter 3: Fiscal Decentralization and Service Delivery 63 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 corporate income tax.73 Often, arrears accumulate in order to finance running costs. However, they are caused by a variety of institutional shortcomings, particularly unrealistic budgeting and ad hoc budget revisions throughout the year, unrealistic and non-binding MTBFs, weaknesses in fiscal reporting and the treasury system, weak commitment control systems, mismatches between LG revenue flows and responsibilities, and poor management of SOEs. What can be done to contain and eventually solve the arrears problem? Deep reforms of LGs and SOEs will be needed to address the structural drivers of arrears accumulation. Recent efforts to introduce a commitment control module in the Treasury system have been only partially successful. The reforms should also focus on reinforcing other aspects of the budgeting process, such as formulation of realistic budgets and revisions for the year; drafting a credible and binding MTBF, and strong and effective commitment controls. This will be temporarily and partially covered by the introduction of the new arrears module for the MOF.74 One way to reduce their arrears is to allow municipal SOEs to collect what is owed to them more effectively. The costs to SOEs of forced collection of user invoices are very high. At present, SOEs need to cover the costs of notaries, bailiffs, and court fees before the forced collection procedure is initiated. Lowering these mandatory costs could allow utility companies to work more actively on collecting overdue bills. Greater flexibility with cost-setting controls, and allowing full-cost recovery should also help reduce the arrears of local public utilities. Finally, allowing LGs more resources and revenue autonomy should be considered part of a long-term solution. 3.7  Reform Options Short Term 1. Enhance the clarity of functional assignments, especially the delegated services currently financed through block grants, by better defining responsibility for rulemaking—the role of line ministries— and responsibility for financing, which should be the responsibility of the whole government. 2. Consider measures to increase local own-revenue capacity—such as introducing new tax sources and measures for better collection of current taxes. This would not only increase LG revenue sufficiency and adequacy, it would enhance the fiscal responsibility and accountability of local officials, and ultimately the efficiency of local spending. 3. Explore avenues for improving the effectiveness of conditional block grants, such as making more transparent their calculation and the selection of municipalities to get transfers. Medium Term 4. Clearly define the objectives of the overall strategy for reforming intergovernmental fiscal transfers, such as arriving at more transparent spending roles for central and municipal governments, improving the revenue adequacy and predictability of municipal budgets, enhancing fairness in 73 The PRO calculates a penalty interest rate for overdue collectables, which was 18 percent annually, which also accumulated interest. 74 The MOF uses web-based software that will allow for systematic and comprehensive monthly reporting of public. The module is capturing over 1,300 public institutions, covering central and local governments, their SOEs, all schools and kindergartens, and public health providers. Through this system, arrears data collected in the end-May survey will be improved through clearer specification of data requirements in the software. 64 Chapter 3: Fiscal Decentralization and Service Delivery SOWING THE SEEDS OF A SUSTAINABLE FUTURE access to public services across municipalities by reforming the transfers system, and bringing back fiscal and budgetary discipline at the local level. 5. Given North Macedonia’s size, two levels of government fit its decentralization needs. However, there is work to be done to move to more efficient use of resources. Here, the government should consider the possibility of asymmetric assignments of responsibilities and further develop and support inter-municipal cooperation in delivering services that require some minimum scale for efficiency and administrative capacity. 6. Improve the measurement of spending needs derived from functional assignments, to help clarify whether services are being underprovided and underfunded, and to better understand the need for higher transfers and granting LGs more revenue- generating capacities. It would also enhance the equalization power of the general VAT grant. 7. Improve LG revenues by increasing the amount of PIT collections going to the municipalities, including against other revenue enhancing options and the institutional capacity of LGs. 8. Enhance the equalization capacity of the general grant by reforming the current formula to better capture expenditure needs, and by allowing different LGs greater revenue collection capacity. Though the current system equalizes some horizontal fiscal disparities, they need to be reduced further. This will require time for reaching consensus on the appropriate level of equalization and for simulating the equalization results for different scenarios. 9. Restructure the system of capital grants. That requires rethinking the objectives to be pursued and the means to achieve them and consolidating the numerous capital transfer programs into a few capital block grants. The budgeting process for allocating capital transfers also deserves reconsideration. 10. Reform the Development Fund so that it has more impact on the broad goal of regional economic convergence. 11. Control local government debt and reduce budgetary arrears. This hinges on building up budgeting process by, e.g., formulating realistic budgets and in-year revisions, drafting a credible and binding MTBF, developing firm and effective commitment control systems, and improving the management of SOEs. Chapter 3: Fiscal Decentralization and Service Delivery 65 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Chapter 4: Tax Policy Options for Increasing Revenue North Macedonia’s strategic decision to use a low-tax environment and other tax incentives to attract FDI has borne fruit, but its fiscal cost has been sizable. Government revenues as a share of GDP have declined and are now much lower than those of its neighbors and other European countries. Contributing to the lower tax revenues resulting from the low- tax incentives is the country’s low revenue efficiency. This chapter explores ways to make the tax system more sustainable and identifies specific reforms that could be adopted in the short, medium, and long term. Among them are adoption of a more progressive PIT structure with higher marginal rates for high- income earners; broadening the PIT base by removing exemptions for specific capital income and gains; broadening the value- added tax base by removing the preferential treatment of selected goods and services; raising the excise taxes on tobacco, coal, and diesel fuel; and using cost-benefit analysis to explore whether the government’s investment incentive strategies have indeed been beneficial. 4.1  North Macedonia’s Tax System in International Context For the past 10 years revenue collection as a share of GDP in North Macedonia has been declining until it is now among the lowest in the region. Between 2007 and 2016 central government revenues fell from 32 percent of GDP to 28 percent as general government revenue also declined, to 30.3 percent of GDP. This is far below the EU average of 45 percent and, among the WB6 countries, only Albania and Kosovo collect less revenue. Tax revenue and SSC—the largest sources of revenue in North Macedonia—are relatively modest compared to both the Western Balkan and other EU countries and have been dropping. In fact, the country’s tax revenue is the lowest of the WB6. The decline is mainly due to lower VAT and SSC, but all other revenue sources except for excises and profit and property tax have also slipped slightly. With its heavy reliance on indirect taxes and SSC revenue, North Macedonia’s tax structure is similar to that of other WB6 countries. Indirect taxes—the VAT, excise taxes, and import taxes—raised about 12 percent of GDP in revenue, with VAT accounting for over half of this. SSC represent the second largest source of revenue, contributing 8.4 percent of GDP to the budget in 2016. PIT contributed only 2.4 percent and CIT just 1.8 percent of GDP. Compared to other WB6 and EU countries, the revenues-to-GDP ratios of North Macedonia are particularly low for VAT, SSC and excises, which together account for 65 percent of total revenue. Income tax collections (PIT and CIT) while at par with those of other WB6 countries, are substantially lower than the EU average. The decline in North Macedonia’s revenue is partly due to the reduction in PIT and CIT tax rates, and the introduction of exemptions that reduced the CIT tax base. Most of these measures were introduced in 2009 to support the economy after the financial crisis and the CIT exemptions have since expired. 66 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE Because its tax rates are lower than those of its peers, North Macedonia has scope to raise more tax revenue. Its standard VAT and SSC rates are far below the EU average (as are its income tax rates); its 10 percent CIT rate is in the middle of the WB6 range, below those of Albania and Serbia; and the flat 10 percent PIT rate is one of the lowest in the region. Figure 4.1. General Government Revenues, 2016 Figure 4.2. Changes in General Government Revenues, 2006–16 Percent of GDP 50 Total revenues Taxes 40 Contributions Import duties 30 Other revenue VAT 20 Other taxes Personal income tax 10 Excises Profit tax Property tax 0 V HR LV A SVN SR B BIH MNE EST SVK BGR LTU MKD ALB KOS -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0 0.5 Percent of GDP Source: Eurostat and World Bank Staff calculations from MOF data. Source: MOF. Table 4.1. North Macedonia: Government Budgetary Operations, 2009–17 2009 2010 2011 2012 2013 2014 2015 2016 2017 Percent of GDP Total Revenues 32.8 32.4 31.7 32.1 30.1 29.7 31.0 30.3 30.9 Taxes and contributions 27.6 27.1 27.0 26.6 25.4 25.8 26.5 26.5 26.7 Taxes 18.3 18.3 18.5 17.9 17.0 17.4 18.0 18.1 18.2 Direct taxes 3.4 3.1 3.1 3.0 3.2 3.6 4.7 4.4 4.5 Personal income tax 2.1 2.1 2.1 2.1 2.1 2.4 2.4 2.4 2.5 Profit tax 1.1 0.8 0.8 0.8 0.9 1.0 2.2 1.8 1.8 Property tax 0.1 0.2 0.2 0.1 0.2 0.2 0.2 0.2 0.2 Indirect taxes 13.2 13.1 13.3 12.7 12.0 12.4 11.8 12.2 12.3 Vat 8.5 8.6 9.1 8.2 7.9 8.3 7.5 7.7 7.7 Excises 3.5 3.4 3.3 3.6 3.2 3.3 3.5 3.7 3.7 Import duties 1.3 1.1 0.8 0.9 0.8 0.8 0.8 0.8 0.8 Other taxes 1.7 2.1 2.1 2.2 1.9 1.5 1.5 1.5 1.4 Contributions 9.4 8.8 8.6 8.7 8.5 8.4 8.6 8.4 8.5 Other revenue 5.1 5.3 4.7 5.4 4.6 3.9 4.4 3.9 4.2 Percent of total revenue Total Revenues 100 100 100 100 100 100 100 100 100 Taxes and contributions 84.3 83.8 85.3 83.1 84.6 86.9 85.7 87.2 86.5 Taxes 55.8 56.4 58.3 55.8 56.5 58.7 58.0 59.5 58.8 Direct taxes 10.2 9.6 9.8 9.4 10.5 12.0 15.3 14.5 14.6 Personal income tax 6.5 6.4 6.6 6.5 6.9 8.0 7.6 8.0 8.0 Profit tax 3.3 2.6 2.6 2.4 2.9 3.2 6.9 5.9 5.9 Chapter 4: Tax Policy Options for Increasing Revenue 67 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table 4.1. North Macedonia: Government Budgetary Operations, 2009–17 (continued) 2009 2010 2011 2012 2013 2014 2015 2016 2017 Property tax 0.4 0.5 0.5 0.4 0.6 0.7 0.7 0.6 0.6 Indirect taxes 40.4 40.5 41.8 39.5 39.8 41.8 38.0 40.1 39.8 Vat 25.9 26.6 28.7 25.7 26.4 28.0 24.1 25.3 25.0 Excises 10.7 10.5 10.5 11.1 10.6 11.1 11.4 12.2 12.1 Import duties 3.8 3.3 2.6 2.7 2.8 2.7 2.5 2.6 2.7 Other taxes 5.1 6.4 6.7 6.9 6.2 5.0 4.8 4.9 4.5 Contributions 28.6 27.3 27.0 27.2 28.1 28.2 27.7 27.7 27.6 Other revenue 15.7 16.2 14.7 16.9 15.4 13.1 14.3 12.8 13.5 Source: North Macedonia authorities. Note: Revenues from the PESR are not included. Table 4.2. General Government Tax Revenues, 2016 or Latest Year Available Percent of GDP Direct Taxes Indirect Taxes Total tax General and social Social Corpora- Unalloc- Property sales, Year security Total Individual Total turnover Excises Other security tions able taxes or contrib. contrib. VAT revenue Albania 2011 22.8 3.9 2.2 1.7 0.0 0.1 13.9 9.1 3.1 1.7 3.8 Bosnia and 2013 37.6 3.1 2.0 1.1 0.0 0.4 18.5 11.8 4.8 1.9 15.5 Herzegovina Kosovo 2013 21.8 3.4 1.7 1.2 0.4 0.3 19.2 10.9 5.9 2.4 n.a. North Macedonia 2016 26.5 4.4 2.4 1.8 0.2 0.2 12.2 7.7 3.7 0.8 8.4 Montenegro Serbia 2016 38.4 6.3 3.6 2.4 0.2 10.5 5.3 1.6 12.3 WB6 average (excl. 30.1 4.2 2.4 1.6 0.1 0.2 12.9 10.6 4.8 1.9 10.5 North Macedonia) Croatia 2016 37.9 6.6 3.7 2.2 0.6 13.0 5.2 1.3 11.7 Bulgaria 2016 29.0 5.4 3.0 2.1 0.2 9.2 5.3 0.8 7.8 Estonia 2016 34.7 7.7 6.0 1.7 0.0 9.4 0.2 0.8 11.8 Latvia 2016 31.6 8.4 6.4 1.7 0.4 8.2 4.8 1.3 8.6 Lithuania 2016 30.2 5.7 4.0 1.6 0.0 7.8 3.3 0.6 12.5 Slovak Republic 2016 32.4 7.3 3.3 3.5 0.3 6.7 3.0 1.0 14.3 Slovenia 2016 36.9 7.4 5.2 1.6 0.6 8.2 5.1 1.3 14.8 STEE7s average 33.2 6.9 4.5 2.1 0.3 8.9 3.8 1.0 11.6 EU28 average 2016 40.0 13.0 9.3 2.6 0.9 7.0 3.8 2.3 13.3 Source: Eurostat and PRO for North Macedonia. Table 4.3. VAT, Corporate and Individual Income Tax, and Social Security Contribution Rates, 2017 Percent GDP per capita VAT thresholds Current standard Corp. income Top individual Social security 2016 (USD) (thousand USD) VAT rates tax rates income tax rates contrib. rates Albania 4,126 47.3 20 15 23 24.5 Bosnia and 4,298 n/a 17 Herzegovina FBiH 10 10 41.5 RS 10 10 33 68 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 4.3. VAT, Corporate and Individual Income Tax, and Social Security Contribution Rates, 2017 (continued) GDP per capita VAT thresholds Current standard Corp. income Top individual Social security 2016 (USD) (thousand USD) VAT rates tax rates income tax rates contrib. rates Kosovo 3,602 36.9 18 10 10 10 North Macedonia 5,264 43.1 18 10 10 27 Montenegro 6,707 23.9 21 9 11 33.8 Serbia 5,348 74.8 20 15 15 37.8 WB6 average (excl. 4,816 19 10 12 21 North Macedonia) Croatia 12,165 40.3 25 20 40 37.2 Bulgaria 7,377 33.9 20 10 10 30.3 Estonia 17,786 21.2 20 20 20 35.4 Latvia 14,063 66.1 21 15 23 34.1 Lithuania 14,893 59.6 21 15 15 41.9 Slovak Republic 16,499 n/a 20 21 25 48.6 Slovenia 21,668 66.4 22 19 50 38.2 STEE7s average 14,921 48 21.29 17.14 26.14 37.96 EU28 average n/a 21.5 21.5 38.4 36.6 Source: Ministry of Finance, KPMG. Notes: PIT rate refers to the top marginal tax rate, VAT rate to the standard rate. 4.1.1  Effective Tax Rates on Factor Incomes and Consumption Measuring effective tax rates (ETR) on factor inputs and consumption gives a macroeconomic picture of potential distortions in factor inputs and the consumption/savings decision. This section estimates macroeconomic ETRs75 that apply to labor, capital, and consumption by assigning the revenue collected from each tax instrument to labor, capital, or consumption to form the numerator. The revenue is then divided by the aggregate payments to labor, capital, or consumption drawn from national accounts (the methodology is described in Mendoza, Razin, and Tesar (1994)). The data sources and assumptions are detailed in Box 4.1. Table 4.4. Effective Tax Rates Compared, North Macedonia and EU Peers Peers’ Effective tax rate: MKD BGR EST LVA LTU SVN SVK EU25 average ETR consumption 15.8 22.4 14 17.7 17.2 23 18.7 18.83 21.4 ETR labor 32.1 24.6 33.2 32 32 35.2 31.9 31.48 34.2 ETR capital 6.7 26.3 9.9 5.5 20.5 14.8 15.40 29.4 Source: Eurostat 2013 Taxation Trends in the EU and own calculations based on National Statistics and data from PRO. North Macedonia has some room to increase the ETR on consumption, which is estimated at 16 percent—lower than the 21 percent EU average and the 19 percent average of peer countries for which data are available (Table 4.4).76 While North Macedonia’s VAT rate of 18 percent is close to the EU average, the effective rate on consumption could be increased by broadening the VAT base, raising excise taxes on cigarettes and fuels, and improving enforcement of the VAT. 75 Eurostat calls macro ETRs “implicit tax rates.” Its methodology requires more data and adds refinements to the one used here; comparisons should be made with cre 76 Data on ETRs are available only for EU peer countries, which are Bulgaria, Estonia, Lithuania, Latvia, Slovakia, and Slovenia. Chapter 4: Tax Policy Options for Increasing Revenue 69 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 The effective rate on labor income of 32 percent does not require major changes. It is close to the EU average of 34 per cent and equal to that of peer countries. The low PIT rate of 10 percent is counterbalanced by a 27 percent SSC rate. Changes to the social security floor and ceiling contributions would improve equity, as they introduce a high effective rate on labor income for the bottom of the income distribution and a low rate for the top. Another equity-improving measure is to raise the top PIT rate, which should be targeted at the top of the income distribution to avoid a drastic increase in the ETR on labor. However, at 7 percent the ETR on capital is so low that options to increase it should be considered. The average in the EU is 29.4 percent and for peer countries 15.4 percent. The rate on capital is low for four reasons: (1) The flat and low-rated PIT schedule limits the taxation of capital income. which is concentrated at the top of the income distribution. (2) The PIT base excludes capital income from several types of securities, the sale of immovable property, and the sale of securities. (3) Revenue from the property tax represents less than 0.5 percent of GDP. (4) The CIT schedule has low rates and allows a range of exemptions. Box 4.1. Effective Tax Rates on Factor Income and Consumption To measure effective rates on factor incomes and consumption, we combined the 2015 central government tax revenue collection data with national statistics from the income and expenditure approaches. We use Public Revenue Office (PRO) data to assign the shares of personal income tax (PIT) revenue from labor and capital. The calculations follow the approach of Mendoza, Razin, and Tesar (1994): tlabor = (PIT * Labor sharePIT + SSC) / Wages tcapital = (PIT * Capital sharePIT + CIT + Property Taxes) / Operating Surplus tconsumption = (VAT + Excises) / (Household Consumption + Government Consumption - GW - VAT - Excises)  The calculations make the following assumptions and simplifications: •• We transfer mixed income from the personal income tax to capital. •• The labor share and the capital share in the PIT are calculated before deductions and are likely to over-estimate the share of labor income in PIT revenue. •• The base for the tax rate on labor does not include employer-paid social security contributions. •• Local taxes are not included. •• The incidence of corporate taxes falls entirely on capital. North Macedonia could increase effective taxation of capital, given the large gap between the ETRs on labor and capital, and the low ETR on capital. The exact size of the gap between the ETRs on labor and capital depends on the assumptions made in the analysis (Box 4.1), its order of magnitude is large, and represents a distinctive feature of North Macedonia’s tax system compared to EU countries. Although the WB6 might be a better comparison group, data on their ETRs are not available. Reducing 70 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE the gap would equalize returns to factor incomes and lower the relative production distortion in the choice of labor versus capital; it would also ensure more equity. Although one rationale for lower ETRs on capital is that capital is more elastic than labor, the current gap in North Macedonia seems too large, and the country could increase taxation of capital without hurting its competitiveness. 4.2  Personal Income Tax North Macedonia has room to increase its top marginal tax rate. It currently taxes personal income at a flat 10 percent. Until 2006 the tax system was progressive, with rates of 15, 18 and 24 percent. In 2007 it was replaced by a 12 percent flat tax, which in 2008 was lowered to 10 percent. North Macedonia collects only 8 percent of its tax revenue from the PIT, equivalent to 2.5 percent of GDP; this is just above the WB6 average of 2.4 percent, but much lower than the 4.5 percent average for the STEE7 and the 9.3 percent EU28 average. This is in part due to North Macedonia’s low rates and an absence of progressive tax rates in its personal income tax schedule. The tax base generally meets international standards although North Macedonia offers deductions and exemptions for specific types of capital income and gains. The tax base includes revenue from salaried income, pensions, business activity and services, capital income (property, dividends, interest from loans and bonds, royalties) and income from capital gains. The annual individual tax allowance of MKD 90,000 (€1,450) is deductible from salaried income and allows for some progressivity in the income tax schedule. Capital gains represent the positive difference between the purchase price and the selling price of real estate, securities, and shares. The base for capital gains is 70 percent of the gain earned, though gains from selling immovable property are reduced or exempted if the seller has resided in the property for a given period. In addition, the sale of securities is tax-exempt. Most capital income from dividends and interest is part of the tax base. Income from property and property rights is given a 30 percent deduction for maintenance and management expenses. Copyrights and patents allow for deduction of expenses ranging from 25 to 60 percent of gross income. Finally, other income, such as e-commerce, receives a 35 percent deduction. Full tax exemptions are applied to some employment-related expenses, awards and scholarships, alimony payments, donations to qualifying institutions, specific types of interest income (on deposits, current accounts, and government bonds) and for a period of 10 years the salaries of employees of firms in the technological industrial development zones (TIDZs). Expanding the PIT tax base would require a review of the rationale for each type of exemption and deduction. A comprehensive review is also necessary to identify whether the tax system is the best policy instrument to achieve the stated goals. For example, the revenue and equity implications of tax exemptions for owner-occupied properties should be weighed against the government’s home- ownership and mobility objectives (See section 4.6 for a full discussion of the tax treatment of owner- occupied properties). Removal of the exemption on gains from the sales of securities is recommended as it would also align North Macedonia with international best practices. Figure 4.3 displays the number of PIT taxpayers in 2017 and the ETRs based on PRO data. The distribution of taxpayers follows a log normal distribution with a long right tail (Panel A).77 Deductions 77 As with most income distributions, the top tail of North Macedonia’s income distribution is well approximated by a Pareto distribution (Shape parameter of 2). This approximation was used in simulating reform scenarios. Chapter 4: Tax Policy Options for Increasing Revenue 71 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 make the tax system slightly progressive (Panel B): the ETR rises from 4 percent for low-income earners to 7 percent for top earners. Figure 4.4 shows the share of total income by source: salary, pensions, and capital income. At the bottom of the income distribution salaries represent 60 percent of total income and pensions 40 percent. Salary income then grows to 80 percent of total income, replacing pensions at the middle of the income distribution. At the top of the income distribution, salaries decrease to 65 percent of total income while capital income rises linearly to reach 35 percent. Thus, capital income is concentrated at the top of the income distribution: it represents 34.4 percent of income for the top 3 percent of earners and only 2.5 percent for the bottom 97 percent. Figure 4.3. Personal Income Tax: Number of Taxpayers and Effective Tax Rate, 2017 A. Taxable income distribution B. Effective tax rate 125 0.08 100 0.07 75 0.06 50 0.05 25 0.04 0 0.03 0 500 1,000 1,500 2,000 0 500 1,000 1,500 2,000 Taxable income, thousand MKD Gross income, thousand MKD Source: PRO administrative tax returns data. Notes: Kernel-weighted local averages. Introducing a higher marginal PIT rate would Figure 4.4. Taxable Income by Source Along the bring in more revenue and make the Macedonian Income Distribution Share of taxable income tax schedule more progressive. It is not likely to 0.9 have much distortionary effect on the economy 0.8 because of the current low-income tax rate and 0.7 the low ETR on capital income compared to peer 0.6 countries (which results in small incentives to 0.5 move capital). The optimal design would depend 0.4 on Government’s objectives for revenue collection 0.3 and how many Macedonians could face a higher 0.2 burden, holding the tax base constant. Two 0.1 important decisions are the top marginal tax rate 0 0 500 1,000 1,500 2,000 and the income threshold for the top bracket. Taxable income, thousand MKD Other parameters to be considered are differential ÂÂSalary ÂÂPensions ---- Capital rates on capital and labor income, deductions on Source: PRO tax returns data. Note: Kernel-weighted local averages. capital income, and the amount of the standard salary deduction. Simulations of different reform scenarios, discussed below, highlight the revenue implications of these choices. 72 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 4.5. Simulated Tax Revenue Gains. Reform Scenario with Uniform Rate for Income Sources A. Simulated revenue gains, elasticity=0 B. Simulated revenue gains, top rate 18% Percentage increase in tax revenue, 2017 base 0.30 0.30 0.25 0.25 0.20 0.20 0.15 0.15 0.10 0.10 0.05 0.05 0 0 400 600 800 1,000 1,200 1,400 1,600 400 600 800 1,000 1,200 1,400 1,600 Threshold for higher MTR, thousand MKD Threshold for higher MTR, thousand MKD ÂÂ21% ÂÂ18% ---- 15% ÂÂ0 ÂÂ0.4 ---- 0.8 Source: PRO tax return data. A reasonable reform scenario would raise the top marginal tax rate from 10 to 18 percent for taxable income above MKD 720 ,000, which would affect the top 2.5 percent of earners. In this scenario, and absent any taxpayer decisions to reduce income when faced with a higher tax rate, PIT revenue would automatically increase by 15.9 percent. Figure 4.5 highlights the tax revenue gains for different reform scenarios: Panel A shows the increase in tax revenue (vertical axis) as a function of the income threshold for the higher marginal tax rate (horizontal axis). The red line shows the MKD 720,000 income threshold the government is currently considering. Figure 4.5 shows the revenue increase as a function of the top marginal tax rate of either 15, 18, or 21 percent. Given North Macedonia’s high ETR on labor, applying the higher rate for the top of the income distribution would ensure that more capital income is taxed and limit the labor ETR increase. With most taxpayers having taxable income of less than MKD 550,000 (minimum taxable income for 4 percent top earners), a threshold below that would impact a much larger share of taxpayers (Figure 4.3) but also raise substantially more tax revenue (Figure 4.5). The considered threshold of about MKD 720,000 (the top 2.5 percent of earners) seems to appropriately balance these objectives. A reduction in incentives to earn more income because tax rates go up is not likely to have much impact on revenue gains. However, a realistic reform scenario should consider taxpayer responses to higher tax rates. Evidence, principally from OECD countries, points to moderate elasticities of personal income with respect to the tax rate in the short and medium term.78 In a survey of the literature by Saez, Slemrod and Giertz (2012), it appears that the elasticity of labor income is about 0.2, though possibly higher for high-income earners. For the top 3 percent of earners in North Macedonia, salaries provide 65 percent of total income. Elasticities of capital income and gains are more uncertain and could be higher. They depend in part on opportunities to shift capital income across jurisdictions and across fiscal years. Capital income and gains account for 35 percent of total income for the top 3 percent of earners. Income from independent activities is also likely to be more elastic than salaried income, in part due to a potential increase in tax evasion on income with weaker third-party information. Therefore, increased enforcement and monitoring of independent business activities by tax administration is encouraged. Panel B of Figure 4.5 shows the revenue gains with a top tax rate 78 While salaried income is unlikely to drop drastically in the short term, higher personal income taxation could affect longer-term investment in education; risk-taking behavior, such as entrepreneurship, and formal registration of new firms. There have been few estimates of long-term elasticities. Chapter 4: Tax Policy Options for Increasing Revenue 73 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 of 18 percent as a function of the chosen threshold for elasticities of 0, 0.4, and 0.8. Given previous studies and the composition of income at the top of the North Macedonian income distribution, a central scenario is to assume an elasticity of taxable income of 0.4. With this elasticity, raising the top rate to 18 percent would increase revenue by 16 percent, 1.4 percent less than in the zero-elasticity scenario. An elasticity of 0.8, which would be in the upper bound of the literature estimates, would bring in additional revenue of 14.5 percent, 2.9 percent less than in the zero-elasticity scenario. Since North Macedonia has a very low top marginal tax rate, and since salary income represents over half of top incomes, the revenue losses from income reduction responses are likely to be minimal. Figure 4.6. Simulated Tax Revenue Gains from Reform Scenario A. Simulated revenue gains, top MTR 18% B. Simulated revenue gains, top MTR 18% Percentage increase in tax revenue Percentage increase in tax revenue 0.22 0.22 0.20 0.20 0.18 0.18 0.16 0.16 0.14 0.14 0.12 0.12 0.10 0.10 400 600 800 1,000 1,200 1,400 1,600 400 600 800 1,000 1,200 1,400 1,600 Threshold for higher MTR, thousand MKD Threshold for higher MTR, thousand MKD ÂÂSame rates labor and capital ---- Flat 15% rate on capital ÂÂSame rates labor and capital ---- 15% capital then 18% MTR all Source: PRO tax returns data and authors’ calculations. Another possible reform would apply differentiated rates on capital income and on salaries and pensions (labor income). One scenario would set a flat 15 percent rate on taxable capital income and rates of 10 percent on taxable labor income up to MKD 720,000, after which the 18 percent rate would apply to additional labor income. The dotted line in Figure 4.6 shows how this scenario would affect government revenues: at the MKD 720,000 threshold (the red line), this would raise tax revenue by 14.9 percent. The full line highlights the previous scenario, where capital and labor had the same rates of 10 percent below the threshold and 18 percent above it: at the MKD 720,000 threshold, the differential taxation of capital and labor reduces revenue by 1 percent. Indeed, while all capital is taxed at a higher rate than it is currently, capital at the top of the income distribution is taxed at 15 percent instead of 18 percent, as the previous scenario proposed. However, there are drawbacks to differential taxation of capital and labor: ( 1) This practice is rare internationally and generates incentives to shift income-type reporting. (2) It would bring in less revenue than the alternative scenario. (3) At the top of the income distribution, capital income would be taxed at a lower rate than labor income, even though North Macedonia’s ETR on capital income is already low. Unless there is reason to believe that capital income is particularly elastic to higher tax rates, applying the same rate to both income types above the threshold seems preferable. A final possibility, illustrated in Figure 4.6, Panel B, highlights the revenue implications of taxing capital income at 15 percent and above the threshold taxing both capital and labor income at 18 percent. At the MKD 720,000 threshold, this would increase revenue by 17 percent—1.1 percent higher than treating capital and labor income the same. By taxing capital at a higher rate on average, this scenario 74 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE rebalances the aggregate ETRs on capital and labor. It is also worth noting that, based on Figure 4.6, if a higher threshold is considered for the increase in the marginal tax rate, taxing capital income at 15 percent below the threshold allows the reform to have a larger revenue impact. Finally, combining tax rate increases with changes Figure 4.7. Simulated Tax Revenue Gains from Reform Scenario with Rate and Base Changes to the tax base would widen the capital base Percentage increase in tax revenue and increase the standard labor deduction. This 0.22 could be done by broadening the capital base by 0.20 reducing permitted deductions by 10 to 30 percent 0.18 or raising the standard deduction for salaries 0.16 and pensions from MKD 90,000 to MKD 96,000. 0.14 Capital income deductions represent 16 percent 0.12 of gross capital income—about 2 percent of total net income. With the current 10 percent tax rate, 0.10 reducing permitted deductions on capital income 0.08 by 10 percent would yield a 0.2 percent increase 0.06 400 600 800 1,000 1,200 1,400 1,600 in tax revenue, and reducing them by 30 percent Threshold for higher MTR, thousand MKD would yield a 0.6 percent increase. With a flat  ÂLabor and capital base change ÂÂ20% drop of capital deductions ---- Expand labor deduction to 96k ---- No change in tax base 15 percent tax on capital income, the revenue Source: PRO tax return data and authors’ calculations. gains would be multiplied by 1.5. In the 10 percent to 18 percent schedule—with the same rate applied to labor and capital—a 10 percent decrease in capital deductions would increase tax revenue by 0.3 percent and a 30 percent reduction would raise it by 0.9 percent. This could be combined with raising the standard annual labor deduction from MKD 90,000 to MKD 96,000, which would decrease tax revenue by 3 percent under the current 10 percent rate. In conclusion, a higher marginal tax rate of 18 percent above a high threshold, a broader capital tax base, and a moderate increase in the standard labor deduction would offer North Macedonia many benefits. The combination measure would substantially increase both tax revenues and PIT progressivity. With an 18 percent marginal tax rate applied to taxable income above MKD 720,000, a 20 percent reduction in permitted deductions on capital income, and raising the standard annual labor deduction to MKD 96,000, PIT revenues could go up by about 12 percent, or 0.35 percent of GDP. This estimate does not take into account behavioral response which might reduce revenue collection. While large efficiency costs are not expected, the measure should be accompanied by heightened enforcement of difficult-to-track revenue sources and top taxpayers to limit evasion and avoidance. Also, an evaluation of individuals’ responses to the higher tax rate is recommended to evaluate efficiency costs and inform future reforms and tax designs. 4.3  Indirect Taxes: Vat and Excise Duties Indirect taxes like VAT, customs duties, and excise taxes are an important source of revenue for North Macedonia. Over the last 10 years, VAT revenue has brought in about 28 percent of total central government revenue. Since 2011 the contribution of excise duties on alcohol, tobacco, mineral oils and motor vehicles has risen from 11 to 13 percent of revenue. However, the contribution of customs duties has fallen from 5 to 3 percent of revenue. Chapter 4: Tax Policy Options for Increasing Revenue 75 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 The importance of indirect taxes is likely to grow as the population ages and the base for income taxes shrinks. Though its population is relatively young compared to ECA countries, the share of those aged over 65 in North Macedonia is expected to increase rapidly from 12 percent in 2015 to 26 percent in 2050. Broad-based consumption taxes like the VAT are also a relatively efficient way of taxing remittances. Official remittances represent about 2 percent of GDP in North Macedonia. However, according to the balance of payments, at 13 percent of GDP in 2016 net cash exchanges are substantial, and anecdotal evidence suggests that some of these represent remittances. 4.3.1 VAT Although this is not currently being considered by the government, North Macedonia could collect more revenue from the VAT by expanding the base (moving toward a single VAT rate) and enforcing compliance. The VAT revenue ratio or C-efficiency ratio79 for North Macedonia hovers around 0.6, which is low compared to WB6 peers. This suggests that revenue is being lost through exemptions and reduced rates, as well as tax avoidance and evasion. About 36 percent of the consumption basket is taxed at the standard VAT rate of 18 percent. The remainder is taxed at a reduced rate of 5 percent or input taxed (exempt without the right to deduct the VAT paid on purchases). Applying a mix of standard and reduced rates creates opportunities for abuse, as sellers can classify goods and services subject to the standard rate as qualifying for the reduced rate. In contrast, a broad-based VAT improves both economic and administrative efficiency. A recent IMF report highlighted weakness in the country’s compliance risk management, a limited formal administrative dispute review process, inadequate coverage of tax audits, significant amounts of tax arrears, and weaknesses in the risk review of VAT refunds (IMF 2016). While North Macedonia has begun to improve its tax administration, further reforms are still being considered. North Macedonia’s VAT schedule of goods and Figure 4.8. VAT Revenue Ratio Compared, 2016 services receiving preferential treatment is 0.9 generally in line with the EU VAT directive, except 0.8 for a few specific goods, but there are goods and 0.7 services where preferential treatment might be 0.6 0.5 removed. In OECD countries, reduced rates are 0.4 generally applied to essentials (e.g. health care 0.3 and food), utilities, socially desirable activities 0.2 (e.g. culture and sport) or activities that support 0.1 employment (OECD 2016). Exemptions are given 0 in circumstances where it is impractical to apply V HR BI H EST NE KOS SRB BGR SVN MKD LVA ALB LTU SVK M the tax (e.g., finance and insurance services), for Source: IMF, Eurostat. equity considerations (e.g., health and education) or for activities that are traditionally a public service (e.g., postal services). In North Macedonia, the scope of reduced rates largely conforms to OECD standards, with some exceptions—products used in agriculture (e.g., fodder, fertilizers, agricultural equipment), baby products, school supplies, computers and software, solar heating systems, and pellet heating equipment. North Macedonia also exempts road tolls and funeral services from VAT, which is uncommon among OECD countries. For some of 79 The C-efficiency ratio is the ratio of actual to theoretical revenues from a perfectly enforced tax applied uniformly on all consumption. 76 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE these products, the reduced rate reflects a government policy—such as the preferential treatment of heating systems to support the environment—and may need to be considered within that context. In the longer term, a more comprehensive reform might expand the VAT base by removing the preferential treatment of goods and services that are not well-targeted to the poor. Indeed, the VAT is not an effective tool for redistribution, even though reduced rates for essential services lead to progressivity (OECD/KIPF 2014). Rich households generally consume more goods and services, and often more expensive goods, than poor households. Hence, in absolute value, richer households can benefit as much as or even more than poor households. Similarly, preferential treatment of goods considered socially desirable, such as books and cultural activities, may effectively result in a subsidy for their consumption by higher-income households, rather than lifting their consumption by lower- income households. A comprehensive reform could include removing the preferential treatment currently granted to publications, pharmaceuticals, hotel accommodation,80 although the VAT Directive does allow it. However, VAT treatment of these goods and services in non-EU countries varies.81 Removing the preferential treatment on some of the goods and services listed above could bring in an additional 2 MKD billion in VAT revenue, about 0.5 percent of GDP. Most of this additional increase comes from removing the preferential treatment on computers and software, hotel accommodation, and gambling. The estimate incorporates the reduction in the consumption of these goods and services from the increase in VAT, but does not include losses in revenue due to tax evasion. Table 4.5. Additional Revenue from Removing Preferential Treatment Consumption expenditure, Additional revenue, without Additional revenue, with excluding VAT (2014) behavioral changes (2016) behavioral change (2016) Products used in agricultural production Not available Computers and software* 5,439 761 706 Baby products and school supplies Not available Solar heating system Not available Pellet heating equipment Not available Road tolls Not available Funeral services Not available Publications 716 100 88 Pharmaceuticals 2,397 335 319 Hotel accommodation 7,315 1024 897 Source: North Macedonia Statistical Office, World Bank staff calculations. Note: The amount shown for computer and software includes spending on other electronic devices; Estimates of price elasticities of demand are based on those for EU member states or OECD countries and are sourced from Copenhagen Economics 2004, and Boeters, Böhringer and Büttner 2010. Where multiple estimates exist, the midpoint of the range has been chosen. Base-broadening can be coupled with compensation to low-income households to mitigate any regressive effects. The success of such large-scale reform relies, in part, on the design of the compensation package for different types of households. Designing a compensation package requires detailed information on the structure of household spending. North Macedonia’s Statistical Office conducts an annual household budget survey and regularly publishes aggregate results. However, the 80 Cultural services would generally be included in the list of goods. However, the EU VAT Directive states that cultural services are exempted in EU member states. Similarly, the directive states that gambling is exempt, but other OECD countries do apply VAT to the gambling margin, defined as the money collected by the operator minus the money paid out. 81 For example, Australia, New Zealand, Canada (other than pharmaceuticals), South Korea, and Singapore apply the standard VAT rate to these goods and services. The EU VAT design is considered to be a first-generation VAT system; those with more comprehensive coverage are considered second-generation. Chapter 4: Tax Policy Options for Increasing Revenue 77 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 size of the survey sample has fallen over time due to non-reporting by households, which has affected its reliability. Confidence in survey reliability would enable analysis of the redistributive effects of both the current VAT regime and any proposed reforms. One possible form of compensation is a cash transfer to the 30–40 percent poorest households. The government is considering a VAT refund that varies as a function of each household’s VAT paid. Specifically the government is considering returning to households 15 percent of VAT revenue collected (over 1 per cent of GDP in 2016), with a cap on the total amount to be returned to each household. However, because this would require data on household VAT receipts, it would be difficult to implement. Cross-country evidence shows that refund policies based on households’ own VAT paid are limited to specific situations—for example, aggregating the receipts to enter in tax lotteries82 and issuing refunds based on debit card consumption. Instead, a means-tested cash transfer to poor households that would limit administrative hurdles could be considered, hopefully as part of the efforts to introduce a guaranteed minimum income (GMI). Depending on the government’s fiscal needs, raising the VAT rate could be considered in the longer term, following base-broadening and improvements in tax administration. North Macedonia’s top statutory VAT rate is 18 percent. The VAT rate in most WB6 countries (Table 4.3) is about 20 percent; only Bosnia and Herzegovina has a lower rate, 17 percent. The authorities are not currently considering a rate increase. As with base broadening, a VAT rate increase can be packaged with transfers to poor households to limit the impact. North Macedonia’s VAT threshold83 is relatively low; the country should consider indexing it to inflation to maintain its value in real terms. In 2015, the registration threshold was reduced to MKD 1 million (about €16,000). When the VAT was introduced in 2000 the registration threshold was set to MKD 1.3 million and in 2010 it went up MKD 2 million. VAT registration thresholds relieve small businesses of the administrative costs associated with VAT compliance. Too high a threshold distorts competition between large and small businesses; too low it can discourage expansion. 4.3.2  Excise Duties In North Macedonia, excise duties are levied on a standard range of goods, namely motor vehicles, alcohol, tobacco, diesel fuel, and other mineral oils. In 2016, they brought in about 3.7 percent of GDP (Figures 4.9 and 4.10), mainly from tobacco (1.7 percent of GDP) and diesel (1.2 percent). Excise duties on tobacco have increased annually since 2012, in line with the scheduled annual increase in excise rates. Unlike the VAT which is a broad-based consumption tax, excise taxes are levied on specific goods considered harmful to health or the environment to discourage their consumption. 82 Tax lotteries are popular worldwide to increase compliance. The lottery of the state of Sao Paulo is one way to measure household consumption. To be eligible; consumers must log-in their receipts on a website, which then aggregates household consumption (Naritomi 2017. 83 Businesses with a turnover above the threshold must register for the VAT;. registration is voluntary for businesses below it. Registered businesses must collect VAT on the goods and services they sell and can claim VAT credits on what they buy. 78 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 4.9. Excise Revenue by Source, 2016 Figure 4.10. Tobacco Excise Revenue, 2007–16 Percent of GDP Percent of GDP 1.8 1.8 1.6 1.7 1.4 1.6 1.5 1.2 1.4 1.0 1.3 0.8 1.2 0.6 1.1 0.4 1.0 0.2 0.9 0 0.8 Other 7 8 9 0 1 2 3 4 5 6 Alcohol Tobacco Diesel mineral oils Motor vehicles 200 200 200 201 201 201 201 201 201 201 Source: MOF and World Bank staff calculations. North Macedonia’s excise tax rates on tobacco and some energy commodities are below the EU minimum requirements. This includes energy commodities such as motor fuels for noncommercial and nonindustrial uses, natural gas, coal, and electricity (Table 4.6). However, the rates on alcohol are above or close to the EU minimum. The Government has recently raised the excise tax on diesel (gas oil) and announced a schedule for future increases. According to the Government’s analysis, the new rates have not adversely affected the tax base. This harmonization with the EU will need to continue, taking into consideration the potential impact of excise increases on the black market. Table 4.6 shows the current gaps in excise tax rates for energy commodities. Introducing an excise on coal is strongly encouraged, given its adverse impact on the environment and the air quality of the country. Table 4.6. Excise Tax Rates, North Macedonia and the EU Minimum Fuel Rate (€ per 1,000 liters, unless otherwise specified) EU minimum North Macedonia Motor fuels Leaded petrol 421 390 Unleaded petrol 359 347 Gas oil 330 242 Kerosene 330 115 LPG (€ per 1,000 kg) 125 78 Natural gas (€ per gigajoule) 2.6 0 Motor fuels used for commercial and industrial use Gas oil 21 242 Kerosene 21 115 LPG (€ per 1,000 kg) 41 78 Natural Gas (€ per gigajoule) 0.3 0 Heating and electricity Gas oil 21 98 Heavy fuel oil 15 0 Kerosene 0 23 LPG (€ per 1,000 kg) 0 78 Natural gas (Euro per gigajoule) 0.15/0.3 0 Coke (€ per gigajoule) 0.15/0.3 1 Chapter 4: Tax Policy Options for Increasing Revenue 79 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table 4.6. Excise Tax Rates, North Macedonia and the EU Minimum (continued) Fuel Rate (€ per 1,000 liters, unless otherwise specified) EU minimum North Macedonia Coal (€ per gigajoule) 0.15/0.3 0 Electricity (€ per MWh) 0.5/1 0 Source: European Commission; North Macedonia Ministry of Finance (Customs Administration). Cigarette prices in North Macedonia are the lowest in the region (  World Bank 2019). In 2016, cigarette prices in Bosnia and Herzegovina and Bulgaria were about double those in North Macedonia. Cigarette price differences with Albania and Serbia also were large enough to encourage cigarette smuggling to those countries. In 2016, the tax share in the price of cigarettes in North Macedonia was lower than in any neighboring country except Albania. Table 4.7. Cigarette Prices and Taxes in North Macedonia and Neighboring Countries, 2016 Price of a 20-cigarette pack of Taxes as a % of price of the most sold brand the most sold brand Net-of- In Int'l dollars US$ at Ad Value tax price, reported Currency (purchas- official Specific valorem Total added Import Total US$ currency reported parity) ing power exchange excise excise tax/sales duties tax excise rates tax Albania 230.00 ALL 5.23 1.88 47.83 0 47.83% 16.7 1.81 66.3% 0.63 Bulgaria 4.90 BGN 7.94 2.78 28.57 38 66.57% 16.7 0 83.2% 0.47 Greece 4.00 EUR 6.62 4.45 41.25 20 61.25% 19.4 0 80.6% 0.86 Serbia 220.00 RSD 5.38 1.98 28.05 33 61.05% 16.7 0 77.7% 0.44 Bosnia and 4.50 BAM 6.61 2.56 26.67 42 68.67% 14.5 1.06 84.3% 0.40 Herzegovina North Macedonia 66.32 MKD 3.45 1.20 46.84 9 55.84% 15.3 0 71.1% 0.35 Source: WHO Global Tobacco Report, analysis by the World Bank Tobacco Tax Group. The currently scheduled increase in tobacco taxation will not be enough to meet EU obligations. Excise rates were stable from 2007 up to July 2013, when the specific rate was increased, while ad valorem rate was decreased. Additionally, a new minimum specific rate was introduced in 2013. Since July of 2016, both specific rates have gone up annually by 0.2 MKD per cigarette, and are scheduled to continue going up in July every year until 2023. Under this schedule, in 2023, the minimum excise rate for cigarettes should reach 3,453 MKD (about €57) per 1,000 cigarettes, but according to the EU tobacco taxation directive 2011/64, excise duty should not be less than €90 per 1,000 cigarettes no matter what the weighted average retail selling price is. North Macedonia can speed up the rise in tobacco excise rates to attain both public health and fiscal aims. Each of the following reforms would accomplish that: •• Increase the minimum specific cigarette excise rate in July each year by 0.5 MKD per 1 cigarette (so that by 2024 it reaches 5,753 MKD per 1,000 cigarettes (€90). •• In July each year, increase the simple cigarette excise rate by 0.25 MKD and the ad valorem cigarette specific excise rate by 1 pp. That will ensure that the minimum excise rate applies to most cigarettes with below average prices (85 MKD in 2018) and will ensure the growth of tobacco revenues by raising the excise burden for both cheap and expensive cigarettes. •• Set the specific excise rate for 1 kg of fine-cut and other kinds of smoking tobacco at the same level as the minimum specific excise per 1,000 cigarettes. 80 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE Such a taxation policy would gradually raise cigarette prices in North Macedonia, but they will still hardly exceed cigarette prices in the rest of the region because the current price difference is very large and at least some neighboring countries will be raising their own tobacco excise rates over the next years. Increases in excise taxation can certainly encourage the black market, but that could be minimized by stronger enforcement by the customs administration. Indeed, in 2014 customs administration adopted a new strategy to reduce the size of the black market and the incidence of smuggling.84 4.4  Corporate Income Tax North Macedonia’s CIT has a low rate of 10 percent and substantial investment incentives, such as a deduction for reinvested earnings. These two factors lead on average to low ETRs85 compared to EU countries (Figure 4.11).86 This can encourage multinationals to locate their subsidiaries in North Macedonia (Devereux and Griffith 2003). The low rates and generous incentives also result in low revenue collection—in 2016, profit taxes (which includes the CIT and the turnover tax paid by small businesses) contributed a mere 6 percent to total revenues. After the financial crisis the contribution had been even lower (Figure 4.12), as North Macedonia temporarily applied a cash flow tax where the tax base consisted of profits distributed to shareholders in order to provide tax relief to businesses. North Macedonia reverted to a classical CIT regime in 2015, which brought in more revenue. Figure 4.11. Effective Average CIT Rates Compared, Figure 4.12. Profit Tax and Net Operating Surplus, 2016 2006–16 Percent Percent of total revenue Percent of GDP 25 8 39 7 37 20 6 35 15 5 33 10 4 31 3 5 29 2 0 1 27 ia ia ia via nia ia ia 8 ia gar Croat Eston Lat ithua vak loven EU2 cedon Bul L Slo S a 0 25 M rth 200 6 200 7 200 8 200 9 201 0 201 1 201 2 201 3 201 4 201 5 201 6 No ÂÂProfit tax, percent of total revenue ---- NOS, percent of GDP Source: Zew (2016). Source: MOF and World Bank staff calculations. Although not currently considered by the Government, North Macedonia could increase its CIT rate by up to 5 pp. This would allow the country to collect more revenue while remaining a competitive destination for foreign investment. In particular, the deduction for reinvested earnings mitigates the effects on investment of any CIT rate increases. Firms can immediately expense investment, which is funded from retained earnings, so the marginal investment decisions of firms that have sufficient retained earnings are not affected by the CIT rate. Indeed, for firms that do not need to issue new equity 84 http://www.customs.gov.mk/images/documents/izvestai/godisni-izvestai/EN/annual-report-2013-EN.pdf 85 The effective average tax rate shows the proportion of pre-tax returns from an investment that accrue to the government. It is affected by the statutory CIT rate, allowable deductions, and how the firm is financed. 86 Comparator countries are based on data availability, regional proximity ,and structural similarities (e.g., population lack of significant natural resources). Chapter 4: Tax Policy Options for Increasing Revenue 81 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 to finance investment, the CIT regime becomes equivalent to a cash flow tax where only economic rents or above-normal returns are taxed. A CIT rate increase might affect the location decisions of multinationals that earn economic rents (i.e., firm-specific rents such as those from patents). However, the effects of the increase would be moderated by the fact that the rate would remain comparable to those levied by regional peers whose rates range from 9 to 15 percent. Raising the CIT rate to 13 percent would increase revenue by about 0.4 per cent of GDP, based on a rough calculation that incorporates some firm responses to this increase.87 Several surveys suggest that a country’s tax burden is not the main factor in business investment decisions. The CIT rate matters for the investment climate and for the incentives for multinational firms to allocate their taxable income domestically or offshore. However, according to a World Bank survey of firms investing internationally, national taxes rank 11th in the top 20 most important factors affecting location decisions (Gómez-Mera,2014). Similarly, according to the 2013 World Bank enterprise survey only 8.7 percent of firms operating in North Macedonia think tax rates are an obstacle (Figure 4.13). While the CIT rate does matter for foreign investors, non-tax considerations like the stability of the economy, the size of the market, labor costs and labor market flexibility, infrastructure, and ease of doing business in general are considered more important. Given North Macedonia’s favorable score, 11th, on ease of doing business ranking there is scope to increase the CIT rate without jeopardizing its competitiveness as an investment destination. Changes to the PIT and CIT regimes need to be Figure 4.13. Top Business Environment Obstacles for Firms Operating in North Macedonia coordinated to lower incentives to misreport Practices of the informal sector labor income and to minimize distortions in the Access to finance choice of legal business entity. The tax treatment Political instability of business income depends on whether the Tax rates business is incorporated (taxed as CIT) or not Inadequately educated workforce incorporated, such as a partnership or sole trader Electricity (taxed as PIT). That is, the CIT is often considered Courts a backstop for the PIT. Currently, the high tax Tax administration burden on labor income (from PIT and SSC) and Customs and trade regulations Corruption the low burden on businesses gives individuals an incentive to supply their labor via a corporation 0 5 10 15 20 25 30 35 Percent and under-report their labor income. Source: World Bank Enterprise Survey. 4.4.1  Technological Industrial Development Zones North Macedonia has used a range of incentives within special economic zones, known as Technological Industrial Development Zones (TIDZ), to attract foreign direct investment into the country. The cumulative investment by firms operating in TIDZ between 2011 and 2015 was about €207 million and the firms employed about 6,800 people—about 1 percent of total employment in 2016 (OECD 2017). Using a broader definition of foreign companies that receive state aid—not just those located in TIDZ—raises the employment estimate to about 20,000 people.88 The additional investment 87 We use a taxable income elasticity of –0.3. De Mooij and Saito (2014 ) reviewed estimates for the elasticity of corporate taxable income and found that most lie between –0.1 and –0.3. 88 Government of North Macedonia estimates, http://vlada.mk/node/13391, accessed 25 April 2018. 82 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE by TIDZ firms has translated into an increase in exports over time. In 2015 net exports of goods from TIDZ reached 3 percent of GDP, about 42 percent of total exports (National Bank 2016). Compared to peers, North Macedonia offers the widest range of tax and non-tax incentives in special economic zones  (OECD 2017). Tax-related incentives in North Macedonia include exemptions from the CIT (for ten years), the PIT, the VAT, customs duties, utility taxes, and building permits. Non-tax incentives include concessionary land leases, grants for construction, free connection to utilities, grants for employee training, and access to support services (e.g., assistance in acquiring work permits). In contrast, other countries mainly offer tax exemptions. The strategy of offering incentives to attract FDI is not sustainable in the long term. The EU accession process restricts the size of the incentives that candidate countries can grant, with restrictions varying depending on the size of the firm and GDP per capita of the region (EC 2014). EU member states are further restricted (OECD 2017) in the type of incentives they can grant (e.g., direct labor subsidies are generally prohibited). Competition between Western Balkan countries to attract FDI using special economic zones has intensified with the number of such zones in the WB6 rising markedly between 2011 and 2015 (OECD 2017). This race increases the fiscal costs of attracting FDI using this strategy. A cost-benefit analysis is necessary to determine whether supporting TIDZ is an efficient use of public revenue. These incentives have fiscal costs for the government in terms of foregone tax revenue, grants, administrative and compliance costs, the distortionary effects of raising revenue, and the distorted resource allocation caused by the incentive. The cumulative direct fiscal cost of TIDZ to North Macedonia between 2007 and 2014 has been estimated at about €66 million, 0.8 percent of GDP (the shortage of data creates enough uncertainty that the estimate is likely to be underestimated; CEA 2016). The government has estimated the cost of attracting FDI through cash grants and tax exemptions for foreign firms locating in the TIDZ and elsewhere to be about €225 million.89 More generally, there are additional concerns beyond the direct costs associated with tax incentives. A study by the World Bank, the IMF, the OECD and the UN has identified the following concerns (World Bank et al. 2015): •• Redundancy. The incentive may merely be subsidizing investment that would have occurred even without the incentive—taxation is not the main factor in business investment decisions. •• Displacement. The incentive may be replacing other investment that the firm would have undertaken without the incentive. Economy-wide, with less than perfect access to capital markets, sector- specific incentives may merely encourage investment by some sectors at the expense of others. Moreover, FDI can displace domestic investment, for example, when it merely transfers ownership from a domestic firm to a multinational. •• Ineffectiveness. Such incentives may be ineffective if the multinational’s home country taxes on a worldwide basis. In such systems, multinationals are taxed based on worldwide income and receive a tax credit for taxes paid overseas. Hence, any increase in tax incentives applied to a multinational can increase their tax liabilities in their home country, though the effect is mitigated if multinationals can defer taxes until income is repatriated. 89 Government of North Macedonia estimates, http://vlada.mk/node/13391, accessed 25 April 2018. Chapter 4: Tax Policy Options for Increasing Revenue 83 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 •• Creation of opportunities for abuse and corruption. Firms can shift taxable income (e.g., through transfer pricing) to a related entity that qualifies for the incentive. •• Higher tax administrative costs and complexity. This and requires solid tax administration capacity to monitor and prevent abuse of incentives. •• Only temporary economic gains. Temporary incentives, such as tax holidays, can bias investment toward ‘footloose’ firms that have low upfront costs and that leave the country once the tax holiday is removed. In contrast, these incentives are not utilized by firms with high start-up costs, such as new enterprises, since they may not be profitable until after the holiday period. •• Mixed effectiveness. The effectiveness of tax incentives is low for developed countries and for FDI looking to exploit a natural resource (e.g., mining or forestry); access the local market; exploit other local strategic assets, such as know-how or technology. Tax incentives are more effective for firms seeking to serve the global market from a low-cost production base. The benefits of investment incentives may include additional investment, job creation, and spillovers; the extent to which North Macedonia has realized these benefits is mixed. Whether tax incentives are effective in attracting investment varies by country and sector. Tax incentives are generally less effective in developing countries, where poor investment conditions (e.g., inadequate infrastructure and poor governance) dominate investment decisions (World Bank 2015b). Such incentives are more effective when they are geared to attracting firms that are looking for a low-cost production base and exporting firms (World Bank 2015b ). North Macedonia’s TIDZ incentives would fit into this latter, more effective, category given its comparative advantage in low-cost skilled labor, its geographic proximity to large markets, and the export requirements for TIDZ firms. In terms of job creation, there is some evidence that FDI has improved employment prospects in the Balkans (Atoyan and Jankulov 2015). On the other hand, there is anecdotal evidence that spillovers from TIDZ firms have been few. Domestic firms have struggled to integrate into the supply chain of foreign firms because they cannot meet technical and safety requirements (IMF 2015). However, there is some positive evidence of cooperation between TIDZ firms and local training and educational institutions (OECD 2017). 4.5  Taxation of Immovable Property Taxation of immovable property is mainly levied by the municipality. Revenue from these taxes is low, about 0.45 percent of GDP, of which 0.25 percent is collected from the transfer of immovable property, with the rest from an annual property tax. Property tax rates levied by municipalities range between 0.10 to 0.20 percent, with most at the lower end of the range (Chapter 3 for a full discussion of local government taxation). The government is considering increasing the revenue collected from property tax by raising the rate for higher-valued properties. Depending on the threshold chosen, the tax increase is likely to be paid mainly by high-income households. However, because the tax is not triggered by a flow of income, its impact on liquidity-constrained households must be considered. For example, retirees may own high–value assets but have low streams of income, and hence might find it difficult to pay the tax. 84 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE Accurate estimation of the tax base is a major challenge to the efficiency and equity of property tax reform. In practice, it might require an update of the cadaster and potential improvements to the valuation method. While expanding the tax base could generate revenue gains as large as the tax rate increase being considered, it presents an important administrative challenge. Obtaining real time data on property values requires frequent transactions, which is possible only in Skopje. Also, the current renewal of the cadaster will not be completed in 2019. Another option for increasing the taxation of immovable property is to lower the scope of PIT discounts for capital gains. Capital gains are discounted—or even exempted—for owner-occupied property or property held for more than five years. Discounts for capital gains are intended to correct for the effects of price inflation, so that the tax applies only to real, not nominal, gains. On this criterion, and given North Macedonia’s low inflation rate, the discounts seem generous. The primary residence commonly receives favorable tax treatment because it is a form of lifetime saving. North Macedonia’s treatment of capital gains is more favorable than that of its peers. Furthermore, countries with a low, flat PIT rate such as Montenegro, Kosovo, and Bosnia and Herzegovina do not offer discounts or exemptions. Those countries with higher rates generally exempt the Capital Gains Tax (CGT) only after 10 years. A potential reform option might be to extend the length of time a property must be held before qualifying for the CGT exemption from 5 years to 10. Table 4.8. PIT Treatment of Capital Gains from the Sale of Real Estate Country CGT rate (%) Discounts and Exemptions Discounted by 30 percent if the seller lived in the property for one year (exempt if the seller who has lived in the property for one year then MKD 10 sells it three years acquisition; of if the property is sold more than five years from acquisition. HRV 24 Exempt after three years Exempt after 10 years (If the gain is reinvested in housing, it is not SRB 15 subject to tax.) MNE 9 No discounts or exemptions ALB 15 No discounts or exemptions KOS 10 No discounts or exemptions BIH 10 Exempt if property is used for personal purposes LVA 15 Private residence exempt under certain circumstances BGR 10 Any one residential property in each year (exempt after five years) SVK 25 (top PIT rate) Exempt after 5 years EST 20 No discounts or exemptions Exempt after 10 years; exempt if individual lived in the property for two LTU 15 years or lived in the property for less than two years but uses the gain to purchase another residence SVN 25 Discount increases over time so that it is exempt after 20 years Source: Deloitte. Chapter 4: Tax Policy Options for Increasing Revenue 85 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 4.6  Tax Compliance and Administration The size of the informal sector in North Macedonia is estimated at 25–40 percent of GDP, depending on the method used.90 International comparisons suggest that informality is a few points higher than in peer countries (see Dzehkova et al. 2014 for a review of the informal sector in North Macedonia and Schneider et al 2010 for a comparison of informal sectors in 162 countries using the MIMIC method). However, indirect estimation suggests that the informal sector has slightly narrowed in recent years. Between 2008 and 2013 self-reported informal employment from the Labor Force Survey dropped from 29 to 22.5 percent. In the same period, in the World Bank Enterprise Surveys the number of firms that report competing against informal firms shrank by 30 percent. Still, in 2013, 55 percent of firms reported competing against informal firms, and 35 percent considered informal competition to be a major barrier to their growth. Thus the Government should continue its efforts to formalize the economy by giving the PRO the tools to achieve its revenue collection goals. The reduction in informality could be due to a combination of low PIT and CIT rates and improvements in tax administration. In its 2016 assessment of the Public Revenue Office, the IMF reported encouraging results: PRO’s results and strategy are transparent, and it has put in place such modern taxpayer services as electronic payment, e-services to business, withholding, and advance payments. PRO is planning to further modernize tax services by introducing pre-filling of tax returns, data cross- checks, and electronic data exchange. These advances are expected to bring in more revenue. PRO is also tackling some shortcomings the IMF assessment identified91: it is automating tax refunds and building a complete taxpayer database with a risk analysis tool. One remaining constraint concerns the personnel—PRO is under-staffed and receives requests that are not directly related to its core mission of tax collection. Moreover, 40 percent of the PRO’s 220 inspectors will retire by 2020, which will put significant stress on the administration. Addressing personnel issues and increasing the funding of the PRO are desirable investments in the efficiency and equity of the tax system. This seems particularly relevant as tax reforms are being considered. For example, raising the top PIT rate could encourage tax evasion and tax avoidance. Limiting exemptions and special regimes could have rewards by also limiting the possibilities for tax evasion and income-shifting. 4.7  Reform Options Short Term 1. Remove the tax exemption on the sale of securities (planned for the end of the year), as is international best practice. Removing this exemption from the PIT base would help to balance the relative tax burden between labor and capital. Currently, the ETR on labor income is much higher than on capital income. While the higher elasticity of capital provides some rationale for 90 Indirect methods use differences in macro-aggregates to estimate the magnitude of the informal sector. A common method compares growth in electricity consumption with growth in GDP. These estimates should not be taken at face value but as a noisy approximation of the size of the informal sector. 91 The main shortcoming identified were delays in issuing tax refunds, low collection of tax arrears, and auditing limitations (assessment of compliance risk, taxpayer database coverage, and the dispute resolution process) 86 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE a higher tax rate on labor, the current gap in North Macedonia is large and may lead to economic distortions. 2. Introduce a higher tax bracket for the PIT, with a top marginal tax rate above 15 percent. That would both increase revenue and introduce greater progressivity into the PIT system. The distortionary effects of higher taxation are likely to be minimal: North Macedonia currently has a low PIT rate and a low ETR on capital income compared to peer countries. 3. Consider broadening the capital income tax base and raising the standard PIT deduction to make the PIT more progressive and realign the ETRs on labor and capital. These base-broadening measures, along with increasing the top marginal tax rate would substantially increase tax revenue. 4. North Macedonia’s schedule of goods and services that receive preferential VAT treatment generally adheres to the EU VAT Directive. However, there is scope to expand the VAT base by removing preferential treatment for selected items, and combining this with a package of cash transfers to compensate low-income households for the increase in living costs. Applying a mix of standard and reduced rates creates opportunities for abuse, while a broad-based VAT improves economic and administrative efficiency. In many countries, the VAT is an ineffective redistributive tool even though reduced rates for essential services lead to progressivity. 5. Accelerate the increase in excise taxes on cigarettes and energy to meet EU requirements. In particular, cigarette prices in North Macedonia are the lowest in the region. Speeding up the increase in tobacco excise rates is a way of meeting both public health and fiscal aims. 6. Conduct a tax expenditure review for the PIT, CIT and VAT to clarify the revenue cost of exemptions, incentives, and special regimes. An accurate estimate of tax expenditures is the first step to analyzing the effectiveness of various incentives and improving budget transparency. 7. Implement the recommendations of the TADAT assessment to modernize tax administration. That assessment reported encouraging progress but identified such challenges as personnel constraints. Investing in the PRO is particularly important at a time when major tax reforms are being considered. Medium Term 8. Conduct a cost-benefit analysis of special economic zones to determine whether supporting TIDZ is an efficient use of public revenue. 9. Review cadaster valuations and increase revenue from the property tax. Accurate estimation of the tax base is a major barrier to efficient and equitable property tax reform. 10. Assess the efficiency, tax revenue, and equity costs of PIT and VAT reforms. A specific risk to assess is how a higher marginal PIT rate impacts tax avoidance and tax evasion. This analysis would help inform future tax policy proposals. Chapter 4: Tax Policy Options for Increasing Revenue 87 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Long Term 11. Bring the CIT rate closer to 15 percent. This would keep North Macedonia’s rate in line with peers, increase revenue, and rebalance the tax burden between capital and labor. While higher CIT rates might discourage investment by multinationals, several surveys suggest that a country’s tax burden is not the main factor in business investment decisions. 12. Review which CIT incentives to retain, based on the tax expenditure review, the cost-benefit analysis of special economic zones, and the availability of non-tax instruments to achieve similar goals. 13. Increase the tenure to qualify for capital gains exemption of owner-occupied housing to 10 years or more to bring it closer to EU practice and increase the share of revenue from property taxation. 88 Chapter 4: Tax Policy Options for Increasing Revenue SOWING THE SEEDS OF A SUSTAINABLE FUTURE Chapter 5: Ensuring the Sustainability of the Pension System Pension reforms are needed to contain the fiscal pressures in North Macedonia caused by population aging. By international standards, North Macedonia spends a comparatively large share of GDP on pensions, which is especially a problem because its population is still relatively young. The high pension spending is due to the transition from a centrally-planned to a market economy, which has resulted in a large number of beneficiaries relative to contributors, relatively generous PAYG pensions, and arbitrary changes in the indexation rules over the last decade. In the next two decades North Macedonia’s rapidly aging population will put additional pressure on the pension system’s sustainability. Without reforms, it will continue to generate a sizable deficit. Over the past decade the government has had several successful rounds of pension reforms but more work is still needed. 5.1 Introduction After several changes in the past two decades, North Macedonia’s pension system now rests on three pillars: (1) a pay-as-you-go (PAYG) pillar; (2) a second, fully funded, mandatory pillar based on individual accounts; and (3) a voluntary private open and occupational pension fund. The PAYG pillar has been reformed several times over the years to make it more fiscally sustainable and to reflect demographic changes, as the old-age dependency ratio—the number of people aged 65 or older to 100 people in working age—went up from 8.2 percent in 1991 to 11.8 percent in 2014. Between 1993 and 1997, the Government slowed pension increases by introducing below-wage indexation, raising the retirement age from 60 to 63 years for men and from 55 to 60 for women, widening the calculation period, and limiting early retirement options to hazardous occupations. In 2005, the retirement age was further risen, to 64 for men and 62 for women, accrual rates were reduced, and Swiss indexation (50 percent to wages and 50 percent to the consumer price index (CPI) was introduced, which further slowed increases in pension spending. In 2006, a mandatory defined contribution component was added to the system with introduction of the second pension pillar, which became mandatory for new labor market entrants. A voluntary third pillar was launched in April 2009. Parametric changes to the system in 2012 reduced the net accrual rate for post-2013 service years among PAYG-only system participants, from 1.8–2.6 percent, depending on gender, to 1.6–1.8 percent. These changes also tightened the criteria for disability pensions. Table 5.1 shows the current basic parameters of the reformed pension system in North Macedonia. Chapter 5: Ensuring the Sustainability of the Pension System 89 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table 5.1. Pension Parameters in North Macedonia Retirement Age 64 M/62 F with 15 years of contributions Service Period for: Until 2001 2001–2012 2013 Non 2nd pillar members with less 2.33% - men 1.8% - men 1.6% - men than 15 years of 2.60% - women 2.05% - women 1.8% - women service Non 2nd pillar members with more Declining (art 228) Declining (art 228) 1.61% - men 2.3%–2.0% - men 2.3%–2.0% - men Accrual rate (Net) than 15 years of 1.61% - women 2.6%–2.1% - women 2.6%–2.1% - women service 2nd pillar members From 2003 2.33% - men 0.75% - men who opted in the 0.75% - men 2.60% - women 0.86% - women system 0.86% - women 2nd pillar members From 2003 0.75% - men who mandatorily 0.75% - men 0.86% - women joined the system 0.86% - women For PAYG-only participants: 79.4% in 2017, declining to 72% in 2040. For voluntary 2nd pillar participants: 79.4% in 2017, declining to 72% in 2040, but with Maximum net maximum PAYG net accrual rates for service before 2006 of 11.65% (men) and 13% accrual rate (women). For mandatory 2nd pillar participants: no maximum PAYG accrual rate. Pensionable base Career earnings, net wagse revalued (valorized) with historical average wage growth 35%–41% (scaled in proportion to service period) of 2002 average net wage indexed the Minimum pension same as regular pensions Maximum pension 2.7 * average net wage in North Macedonia in previous calendar year. Indexation post- “Swiss” formula: 50% average gross wage rate of change plus 50% CPI rate of change. retirement Professional incapacity: at least 50% reduced work capacity, minimum age of 45, and 1/3 of contribution period above the age of 20 (26 for tertiary education); Eligibility for disability pension General incapacity: 80–100% incapacity, minimum age of 20 (26 for tertiary education), and at least of 1/3 of contribution period above age of 20 (26 for tertiary education); for injury at work, regardless of service period. 80% of the old age pension for full incapacity in case of work injury and professional disease; declining to 72% by 2040; Disability pension Professional incapacity: 38% men/ 44% women is the replacement rate upon attainment of age; pension supplements of 10–20% for 80–100% incapacity. Age 50 for widow Eligibility for Age 55 for widower survivor’s pension Children aged 15 or under and until age 26 if students or lifetime if incapacitated Survivor’s pension 70% of contributor’s entitlement if 1 person; 80% if divided by 2; 90% if divided by 3; and 100% if divided if 4 or more. Second-pillar New entrants (regardless of age) since 2006 mandatory, voluntary opt-in for others. participation Total: 18%; Pension PAYG: 12%; contribution rate Second Pillar: 6%. Source: Pension and Disability Insurance Fund (PIOM); Ministry of Labor and Social Policy (MLSP). 90 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE 5.2 Recent Developments and Features of North Macedonia’s Pension System Pension system finances have worsened in the last decade, and the initial successes of the 2005– 06 reforms in stabilizing the pension deficit have been reversed. A reluctance to undertake further reforms, ad hoc indexations, and simultaneous reductions in contribution rates have resulted in pension deficits rising from 2.9 percent in 2007 to 4.6 percent in 2016 (Figure 5.1). Overall and current deficit trends92 show that this is due more to booming expenditures and less to falling contribution revenues. North Macedonia’s pension system is notably expensive considering its current demographic structure. While the population is relatively young, with the lowest old-age-dependency ratio among peer countries, its spending on pensions as a share of GDP is in line with that of older EU-28 countries (Figure 5.2). The country’s high pension spending is due to several features of its pensions system that are discussed below. However, the relatively young current structure of the Macedonian population notwithstanding, falling fertility rates and population aging are signals that further pension reforms are needed to assure the viability of the pension system. Indeed, the share of the population above the retirement age is projected to double by 2050 (Figure 5.11). Figure 5.1. Pension System Finances, 2001–16 Figure 5.2. Total Pension Spending as a Share of GDP and Old Age Dependency Ratio, 2015 Percent of GDP Percent 12 20 18 GRC 10 AUT ITA 16 DNK FRA BEL PRT 8 14 NLD CHE FIN 12 SRB ESP DEU LUX GBR MNE NOR HRV SWE 6 10 MKD SVN CZE EST ALB BIH BGR 8 ROU SVK ISL LVA 4 HUN LTU 6 IRL MLT 2 4 2 0 0 1 3 5 7 9 1 3 5 6 200 200 200 200 200 201 201 201 201 15 20 25 30 35 40 ÂÂOverall pension expenditures ÂÂContribution revenues Old age dependency ratio (65+/15–64) ÂÂOverall pension deficit ÂÂCurrent PAYG pensions deficit Source: PIOM and State Statistical Office of North Macedonia. Source: UN Population Prospects 2017 Revision and World Bank staff notes. The pension reforms over the past decade have markedly improved pension system dependency. Erosion in the system dependency ratio in North Macedonia,93 as in other transition economies, started with the transition from a centrally planned full-coverage system to a market economy. Massive job destruction in the early 1990s pushed the ratio down from more than three contributors to one pensioner before the transition to almost one. Following pension reforms of 2005 and 2006, the ratio has improved. At the end of 2016, the Macedonian PAYG system served 305,766 beneficiaries and 570,168 insured individuals. Each pensioner is currently supported by 1.86 contributors, up from 1.31 in 2005 (Figure 5.3). In addition to reforms that broadened labor participation, improvements in pension system dependency trends have been achieved by raising the retirement age, abolishing 92 The current deficit is defined here as pension benefit payments minus PAYG pension contributions. The overall deficit incorporates other PAYG system spending, i.e., health contributions for pensioners and transfers of second-pillar contributions to the second pillar pension fund. 93 System dependency is defined here as a ratio of the number of contributors to the number of pension beneficiaries. Chapter 5: Ensuring the Sustainability of the Pension System 91 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 early retirement, and tightening disability conditions. As a result, the system dependency ratio in North Macedonia has surpassed levels in many transition and EU countries (Figure 5.4). Nevertheless, pension reforms have been unable to prevent or curb the deterioration in pension system finances. The reform process is incomplete. Figure 5.3. North Macedonia – Insured Individuals, Figure 5.4. System Dependency Ratios Compared, 2013 Pensioners, and Pension System Dependency Ratio, (2016 for North Macedonia) 2001–16 Thousands 600 1.9 3.0 500 1.8 2.5 1.7 2.0 400 1.6 1.5 300 1.5 1.0 200 1.4 0.5 100 1.3 0 DNK ROU HRV BGR SVN PRT HUN FRA EST ITA GRC LVA Euro area FIN DEU CZE AUT POL BEL SVK MKD ESP MLT NLD NOR SWE LUX IRL CYP 0 1.2 1 3 5 7 9 1 3 5 6 200 200 200 200 200 201 201 201 201 JJInsured individuals, lhs JJPensioners, lhs ÂÂSystem dependency ratio, rhs Source: PIOM, Annual Report 2017. Source: EU Aging Report 2015. North Macedonia’s mandatory retirement age is still among the lowest in Europe, and it is gender specific. Though reforms were introduced in North Macedonia earlier in the decade, retirement ages are only 64 for men and 62 for women, and in most EU countries, retirement for both is the same. At first glance, pension reforms in North Macedonia should feature a raise of the retirement age. However, there are factors that need to be taken into consideration when determining retirement age increases, and their gender dimensions. One is life expectancy at retirement—15.3 years for men and 18.9 for women—which is lower in North Macedonia than in most peer countries (Figure 5.5). This suggests that a policy of raising retirement ages need not be aggressive but simply aimed at maintaining a sustainable ratio of the active contributory period to retirement years. North Macedonia has reduced opportunities for early retirement and tightened eligibility criteria for disability pensions. However early retirement options still exist for hazardous occupations. General early retirement, common in other countries in the region and in the EU, was abolished in North Macedonia in 2005 (Table 5.2). A potential for early retirement through disability was tightened with the introduction of stricter disability eligibility criteria. General disability eligibility was limited to minimum age of 45 years, one-third of paid contributions for service above 20 years, and a minimum incapacity level of 50 percent. (Figure 5.6 for the positive impacts of these measures). Until 2002, disability pensioners represented about 21 percent of the total. In 2016, that figure was less than 13 percent. The only remaining early retirement option is the extended service period for hazardous occupations, which awards 14–18 months (depending on hazard levels) for each 12 months worked and reduces their retirement age accordingly. However, the supplementary contribution rate for “18 for 12 months” occupations is only 9 percent—well below its actuarial balance. The idea of mandatory early retirement in hazardous occupations, its current financing arrangement, and the list of occupations eligible for it through the extended service period (“beneficirani staz”) is being reviewed, and expected to be reformed in 2019. 92 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 5.5. Remaining Life Expectancy at Retirement Male Years 24 22 20 18 16 14 12 10 BEL GRC ESP ITA FRA LUX MLT IRL SWE DNK NLD PRT GBR AUT DEU FIN NOR LTU LVA ALB RUS BLR ARM BGR HUN ROU CZE SVK POL HRV SRB MKD BIH KAZ KGZ AZE TUR Female Years 28 26 24 22 20 18 16 14 12 10 L P A A L D R R A R S R B L D B R GRC MLT BE ES IT FR LUX IR PRT SWE NL DNK NO DEU GB FIN AUT LTU LV BG ARM ROU HUN CZE HRV SVK RU BL AL PO MK SR BIH KAZ KGZ AZE TU Source: Inverting the Pyramid, World Bank, 2014; HSTC= high spending transition countries; YC= “young” countries. Figure 5.6. North Macedonia: Categories of Pension Beneficiaries, 1991–2016 Thousands 200 180 160 140 120 100 80 60 40 20 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 199 199 199 199 199 199 199 199 199 200 200 200 200 200 200 200 200 200 200 201 201 201 201 201 201 201 JJOld-age JJDisability JJSurvivors Source: PIOM. Chapter 5: Ensuring the Sustainability of the Pension System 93 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table 5.2. Regular, Early, and Late Retirement in OECD, EU, and FYR Countries Regular retirement age Early retirement age Early retirement Late retirement Annual Men Women Men Women without decrement bonus decrement Australia 67 n.a. n.a. n.a. Austria 65 62 60 4.2 4.2 Belgium 65 60 0 35 yos 0 Canada 65 60 7.2 8.4 Chile 65 n.a. n.a. n.a. Croatia 65 60 1.8–4.2 41 yos 1.8 Czech Rep 65 62–65 60 59–60 2–6 6 Denmark 67 n.a. n.a. 5.6 Estonia 63 60 4.8 10.8 Finland 65 62 7.2 4.8 France 65 56–60 5 41 yos 5 Germany 67 63 3.6 45 yos 6 Greece 65 55–60 6 37 yos 0 Hungary 65 63 3.6–4.8 6 Iceland 67 62 7 6 Ireland 66 65 n.a. n.a. n.a. Israel 67 62 n.a. n.a. Italy 65 60 any age/61 2.3–2.9 2.6–2.9 or 0 Japan 65 60 6 8.4 Korea 65 60 6 6 Luxembourg 65 57/60 0 n.a. Mexico 65 60 0 0 North Macedonia 64 62 n.a. n.a. n.a. n.a. 40% (new Montenegro 67 62 62 4.25 40 yos service >40) Netherlands 65 n.a. n.a. n.a. New Zealand 65 n.a. n.a. n.a. Norway 67 62 3.8–4.7 4.9–5.4 Poland 65 60 n.a. n.a. 4.3 Portugal 65 55 6 4–12% Rep Srpska 65 60 58 0 Serbia 65 60.5 54–60 53–60 4.1 40–45 yos 0 Slovak Rep 62 60 6.5 6.5 Slovenia 63 58 1.2–3.6 1.2–3 Spain 65 61 6.0–7.5 2.0–3.0 Sweden 65 61 4.1–4.6 4.9–6.1 Switzerland 65 64 63 62 4.5 5.2–6.5 Turkey 65 n.a. n.a. 0 UK 68 n.a. n.a. 10.4 USA 67 62 5.0/6.7 8 Source: World Bank Pension Database, OECD, Eurostat. 94 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE While accrual rates in North Macedonia are cohort and gender-specific, they are more generous for older cohorts than in peer countries. The calculation of pension benefits in North Macedonia rests on the traditional pension accrual formula.94 A comparison of effective gross PAYG accrual rates in North Macedonia and the EU95 (Table 5.3) shows that for non-switchers (individuals in the first pillar only), North Macedonia’s PAYG benefit levels are slightly below the EU average but above those of most new EU member states. Most EU countries have an accrual formula award period- and gender- universal pension accruals. North Macedonia, however, links the accrual determination to specific cohorts, gender, and contribution periods. All second-pillar participants, voluntary and mandatory, have PAYG accruals that are far less than their share of paid-in contributions.96 The objective of setting such accrual rates at the start of the 2005 reforms was to have a smooth transition from the PAYG system to a multi-pillar system, and for effective accrual rates to gradually reach sustainable levels. However, the outcome has been the opposite—large differences in initial pension levels between PAYG only, voluntary and mandatory multi-pillar members, and a fiscally unsustainable PAYG system. Pension accruals that were reduced more than proportionally for voluntary switchers have resulted in first multi-pillar pensions significantly below the old first pillar PAYG pensions. The pension benefits of first multi-pillar pensioners (about 100 new pensioners by year-end 2017)97 have been reduced by 20 to 60 percent, depending on the length of their service.98 This socially unacceptable reduction, which pensioners did not expect, has led to dissatisfaction with the reform process, and triggered immediate attention and action from the Government.99 Table 5.3. Effective Gross Accrual Rates, EU and North Macedonia, 2013–60 Country 2013 2020 2030 2040 2050 2060 Austria 1.2 1.1 1.2 1.2 1.1 1.1 Belgium 1.5 1.5 1.4 1.4 1.4 1.4 Bulgaria 1.1 1.2 1.2 1.2 1.2 1.2 Cyprus 1.3 1.3 1.3 1.3 1.3 Czech Republic 1.8 2.0 1.8 1.7 1.9 1.9 Estonia 0.6 0.6 0.5 0.4 0.4 0.3 Finland 2.9 2.3 1.9 1.9 1.9 1.9 France 1.7 1.7 1.7 1.7 1.7 1.7 Greece 2.0 1.8 1.4 1.4 1.4 Hungary 2.1 2.0 2.0 2.0 2.0 2.0 Italy 1.9 1.8 1.7 1.7 1.7 1.7 Latvia 1.1 1.0 1.0 0.8 0.7 0.6 94 Pension accrual represents a percentage of the individual's pension base earned for one year of pensionable service with paid contributions. A pension accrual of 1.5, for example, indicates that a person gets 1.5 percent for each year of service. The base is commonly lifetime earnings, revalued (valorized) to current prices. 95 Macedonian accrual rates legislated and applied in net terms (Table 5.1), i.e. as fraction of net wage base “accrued“ for retirement. Gross accrual rates are obtained by multiplying net rates by the ratio of net to gross wage. 96 Voluntary and mandatory second pillar participants pay 2/3 of their pension contribution to the PAYG pillar. As Table 5.1 shows, in all periods, projected accrual rates for these categories has been less than 2/3 for PAYG-only participants. 97 The first pensioners that in 2016 retired from both pension pillars are those who voluntarily switched into the new system in 2003 and accepted the new rules accepting that opt-back would not be possible. 98 Three factors contribute to the differences (1) reduced pre-2003 PAYG accrual rates for voluntary participants (accrual cap of 13% for women and 11.65% for men); (2) more than proportionately reduced PAYG accrual for the service period after 2013; and (3) how the programmed withdrawals formula is applied to men and to women. 99 Several retirees have filed or are preparing to file lawsuits claiming they were misled by either the reforming Government or the Pension Fund sales force. There is also strong pressure from recently retired multipillar participants to allow them to opt back to the first pillar. Negative public reaction led the pension regulator MAPAS to propose opt-back to all individuals that were younger than 40 who had more than 5 service years at the second pillar inception. The Government has formed a working group to analyze possible options and propose a solution as quickly as possible. Chapter 5: Ensuring the Sustainability of the Pension System 95 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table 5.3. Effective Gross Accrual Rates, EU and North Macedonia, 2013–60 (continued) Country 2013 2020 2030 2040 2050 2060 Lithuania 0.5 0.4 0.4 ;0.4 0.4 Luxembourg 1.8 1.7 1.7 1.6 1.6 Malta 1.9 1.7 1.7 1.7 1.7 Netherlands 2.0 2.0 2.0 2.0 2.0 2.0 Portugal 2.1 2.1 2.2 2.2 2.2 2.2 Slovakia 1.2 1.1 1.1 1.1 1.2 Slovenia 1.5 1.5 1.5 1.5 1.5 1.5 Spain 2.3 2.1 1.7 1.7 1.6 1.6 Sweden 1.0 1.0 0.9 0.9 0.9 0.8 EU28 1.6 1.6 1.5 1.4 1.4 1.4 MKD PAYG only 1.36 (1.57) 1.36 (1.5) 1.29 (1.43) 1.16 (1.16) MKD voluntary* 0.33 (0.37) 0.43 (0.44) 0.57 (0.65) 0.51 (0.58) MKD mandatory* 0.51 (0.58) 0.51 (0.58) 0.51 (0.58) Source: EU Aging Report 2015. Note: Data for North Macedonia (MKD) reflect the effective accrual rate for new retiree with 35 service years for men (women). Pension indexation in North Macedonia has been among the more generous in Europe. Best international practice suggests wage valorization of the pension base, and price indexation of pension benefits,100 as most EU countries do (Table 5.4). North Macedonia indexes pensions using the Swiss formula: 50 percent based on wage changes and 50 percent on price changes. Indexing pensions using a combination of wage and price changes transfers a share of economic growth to pensioners, diverting it from other national priorities. If real wages grow, real pensions grow, independent of pensioner activity, but if real wages fall, so do pensions. Price indexation, on the other hand, insulates pensioners from economic swings. A significant number of developed countries—Germany, Sweden, Netherlands, Norway, and to an extent Switzerland—use similarly generous pension indexation, as do new EU member states and transition economies Bulgaria, the Czech Republic, the Slovak Republic, Romania, Bosnia and Herzegovina’s Republika Srpska, and to a certain extent Latvia. In many of these countries, however, more generous indexation has been accompanied by pension reforms to build fiscal sustainability, such as adjustment of the retirement age, revenue-balancing coefficients, and accrual rate adjustments. Table 5.4. Pension Valorization and Indexation Rules, OECD and Eastern Europe Countries Country Valorization Indexation Austria Wages Discretionary Belgium Prices Prices Canada Wages Prices, conditional on pension fund finances) Finland 80% wage growth, 20% prices 20% wage growth, 80% prices France Prices Prices Germany Wages, adjusted for sustainability Wages, adjusted for sustainability Greece Prices 50% wage growth, 50% GDP growth Iceland Statutorily fixed rate Prices 100 Valorization upgrades past earnings to current period wages prior to awarding first pension, while indexation adjusts all pensions in payment. Wage valorization maintains the initial replacement rate stable across generations of new pensioners. Price indexation keeps the real purchasing power of pension benefits constant. 96 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 5.4. Pension Valorization and Indexation Rules, OECD and Eastern Europe Countries (continued) Country Valorization Indexation Prices for low pensions, 90% or 75% of prices Italy GDP for higher pensions Japan Wages Wages until age 67, prices thereafter Korea Wages Prices Luxembourg Wages Wages Netherlands Wages, conditional on pension fund finances Wages, conditional on pension fund finances Norway Wages Wages - 0.75 Prices and supplement depending on GDP Portugal 25% wages,75% prices growth Spain Prices 0.25% to prices +0.5% Wages (minus 1.6% conditional on scheme Sweden Wages finances) Switzerland Statutorily fixed rate (2%) 50% wages, 50 %prices United Kingdom Wages Prices United States Wages Prices Eastern Europe 50% wages, 50% prices by law, with ad hoc Bulgaria Wages adjustments in recent years Variable formula based on price and salary Croatia trends (in a proportion of 70:30, 50:50 or Same as the Valorization formula 30:70, whichever is more favorable Czech Republic Wages 50% wages, 50% prices Estonia 50% wages, 50% prices 80% wages, 20% prices Hungary Wages Prices Latvia Wages Prices + 50% GDP Lithuania Wages Ad-hoc North Macedonia Wages 50% wages, 50% prices Montenegro 25% wages, 75% prices 25% wages, 7%5 prices Poland Wage bill growth Prices + at least 20% of real wages Romania Wages 50% wages, 50% prices Prices, and supplementary up to a ceiling, Serbia Wages depending on GDP growth Slovak Republic Wages 50% wages, 50% prices Slovenia Wages with discretion element Wages Turkey Prices + 30% GDP Prices (except for civil servants) Source: OECD Pensions at a Glance 2017, World Bank Database, and MISSOC: http://www.missoc.org/MISSOC/INFORMATIONBASE/COMPARATIVETABLES/MISSOCDATABASE/ comparativeTableSearch.jsp. Ad hoc indexations have been frequent in North Macedonia for the past decade. Sticking to an indexation rule is difficult as an economy grows. Politicians are often tempted to share the growth dividend with pensioners instead of financing other pressing public needs. Even though the indexation rule is already generous, the Government of the Republic of North Macedonia has often re-indexed pensions at rates determined ad hoc. In 2008, pensions went up by 21.7 percent—about 14 pp above the regular indexation rule. They were then under-indexed in 2009 and 2010 (Figure 5.7).101 Supplementary 101 This outcome was partly caused by incorporation of some labor benefits (like transport and hot meals) into the net wage. However, the result was a replacement rate spike in 2008 that was alleviated in 2009 and 2010. Chapter 5: Ensuring the Sustainability of the Pension System 97 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 pension increases of 5 percent were announced in 2013, 7.5 percent in 2014, and 10 percent in 2015. In 2013 the raise in pension benefits was an average of 5 pp above regular indexation.102 In 2014, 2015, and 2016, ad hoc indexation was constrained by a 5 percent indexation cap, and each beneficiary instead received flat payments of MKD 600–650, leading in effect to indexation of 1–10 percent, depending on pension level. Effectively, these supplementary indexations, set above rates of wage growth, have transferred the bulk of national productivity growth to pensioners, instead of more pressing national priorities.103 Poverty rates are lower among older cohorts than for other vulnerable groups in North Macedonia and do not (socially) justify the accelerated growth of real pensions and higher pension spending. Figure 5.7. Expected and Actual Indexation of Figure 5.8. North Macedonia: Average Pension Pensions in North Macedonia, 2008–16 Replacement Rate and Pension Spending, 2001–16 Percent Percent of GDP 25 65 10.5 60 10.0 20 55 9.5 15 50 9.0 10 45 8.5 40 5 35 8.0 0 30 7.5 8 9 0 1 2 3 4 5 6 1 3 5 7 9 1 3 5 6 200 200 201 201 201 201 201 201 201 200 200 200 200 200 201 201 201 201 JJExpected indexation JJActual indexation JJTotal pension expenditures, rhs ÂÂAverage net replacement rate, lhs Source: PIOM and State Statistical Office. Source: PIOM and State Statistical Office. As a consequence of supplementary ad hoc indexations, pension spending and average pension replacement rates (the average pension as a share of the average wage) have risen considerably. Average replacement rates have been growing continuously since 2011 (Figure 5.8). In 2016, gross new old-age replacement rates in North Macedonia were about 51 percent, among the highest in transition economies and the EU (Figure 5.9). Despite strong real GDP growth, a widening of the contribution base, and job growth that outstripped the growth of new retirees in the last five years—which should have brought down the share of pension spending in GDP—that share (of pension spending) grew to more than 10 percent. High benefits require high contribution rates, yet North Macedonia has cut the contribution rate in an attempt to reduce labor costs. The current pension contribution rate of 18 percent is one of the lowest among peer countries (Figure 5.10) as a result of a series of cuts designed to heighten labor competitiveness by reducing labor market costs. Between 2008 and 2010 the contribution rate was lowered first from 21.2 to 19 percent and then to 18 percent. The second-pillar contribution rate was simultaneously dropped from 6.7 percent to the current 6 percent. The government’s initial plan was to reduce the contribution rate to 15 percent from 2011 on, but it realized that in combination with ad hoc pension indexations this would result in a highly detrimental pension deficit. 102 Regular indexation was 2%. Supplementary 5% indexation was done by distributing MKD 550 to all beneficiaries, which averaged a 5% increase. Effectively, the overall pensions increase ranged from 3% for high-pension beneficiaries to 10% for low-pension ones. 103 Had the indexation rule been implemented vigorously since 2011, in 2016 the pension spending share in GDP would have been 8–8.5%, about 2 pp less than its current level. 98 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 5.9. Replacement Rates in North Macedonia The performance of the second pension pillar and Selected Countries, 2015 Percent has been positive. By the end of 2016, it served 80 427,000 people (75 percent of pension-insured 70 Macedonians) and managed individual pension 60 assets worth MKD 48 billion (€780 million), the equivalent of about 8 percent of GDP. The 50 annualized rate of return on second-pillar 40 accounts since its inception reached 5.6 percent 30 in 2016—3.5 percentage points above inflation as 20 measured by the CPI. Returns have outperformed 10 the nominal wage growth rate,104 suggesting that 0 one MKD diverted into the second pillar would HRV BGR NLD IRL CZE HUN LVA ROU SWE SVN BEL DNK EST EU28 DEU NOR FIN Euro area FRA MKD 2016 AUT SVK POL PRT ITA ESP Source: The EU Aging Report 2015. ultimately yield a higher pension than one MKD contributed to the PAYG system. Some 69 percent of second pillar accumulation is invested in public debt and deposits. Fund management fees,105 which are currently less than 0.5 percent of net assets and are expected to decline in the future, are among the lowest in second=pillar countries. The voluntary third pillar, launched in April 2009, has not attracted much interest. Through yearend 2016 it comprised 23,000 individuals and MKD 1 billion in accumulated funds, the equivalent of 0.2 percent of GDP. Modest voluntary pension savings are common in post-socialist countries, regardless of their income level. Figure 5.10. Public Pension Insurance Contribution Rates, Selected Developed and CEE Economies, 2016–17 Percent 35 30 25 20 15 10 5 0 Croatia Japan Switzerland Canada Australia Finland Netherlands Turkey Poland Iceland Hungary Romania Greece Serbia France Israel BIH Federation Sweden Italy Russian Federation Montenegro Germany North Macedonia Slovak Republic Bulgaria Belgium Luxembourg Denmark Chile Korea Mexico BIH RS Sources: OECD Pensions at a Glance 2017; World Bank Pension Database. 5.3  Sustainability of the Pension System 5.3.1  Baseline Scenario A status quo (baseline) simulation was used to assess the outlook for North Macedonia's pension system. The simulation shows the long-term social and fiscal performance of the system assuming there are no reforms, so that the current rules and parameters remain in place. The projection relies 104 While the returns since inception have outperformed nominal wage growth by 0.7 pp, since 2008 real second pillar returns were 3.5 pp above real wage growth. 105 Managers of mandatory pension funds can charge up to 3 percent on contributions and up to 0.4 percent on net assets annually. Chapter 5: Ensuring the Sustainability of the Pension System 99 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 on the World Bank’s Pension Reform Options Simulations Toolkit (PROST) model, adjusted to North Macedonia’s pension system parameters.106 The PROST model incorporates North Macedonia’s most recent demographic and economic data (through 2016), projected economic and demographic trends, and macroeconomic assumptions. The simulation period ends in 2080. Though North Macedonia still has a relatively young population, its share of elderly is projected to increase rapidly. With life expectancies at birth of 74 years for men and 78 years for women, and an old-age dependency ratio of 17.7, North Macedonia is “younger” than most other ECA countries. According to UN demographic forecasts, the current fertility rate of 1.51 should gradually improve to 1.80 by 2080. The baseline projection assumes that the ratio of contributors of any given age to the general population at any given age holds constant over the simulation horizon.107 Assuming no change in net migration and that aging accelerates as the UN projects, by 2080 North Macedonia’s population would shrink by more than 10 percent, and the share of Macedonians past retirement age would more than double, from 15 to 31 percent (Figure 5.11). The baseline simulation for North Macedonia Figure 5.11. North Macedonia - Demographic Projection, 2017–80 takes into account the GDP, wage, and inflation Total population, thousands values expected for 2017. For 2018 and 2019, it 2,500 uses current data and the government’s 2017 economic program estimates. From 2020 onward, 2,000 GDP growth is conservatively assumed to fall to its long-term trajectory of about 2 pp, and real 1,500 wage growth to be 2.5 percent, which would 1,000 keep the wage bill as a share of GDP at about 50 percent in the long run. It is assumed that 500 the rate of inflation rate stays close to 2 percent through 2080 (Table 5.5). The projections assume 0 6 4 2 0 8 6 4 2 0 201 202 203 204 204 205 206 207 208 contribution collection rates (collected/declared), JJ0–14 JJ15–Retirement age JJRetirement age + currently at 100 percent, are unchanged. Sources: UN Population Projections; World Bank PROST model for North Macedonia. Table 5.5. Macroeconomic Assumptions for Baseline Simulation, 2018–80 Percent 2018 2020 2030 2040 2050 2060 2080 Real GDP Growth 3.4 3.8 2.0 2.0 2.0 2.0 2.0 Real wage Growth 3.8 4.4 2.7 2.6 2.9 2.9 2.5 Inflation Rate 1.6 2.0 2.0 2.0 2.0 2.0 2.0 Source: World Bank staff estimates. Without reforms, the pension system dependency ratio will worsen rapidly. The baseline simulation assumes that the labor force participation rate and the employment rate stay close to their current levels, and for the whole simulation period, the retirement age stays at 64 for men and 62 for women. If so, because of demographic, labor market, and pension system developments, the number of contributors will saturate over the next decade but then start contracting. However, the number of 106 PROST is a standard World Bank tool used to analyze pension system features in more than 100 countries. 107 To illustrate, in 2016 the number of contributors as a share of population from 15 to retirement age was 37 percent for women and 44 percent for men. These ratios are kept constant through 2080. 100 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE beneficiaries will grow fast, with the system dependency ratio (ratio of contributors to pensioners) deteriorating rapidly. By 2045 the number of pensioners will equal the number of contributors and by 2080 will be about 10 percent above it (Figure 5.12). According to the projections, old-age pension Figure 5.12. North Macedonia: Pension Contributors and Beneficiaries, Baseline Simulation, 2017–80 replacement rates will remain high in the short and Thousands medium term, with prevailing inter-generational 600 imbalances. Without further reforms, the new 550 gross replacement rates of PAYG-only participants (nonswitchers) will gradually decline from 50 to 500 40 percent, still relatively high, in the next decade 450 (Figure 5.13). Average PAYG replacement rates will 400 decline very slowly (Figure 5.14). New and average 350 replacement rates for voluntary and mandatory new system participants (“voluntary switchers” 300 and “mandatory switchers”) are, and will continue 250017 024 031 038 045 052 059 066 073 080 2 2 2 2 2 2 2 2 2 2 to be, lower than for new PAYG-only retirees. As ÂÂPensioners, in thousands ---- Pension contributors, in thousands Figures 5.13 and 5.14 show, in a decade or two the Source: World Bank PROST model for North Macedonia. second pillar annuity generated from a 6 percent contribution rate with moderate rate of return assumptions,108 could compensate for a reduction in PAYG accrual for switchers. This socially unjust intergenerational outcome in the short and medium term, especially for voluntary switchers who have already started retiring, requires urgent action. It has resulted in a negative attitude to the second pillar, although the source of the problem lies in the first pillar parameters. Unless the PAYG accrual rates and accrual caps are aligned across generations, or the second pillar contribution rate is increased, in the next decade or two second-pillar participants will end up worse off than the old system PAYG-only participants. Figure 5.13. North Macedonia: Baseline New Pension Figure 5.14. North Macedonia: Baseline Average Replacement Rates, 2018–80 Pension Replacement Rates, 2018–80 Percent Percent 80 50 70 40 60 50 30 40 30 20 20 10 10 0 0 8 3 8 3 8 3 8 3 8 3 8 3 0 8 3 8 3 8 3 8 3 8 3 8 3 0 201 202 202 203 203 204 204 205 205 206 206 207 208 201 202 202 203 203 204 204 205 205 206 206 207 208 ÂÂNew non-switchers PAYG RR ÂÂNew voluntary switchers PAYG RR ÂÂNon-switchers average RR, men ÂÂVoluntary switchers ave. PAYG RR, men ÂÂNew mandatory switchers PAYG RR ÂÂNew voluntary switchers PAYG + funded RR ÂÂMandatory switchers ave. PAYG RR, men ÂÂVolunt. sw. PAYG + funded ave. RR, men ÂÂNew mandatory switchers PAYG + funded RR ÂÂMandatory switchers PAYG + funded average RR, men Source: World Bank PROST model for North Macedonia. Note: PAYG = Pay as you go; RR = Rate of return. 108 Simulation assumes steady-state real annual rates of return of 3 percent on second pillar accumulation and 2 percemt during the decumulation (anuity payment) phase. Chapter 5: Ensuring the Sustainability of the Pension System 101 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Without reforms, over the next decade PAYG Figure 5.15. Baseline Pension Expenditures, Revenues and Deficit, 2017–80 spending, and deficits, will grow. Worsening Percent of GDP system dependency and the slowly declining 15 replacement rates of old system participants (non-switchers) will result in higher PAYG spending 10 and a higher deficit (Figure 5.15). By the early 2030s, PAYG expenditures will reach 12 percent of 5 GDP, and the overall pension deficit will equate to 0 about 6 percent of GDP. Unless its parameters are adjusted, the country's pension system’s already -5 high pension spending will crowd out even more funds for other public needs than is already the -10 7 4 1 8 5 2 9 6 3 0 201 202 203 203 204 205 205 206 207 208 case. ÂÂBaseline: total PAYG expenditures ÂÂBaseline: total PAYG revenue ÂÂBaseline: total PAYG deficit Source: World Bank PROST model for North Macedonia. 5.3.2  Alternative Policy Simulations Possible policies to address the long-term sustainability of the pension system in North Macedonia and the adequacy of pension incomes were also simulated. Several policies—some already considered by the government—were tested for possible relief from the imbalances detected in the baseline simulation (Table 5.6). The NOCAP simulation eliminates the cap on PAYG accrual rates for voluntary second-pillar participants.109 It attempts to estimate the fiscal costs of setting the voluntary switcher PAYG accrual rates at the same level as non-switchers. The INDEX simulation tests the impact of indexing pension payments to the CPI (best international practice) rather than using the Swiss formula. The RETAGE simulation gradually increases and equalizes retirement ages and tests how that would affect pension levels and spending. In that simulation, the retirement age goes up by four months per year for women and two months per year for men until by 226 they are equal at 65; it grows in parallel with life expectancy thereafter.110 The 40+RET simulation transfers all second-pillar members older than 40, both voluntary and mandatory, back to the first pillar, with standard PAYG benefits in exchange for their second-pillar accumulations. This simulation also investigates the possible impact of applying the cutoff age for entry into the second pillar retroactively. In the ARHARM simulation, a single gross accrual rate for all pre-2006 service for nonswitchers is set at 0.9 percent. The same rate is applied post-2006 for nonswitchers, but for all switchers the rate is reduced proportionally by the ratio of PAYG rate to total contributions. This simulation harmonizes accrual rates by simultaneously reducing them for nonswitchers and raising them for switchers. The SOCPEN simulation introduces a social pension of 40 percent of the average pension for all Macedonian citizens older than 65 that are not eligible for a contributory pension (including the survivor pension).111 Figures 5.16 to 5.20 summarize the results. 109 For voluntary switchers the accrual rate for service before 2006 is capped at 11.65 percent for men and 13 percent for women regardless of years worked. 110 Increasing the retirement age with life expectancy keeps the duration of the retirement period unchanged for all future generations and does not erode second-pillar initial annuities. 111 The Government of the Republic of North Macedonia has announced plans to Introduce a social pension along the lines of SOCPEN. Because a social noncontributory pension is not an insurance-based benefit, it should be financed by government transfers rather than the insurance pension system. A nontargeted social pension raises questions about its effectiveness and social justification, especially with a social assistance system that is well-targeted and is performing well. 102 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 5.6. Individual Pension Policy Scenarios Simulated with the North Macedonia Pension Model Scenario Individual Policy Measures Baseline + lifting the net accrual cap of 11.65% (men) and 13% (women) for voluntary NOCAP switchers INDEX Baseline + pensions indexed with CPI inflation starting in 2018 Baseline + retirement age raised to 65 for both genders by 2026 (4 months a year for RETAGE women, 2 months for men) and then automatically indexing retirement age to life expectancy 40+RET Baseline + all second pillar participants younger than 40 return to the PAYG pillar Baseline + harmonization of PAYG gross annual accrual rates at 0.9% (0.6% for second-pillar ARHARM participants) starting in 2018 for past and future service period Baseline + all citizens older than 65 receive a social pension of 40% of the average old age SOCPEN pension (currently about €100) Figure 5.16. Option Simulations: Average PAYG Figure 5.17. Option Simulations: Average PAYG Replacement Rates, Male Nonswitchers, 2017–77 Replacement Rates, Male Voluntary Switchers, 2018–75 Percent Percent 45 50 40 35 40 30 30 25 20 20 15 10 10 5 0 0 7 2 7 2 7 2 7 2 7 2 7 2 7 8 4 0 6 2 8 4 0 6 2 201 202 202 203 203 204 204 205 205 206 206 207 207 201 202 203 203 204 204 205 206 206 207 ÂÂBaseline: male non switchers, average PAYG RR ÂÂARHARM ---- RETAGE ÂÂBaseline: voluntary switchers, average PAYG RR ÂÂARHARM ---- RETAGE ÂÂINDEX ÂÂNOCAP ÂÂ40+RET ÂÂINDEX ÂÂNOCAP Source: World Bank PROST model for North Macedonia. Figure 5.18. Option Simulations: Average PAYG Figure 5.19. Option Simulations: PAYG Total Spending, Replacement Rates, Male Mandatory Switchers, 2017–80 2017–77 Percent Percent of GDP 40 16 35 15 30 14 13 25 12 20 11 15 10 10 9 5 8 0 7 8 4 0 6 2 8 4 0 6 2 7 4 1 8 5 2 9 6 3 0 201 202 203 203 204 204 205 206 206 207 201 202 203 203 204 205 205 206 207 208 ÂÂBaseline: mandatory switchers, average PAYG RR ÂÂARHARM ---- RETAGE ÂÂBaseline: total PAYG expenditures ÂÂARHARM ---- RETAGE ÂÂINDEX ÂÂINDEX ÂÂNOCAP ÂÂ40+RET ÂÂSOCPEN Source: World Bank PROST model for North Macedonia. Chapter 5: Ensuring the Sustainability of the Pension System 103 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Individual policy measures could only partly Figure 5.20. Option Simulations: PAYG Deficit, 2017–80 stabilize North Macedonia’s pension system. By Percent of GDP 0 the 2030s raising the retirement age (RETAGE) -1 would prolong workers’ stay in the labor market, -2 increase new replacement rates, slow the -3 retirement flow, and reduce pension spending -4 and deficits by up to 2 percent of GDP annually -5 (Figures 5.19 and 5.20). Similarly, by the 2030s -6 -7 indexing pensions to the CPI instead of using -8 the Swiss formula (INDEX), would reduce pension -9 deficits by the same amount but at the expense -10 7 1 5 2 3 4 8 9 6 0 201 202 203 203 204 205 205 206 207 208 of declining average replacement rates (Figure  ÂBaseline: total PAYG deficit ÂÂARHARM ---- RETAGE 5.16). The ARHARM simulation stabilizes pension ÂÂINDEX ÂÂNOCAP ÂÂ40+RET ÂÂSOCPEN finances through a rapid reduction of new Source: World Bank PROST model for North Macedonia. nonswitcher pensions and by harmonizing their pension accruals with the (upgraded) accruals of second-pillar participants. RETAGE, INDEX, and ARHARM all prevent the further buildup of PAYG pension spending and deficits, though they each impose a burden on a different group: future pensioners who must work longer (RETAGE), all current and future pensioners (INDEX), or future nonswitchers (ARHARM). The other three simulations—NOCAP, 40+RET, and SOCPEN—generate additional costs in either the short or long term. Removing the constraining accrual cap for voluntary switchers (NOCAP) immediately raises their new replacement rates close to the that of nonswitchers (Figure 5.17). The costs of NOCAP are small in the first 10 to 15 years and do not exceed 0.3 percent of GDP even at their mid-2030 peak, when most voluntary switchers would have retired (Figures 5.19 and 5.20). The 40+RET simulation initially generates savings from lower transition costs and repatriated second-pillar savings (Figures 5.19 and 5.20) but in the long run the cost is unsustainable and causes deficits of more than 9 percent of GDP as returnees to the PAYG pillar start retiring on high PAYG-only pensions (Figure 5.16). The SOCPEN simulation demonstrates the impact of introducing a universal social benefit for the elderly (the “social pension”). Though with low costs initially (0.09 percent of GDP in 2018), the social pension generates a long-term extra cost of about 1 percent annually.112 The NOCAP and SOCPEN simulations have positive social effects but add costs to the already tight pension finances. NOCAP should be considered when fiscal space allows, supported by cost saving policies. Furthermore, whether an untargeted SOCPEN, financed from general revenues, would be effective outside the current social assistance system raises serious concerns. The search for a consistent package of measures to improve the equity, adequacy, and fiscal sustainability of the country’s pension system requires both a short- and a long-term perspective. With some policy options demonstrating that they can stabilize the pension system in North Macedonia, it could be expected that adopting several cost-saving measures simultaneously would have even more impact. The policy package tested (MIX1) assumes a combination of CPI-based indexing (as in the INDEX simulation) and higher retirement ages (as in the RETAGE simulation). The MIX1 package also assumes that (1) pension contributions go up to 20 percent—closer to peer country rates—with 6 percent still diverted to the second pillar, and (2) there is a three-year pension indexation freeze between 2018 and 2020 in order to counteract the expansionary impact of the bonus indexations that have taken place since 2013. Table 5.7 summarizes the elements of the MIX1 policy package. 112 Pension coverage of elderly age 65+, currently standing at 99%, would decline as the current generations of contributors (of which is on the average only 45% covered by social insurance) approach retirement age. 104 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 5.7. Pension Policy Package Simulated with the North Macedonia Pension Model Baseline + INDEX + RETAGE + Pension freeze 2018–20 + Contribution rate up to 20% (of which MIX1 second-pillar contributions remain at 6%) Figure 5.21. Policy Package MIX1, Average Replacement Figure 5.22. Policy Package MIX1, PAYG Spending, Rates, 2017–77 2017–80 Percent Percent of GDP 45 13 40 12 35 11 30 25 10 20 9 15 8 10 7 5 0 6 7 2 7 2 7 2 7 2 7 2 7 2 7 7 4 1 8 5 2 9 6 3 0 201 202 202 203 203 204 204 205 205 206 206 207 207 201 202 203 203 204 205 205 206 207 208 ÂÂBaseline: male non-switchers PAYG RR ---- MIX1: male non-switchers PAYG ave. RR ÂÂBaseline: total PAYG expenditures ---- MIX1: total PAYG expenditures ÂÂBaseline: male voluntary sw. PAYG RR ---- MIX1: male voluntary sw. PAYG ave. RR ÂÂBaseline: male mandatory sw. PAYG RR ---- MIX1: male mandatory sw. ave. PAYG RR Source: World Bank PROST model for North Macedonia. Figure 5.23. Policy Package MIX1, PAYG Deficit, 2017–80 A strong and decisive policy package for North Percent of GDP Macedonia could yield sustainable short- and 1 long-term gains. The MIX1 combination, which 0 -1 combines several tough policy measures, would -2 reverse pension trends in North Macedonia -3 and make them sustainable (Figures 5.21–5.23). -4 Pension spending and deficits would decline -5 steadily throughout the simulation period. The -6 combination of a temporary pension freeze and -7 CPI-based indexation of the pensions would yield -8 7 1 2 lower average replacement rates than the baseline 4 8 5 9 6 3 0 201 202 203 203 204 205 205 206 207 208 ÂÂBaseline: total PAYG deficit ---- MIX1: PAYG deficit scenario; adjustment of the retirement age would Source: World Bank PROST model for North Macedonia. improve the system dependency ratio; and higher contribution rates would bring in more pension revenue. Savings in the PAYG system would be distributed across all current and future pensioners. The fiscal outcome of MIX1 would clearly be much better than the measures separately and keep the system sustainable in the long run. Adopting such a policy package would help alleviate the burning issue of first pensions for voluntary switchers by removing the accrual rate cap (adding NOCAP) and harmonizing the accrual rates across generations (adding ARHARM). Dropping one or more of the MIX1 measures would weaken its short-term impact and require additional transfer from the central budget to resolve the challenge of PAYG accrual rate differentials and low first multipillar pensions. After the fiscal savings of a tough policy package such as MIX1 materialize in the pension system, policymakers in North Macedonia could reform the system of early retirement for hazardous occupations by extending the service period and, in the medium and long term, consider raising the contribution rate for the second pillar, or partially relaxing some of the tougher measures. Chapter 5: Ensuring the Sustainability of the Pension System 105 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 5.4  Conclusions and Reform Options Stabilizing North Macedonia’s pension system requires immediate and determined action. Without reforms, over the next decade or two, system’s fiscal conditions will only worsen. A package comprising a temporary (3-year) pension indexation freeze, CPI indexation, a modest increase in the retirement age for both men and women, and an additional 2 percent in contributions for PAYG financing would stabilize the system fiscally for the long term. In the short term the reform package would also be strong enough to open up the fiscal space necessary to equalize the accrual rates of voluntary second- pillar participants, harmonize the accrual rates of all system participants and, in the medium term, raise the second pillar contribution rate. Adopting only some of the measures in the policy package may delay system stabilization and create additional problems. 5.4.1  Reform Options Short term 1. Legislate CPI indexation of pension benefits and refrain from any ad hoc pension increases. This would fix the real purchasing power of pension payments over a lifetime and stabilize pension spending in the medium term. 2. Consider a gradual increase and equalization of the retirement age for men and women. Increments added to the retirement age should exceed those of life expectancy, and the rate of increase for women should exceed the rate for men for catch-up purposes. 3. Given the frequent ad hoc increases in pensions in recent years, a temporary (3-year) pension indexation freeze should be considered to counterbalance the increases. 4. Gradually increasing the PAYG contribution rate by 2 pp would generate additional revenues and reduce the share of pension deficit financing in general revenues. 5. Reforms to early retirement for hazardous occupations are also recommended. These should include restricting early retirement eligibility to work in extreme conditions, improving conditions in the labor market, and reducing the costs of early retirement. 6. It might be advisable to explore removing the accrual cap for voluntary switchers or gradually equalizing PAYG accrual rate differentials between nonswitchers and second-pillar switchers (or both). This measure would eliminate or reduce the reasons for the low multipillar pensions and make the system fairer. Medium to long term 7. After the retirement age for women equalizes with that of men, consideration should be given to linking retirement age to North Macedonia’s average life expectancy. In the long term doing so would preserve the ratio of time spent in retirement to active service period. 106 Chapter 5: Ensuring the Sustainability of the Pension System SOWING THE SEEDS OF A SUSTAINABLE FUTURE 8. Consider raising the contribution rate for the second pillar. Not only past performance of the second pillar in North Macedonia indicates that a denar from the second pillar yields higher pension than a denar contributed to PAYG pillar, higher second pillar contribution implies lower PAYG costs for future generations. This reform, however, requires fiscal space. Chapter 5: Ensuring the Sustainability of the Pension System 107 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable North Macedonia’s social assistance spending is not only relatively low but it is also fragmented, although social protection spending (with social insurance, including pensions) is quite large. Social assistance is deficient in several areas. Recent spending has been tilted toward non-means-tested programs even though poverty is relatively high and income-generation opportunities for the poor are scarce. This raises allocative efficiency concerns. Cost-effectiveness is also a challenge, particularly for programs like the parental allowance, for which hikes in spending have not had the intended impacts on population growth. Social financial assistance benefits are barely adequate, but there are few financial incentives for social assistance beneficiaries to seek employment. It will also be necessary to address the gaps in program coverage that leave many households mired in poverty. 6.1  The Social Protection System North Macedonia’s social protection system consists of contributory (mandatory social insurance) and noncontributory (state-financed social assistance) benefits and employment promotion programs, all of which are useful in protecting poor and vulnerable Macedonians. The main contributory program is pensions. The pension scheme is a pay-as-you-go plan with two private contribution pillars. It covers the most social protection beneficiaries and absorbs over half of social assistance spending. Unemployment benefits are also contributory and, together with a range of state-financed job search tools and active labor market programs (ALMPs), embody the country’s labor market policies. The main social assistance benefits, in terms of coverage and budget expenditure, are social financial assistance (SFA), permanent financial assistance (PFA), the parental allowance (PA), and disability- related benefits. SFA is the only explicitly anti-poverty program.113 Means-tested, it targets able-bodied persons who cannot meet North Macedonia’s subsistence minimum, which was MKD 2,831 per month in 2017. The amount increases by a coefficient of 0.37 for each additional household member, up to a limit of five. This guaranteed minimum income bridges the gap between the minimum subsistence level and total household income. After three years of this benefit, the amount is halved. The means-tested PFA and CA programs also mitigate poverty and social exclusion by targeting segments of the population that are unable to work, including children. The PFA is a benefit for people unable to work and in financial need. A person with mental or physical disabilities is considered unable to work.114 PFA benefits in 2017 were MKD 4,247 for one beneficiary, MKD 5,946 with one co- beneficiary, and MKD 8,494 with two co-beneficiaries. The CA is provided until the child reaches age 18 and requires a confirmation of school attendance. It also requires that at least one parent be 113 Related smaller means-tested programs include reimbursement for electrical power bills (the “energy subsidy”) for SFA recipients and conditional cash transfer (CCT) programs. The CCT for secondary education (CCT education) is a top-up to SFA benefits when there is a child in secondary school and attends at least 85 percent of classes. The subsidized employment program, which targets specified groups of vulnerable people (including SFA beneficiaries), has in four years provided on-the-job training and employment to over 250 welfare beneficiaries. 114 Single women during pregnancy and until the child reaches 3 and people older than 65 are also eligible if they are not financially secure. 108 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable SOWING THE SEEDS OF A SUSTAINABLE FUTURE employed. The monthly benefit is MKD 744 for a child under 15 and MKD 1,180 per month for a child between 15 and 18. The maximum benefit is MKD 1,870 per family. Other noncontributory (categorical) social assistance programs are the personal care allowance (PCA) and disability benefits, family allowances like the PA provided to mothers who gave birth to a third child, one-off assistance for first newborns, and other smaller benefits and programs. There is a plethora of disability benefits, which have expanded over time with introduction of new disability- related benefits and redefinition of what qualifies as disability. The social protection system delivers a range of social services, institutional and noninstitutional, that—together with the cash transfers—are planned, regulated, and supervised by the Ministry for Labor and Social Policy (MLSP). Provision of social and child protection benefits and services is guided by strategic documents and regulated by the Social Protection Law, the Child Protection Law, the Family Law, and related by- laws.115 Strategic documents adopted by the Government of North Macedonia, such as the National Strategy for Alleviation of Poverty and Social Exclusion for 2010–20, the National Program for the Development of Social Protection 2011–21, and the Employment and Social Reform Program 2020, set social protection objectives and specify what must be done to achieve them. Benefit programs prescribed by the social and child protection laws are the exclusive mandate of the MLSP. The benefits and services are carried out by the MLSP’s network of centers for social work and a number of social welfare institutions. 6.2  Social Assistance Spending At about 14 percent of GDP, North Macedonia’s total spending on social protection, including social insurance, social assistance, and labor market programs, is high compared to peer countries  (Figure 6.1). At 10.5 percent of GDP in 2017, pension spending (old age, disability, and survivor’s, including administrative and transition costs, as well as health contributions for pensioners) makes up most of the total spending on social protection. The share of labor market policies is less than 0.1 percent of GDP, though that is standard for the region.116 Finally, North Macedonia spends 0.07 percent of GDP on institutional and 0.02 percent of GDP on noninstitutional social services—lower than regional peers. Conversely, North Macedonia’s spending on social assistance, at 1.2 percent of GDP annually for 2014–17, is among the lowest in the region. Average social assistance spending for ECA countries is 2.2 percent. Albania, Montenegro, and Serbia spend a slightly higher share of GDP than North Macedonia on social assistance, and Kosovo spends 2.84 percent of GDP on it (including spending on noncontributory pensions). Bosnia and Herzegovina’s spending of 4.72 percent of GDP is an exception related to war veteran benefits. Small transition economies like Croatia, Estonia, and Lithuania spend notably more on social assistance than North Macedonia, but Bulgaria and Latvia spent relatively little. 115 Law on Social Protection, Official Gazette no. 79/2009, and subsequent changes and amendments (latest clean version published in Official Gazette no. 148/2013); Law on Child Protection, Official Gazette no. 23/2013, and subsequent changes and amendments; Family Law, Official Gazette no. 80/1992 and subsequent changes and amendments (latest clean version published in Official Gazette no. 153/2014). 116 The number and scope of ALMPs has significantly increased in recent years. However, there has been no comprehensive evaluation of ALMPs and their effects. Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable 109 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 6.1. Social Protection Spending, 2014 or Latest Data Percent of GDP 20 18 16 14 12 10 8 6 4 2 0 14 10 13 13 14 09 14 12 12 11 13 12 14 12 13 14 09 13 14 14 14 14 14 a ine ania erbia negro donia ania lgaria roatia larus ublic tonia atvia ldova oland urkey bania BIH orgia enia aijan hstan istan sovo Ukr Ro m S nte ace Lith u B u C e B z Re p Es L M o P T A l G e m b Ar Azer aza k Taj i k K o Mo rth M gy K No Kyr JJSocial assistance JJLabor market JJSocial insurance Source: World Bank, Europe and Central Asia (ECA) Social Protection Expenditure and Evaluation Database (SPeeD). North Macedonia’s spending on social assistance—cash transfers pursuant to the social welfare and child protection laws—has been rising since 2008  (Figure 6.2). However, the number of households receiving SFA has dropped more than 50 percent since 2005, from 67,000 to 25,300 (Figure 6.3). Correspondingly, SFA spending has dropped from 0.6 to 0.2 percent of GDP in 2017. The CA program was also reduced, in terms of both coverage, from 21,000 beneficiaries to less than 5,000, and spending, from 0.09 to less than 0.02 percent of GDP.117 Figure 6.2. Spending on Social Assistance Programs, Figure 6.3. Number of Recipients for Selected 2005–17a Benefits. 2005–17 Percent of GDP Recipients, in thousands 1.4 80 1.2 70 60 1.0 50 0.8 40 0.6 30 0.4 20 0.2 10 0 0 5 6 7 8 9 0 201 1 201 2 201 3 201 4 201 5 201 6 5 6 7 8 9 0 11 2012 2013 2014 2015 2016 201 7 200 200 200 200 200 201 200 200 200 200 200 201 20 JJSFA (incl. one off financial assistance and energy subsidy ÂÂSocial financial assistance (houdeholds)ÂÂChild allowance ÂÂParent allowance JJChild and parental allowance, and one-off assistance to new borns JJDisability related benefits JJWar veterans pension JJOther social assistance Source: World Bank, SPeeD. Source: World Bank, SPeeD. Note: a. Includes health protection for beneficiaries; Spending data for war veterans were not available for 2015; it was assumed they held at the 2014 level. The structure of North Macedonia’s social assistance is inequitable, with an increasing share going to categorical programs. While spending on means-tested programs has been cut back, since 2008 the budget for non-means-tested family and child protection programs has significantly increased (Figure 6.4). This was primarily due to the 2008 introduction of the parental allowance program, which has significantly expanded since. Disability-related social benefits have also increased, though less dramatically. 117 The decline in SFA and CA coverage was primarily driven by improvements in benefit administration, especially introduction of the new cash benefit information system (CBMIS). 110 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable SOWING THE SEEDS OF A SUSTAINABLE FUTURE High poverty, and the limited income-generation Figure 6.4. Means-Tested and. Non-Means-Tested Benefit Spending, 2005–17 opportunities available to poor and vulnerable Percent of GDP Macedonians, certainly does not justify the 1.4 dramatic shift in spending from means-tested 1.2 programs like SFA and the CA to non-means- tested programs. Although poverty has declined 1.0 post-crisis and the living conditions of the poor 0.8 have improved somewhat, poverty and inequality 0.6 are still high compared to regional peers and 0.4 most EU countries (after controlling for economic 0.2 development).118 North Macedonia’s relative poverty rate119 was 22 percent in 2016 (SILC). In 0 5 6 7 8 9 0 1 2 3 4 5 6 200 200 200 200 200 201 201 201 201 201 201 201 2014, only 7 percent of the extreme poor were JJNon means-test JJMeans-tested ÂÂTotal SA spending employed, and only 14 percent of the poor.120 Yet Source: World Bank, SPeeD. spending on means-tested programs slid from 0.82 percent of GDP in 2008 to 0.6 percent in 2017. Meanwhile, spending on categorical programs more than doubled. The PA program is a clear outlier in terms of public spending. It is provided (effective January 1, 2009) to mothers who give birth to a third child. The benefits are very high compared to other social assistance programs, at MKD 8,362 per month, and it is granted until the third child is of age 10. By 2017, the program’s caseload exceeded 26,000, and it consumed 0.4 percent of GDP. Cross-country comparison reveals that though WB6 spending priorities differ, there is a clear bias toward categorical programs ( Figure 6.5). As noted by Gotcheva and Sundaram (2013), categorical programs are easy to increase in good times but very difficult to shrink, even when the fiscal pressure is significant.121 In North Macedonia and Serbia, spending levels on means-tested and categorical programs are roughly equal, but in Albania (due to disability benefits) and Bosnia and Herzegovina (primarily due to war-related benefits) a significant bias exists towards categorical benefits. That said, in recent years, both Albania and Bosnia and Herzegovina have initiated reforms to reduce spending on categorical benefits. In its new Government Program (2017), North Macedonia recognizes the need to reconsider the design and cost of its social assistance programs. The costly PA program, which does not seem to have stimulated population growth as was intended, has been thoroughly discussed.122 North Macedonia’s total fertility rate in 2015 was 1.49 live births per woman; the EU average is 1.58. Over the past 10 years, due to the falling birth rate and the rising general mortality rate, by 2015 the natural population increase had dropped from 2.0 per 1,000 in 2005 to 1.3.123 However, even if the government decided to discontinue the PA program in 2018, there would still be significant transitional costs associated 118 Source: Poverty Reduction, Shared Prosperity, and Inequality in North Macedonia in the Post-Crisis Period (2009-2013), World Bank, 2016. 119 Also referred to as the at-risk-of-poverty rate, which shows the share of the population living below 60 percent of the median household per adult equivalent income. 120 Employment for the bottom 40 was higher at 28 percent, but still considerably behind the national rate. 121 Social Safety Nets in the Western Balkans: Design, Implementation, and Performance, published in Laderchi and S.Savastano (Eds.), Poverty and Exclusion in the Western Balkans – New Directions in Measurement and Policy, New York: Springer. 122 Empirical studies generally find little relationship between cash benefits and fertility (McNown and Ridao-Cano 2004, or Gauthier and Hatzius 1997. 123 Continuing emigration will also reduce the North Macedonian population. Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable 111 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 with current entitlements. The benefits for those who started receiving PA benefits in 2018 would continue through 2028, reducing short-term savings from terminating the program. Table 6.1 shows annual budget estimates if the program were terminated in 2018 and monthly entitlements remained constant in nominal terms at MKD 8,362. The costs were estimated using Statistical Office data on the number of third children born between 2009 and 2016, and assuming that the number of births in 2017 was the same as in 2016. Figure 6.5. Spending on Social Assistance by Targeting Method, Local Currency Units A. Albania B. Bosnia and Herzegovina Billions Percent Billions Percent 25 9 1.2 10 8 8 1.0 20 7 6 6 0.8 15 4 5 0.6 4 2 10 3 0.4 0 5 2 0.2 -2 1 0 0 0 -4 0 2 4 6 8 0 2 4 5 6 7 8 9 0 1 200 200 200 200 200 201 201 201 200 200 200 200 200 201 201 JJPoverty alleviation programs (LCU) JJOther SA ---- GDP growth, rhs JJPoverty alleviation programs (LCU) JJMeans-tested SA JJOther SA ---- GDP growth, rhs C. North Macedonia D. Serbia Billions Percent Billions Percent 8 8 70 10 7 60 8 6 6 50 6 4 5 40 4 4 2 30 2 3 0 20 0 2 -2 10 -2 1 0 -4 0 -4 0 2 4 6 8 0 2 4 6 0 01 2 3 4 5 6 7 8 9 0 11 12 13 200 200 200 200 200 201 201 201 201 200 20 200 200 200 200 200 200 200 200 201 20 20 20 JJPoverty alleviation programs (LCU) JJMeans-tested SA JJOther SA JJPoverty alleviation programs (LCU) JJMeans-tested SA JJOther SA ---- GDP growth, rhs ---- GDP growth, rhs Source: World Bank, SpeeD. Table 6.1. PA Recipients and Possible Spending Scenario, 2018–28 Year Number of beneficiaries Annual budget, MKD million (2017 prices) 2018 30.506 3.061.1 2019 29.028 2.912.8 2020 25.885 2.597.4 2021 22.656 2.273.4 2022 19.387 1.945.3 2023 15.943 1.599.7 2024 12.357 1.240 112 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 6.1. PA Recipients and Possible Spending Scenario, 2018–28 (continued) Year Number of beneficiaries Annual budget, MKD million (2017 prices) 2025 8.763 879.3 2026 5.262 528 2027 1.754 176 2028 0 0 Source: Ludovico Carraro, Report on alternative reform scenarios, Oxford Policy Management, 2017. 6.3  Performance of Social Assistance Programs North Macedonia’s social protection system has serious performance problems. Here, the main social assistance programs are assessed in terms of coverage; benefits and beneficiary distribution; contribution to consumption, poverty and inequality effects; and cost-efficiency.124 The performance assessment also sheds light on how North Macedonia compares to regional peers. 6.3.1 Coverage Poverty rates are relatively high in North Macedonia, but social assistance programs reach a third of the poorest quintile  (Table 6.2).125 The result is even more striking given that the bottom quintile roughly corresponds to North Macedonia’s poor population. Between 2010 and 2012, overall coverage of social assistance programs increased from 8.5 to 10.3 percent, where it held steady, registering 10.2 percent in 2016. This pattern was primarily driven by the increased coverage of the bottom quintile by family and child protection benefits. Table 6.2. Social Protection Coverage by Quintile Coverage Total Q1 Q2 Q5 All social protection 46.3 55.1 45.7 44.8 All social assistance 8.5 27.2 5.1 1.2 2010 SFA, PFA, and related allowances 6.9 24.3 4.0 0.8 Child allowances + scholarships + personal, family or disability allowances 1.7 3.8 1.1 0.4 All social protection 49.5 54.6 52.6 39.9 All social assistance 10.3 27.3 8.4 0.6 SFA, PFA, and related allowances 7.3 21.3 5.8 0.6 2012 One-off monetary assistance for a newborn child and parental allowance 2.1 4.7 0.7 0.0 Child allowance 0.9 0.6 0.5 0.0 Personal, family, or disability allowances for war veterans 0.2 0.6 0.1 0.0 Scholarships 0.3 0.1 1.4 0.0 124 The analysis presented in this section draws on data from the Household Budget Survey (HBS) for 2010, 2012, and 2016 and covers social transfers for which information is available. The poverty rate is set at the 20th percentile of population distribution, which makes it possible to compare results over time and across countries. The data include both direct and indirect beneficiaries; the rates presented indicate the share of the population living in households where at least one member receives a transfer. The analysis used the World Bank ADePT software. For consistency checks, results calculated with HBS were compared with administrative data and Survey of Income and Living Conditions data. Although, as expected, the numbers of beneficiaries total transfer value was not exactly the same, they followed the same trend over time. 125 The r findings on both coverage and poverty impact of social transfers, however, need to be treated with caution since they tend to be under-estimated in analysis of household survey data. This happens due to generally low coverage of some programs and underreporting of benefit receipt. CA and disability related programs are not explored in detail because the dataset contains too few observations. Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable 113 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table 6.2. Social Protection Coverage by Quintile (continued) Total Q1 Q2 Q5 All social protection 46.2 57.7 40.1 42.0 All social assistance 10.2 33.9 7.2 0.9 SFA, PFA, and related allowances 5.2 19.3 4.3 0.2 2016 One-off monetary assistance for a newborn child and parental allowance 4.7 14.6 2.9 0.0 Personal, family, or disability allowances for war veterans 0.2 1.0 0.0 0.1 Scholarships 0.3 0.1 0.1 0.7 Source: HBS data. North Macedonia does not compare well with regional peers even though between 2010 and 2016 its social assistance coverage of the lowest quintile broadened markedly, from 27.2 to 33.9 percent.126 Other countries whose social assistance spending is similar, or even much lower, have much better coverage rates. For example, coverage of the poorest quintile is 80 percent in Latvia and 90 percent in Romania (Figure 6.6). Figure 6.6. Social Assistance Coverage of the Poorest Quintile, 2016 or Latest Data Percenta 100 90 80 70 60 50 40 30 20 10 0 2 2 0 7 2 8 8 8 4 2 3 0 6 3 1 6 0 2 1 2 201 201 201 200 201 200 200 200 200 201 201 201 201 201 201 201 201 201 and 7 201 201 e o rgia mania Latvia ngary elarus uania ation lgaria tonia oland kraine hstan menia ldova osovo donia erbia lbania osniaa 200 negro kistan r Es P ak Mo K S B vin nte i G Ro Hu B Lith Fede Bu U Kaz Ar Ma ce A go Mo Taj s ian o rth Herze Ru s N Source: World Bank, SPeeD. Note: A. Information on ECA countries is based on standardized methodology for assessment of social assistance programs using different household surveys and harmonized ECAPOV aggregates of consumption per capita. Households are ranked into quintiles before transfers. Figure 6.7. Last Resort Social Assistance: Poorest Coverage of means-tested social assistance (SFA Quintile Coverage, 2016 or Latest Data and PFA) in general but particularly of the bottom Percenta quintile is low. About 5 percent of Macedonians 40 35 live in households where at least one member 30 receives targeted social assistance and related benefits. According to HBS data, only 19.3 percent 25 of the those in the poorest quintile receive SFA; 20 clearly the program does not reach many low- 15 income households. SFA coverage is thus low 10 compared to similar last-resort programs in 5 Albania and Kosovo, but it is similar to that of 0 011 16 012 16 011 012 10 08 10 013 013 07 04 08 07 10 S 2 20 2 20 E 2 L 2 20 20 20 2 2 20 20 20 20 20 KO ARM ALB MKD MN PO SRB BGR LVA MDA UKR HUN EST LTU BIH KAZ similar programs in other WB6 countries (Figure Source: World Bank, SPeeD. 6.7). Note: A. Information on ECA countries is based on standardized methodology for assessment of social assistance programs using different household surveys and harmonized ECAPOV aggregates of consumption per capita. Households are ranked into quintiles before transfers. 126 Possibly due to the increased coverage of family and child protection programs. 114 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable SOWING THE SEEDS OF A SUSTAINABLE FUTURE 6.3.2  Distribution of Benefits and Beneficiaries Between 2010 and 2016 SFA coverage of the bottom quintile fell from 24.3 to 19.3 percent, but because coverage of the higher quintiles dropped more, the benefits distribution shifted toward the bottom quintile(Table 6.2). The change can largely be attributed to introduction of the Cash Benefit Management Information System (CBMIS), which reduced inclusion errors and benefit leakage and increased the efficiency of administration. Moreover, although between 2010 and 2012 the proportion of social assistance beneficiaries in the bottom quintile dropped from 69.8 to 58.4 percent in 2012, in 2016 it rose to 71.5 percent (Figure 6.8). Figure 6.8. Distribution of Social Assistance Figure 6.9. Distribution of Social Assistance Benefits Beneficiaries by Quintile by Quintile Percent Percent 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 JJ2010 JJ2012 JJ2016 JJ2010 JJ2012 JJ2016 Source: HBS data. Source: HBS data. Social assistance is well-targeted: more than Figure 6.10. Last Resort Social Assistance, Percent of Benefits Received by the Poorest Quintile a 70 percent of all transfers go to the poorest Percent quintile (Figure 6.9), which places North 100 Macedonia among the best performers in both 90 the WB6 countries and all of ECA. Furthermore, 80 70 Between 2012 and 2016, the benefit for the poorest 60 quintile went up from 57.8 to 70.3 percent. This is 50 largely due to the targeting accuracy of the SFA 40 program, which also compares very favorably to 30 those of regional peers (Figure 6.10). In fact, the 20 10 leakage of social assistance to the richest quintile 0 8 11 8 11 6 4 0 13 0 0 7 13 6 12 7 12 (1.4 percent of all benefits) is among the lowest 200 E 20 200 S 20 201 200 201 20 201 201 200 20 201 20 200 L 20 LTU MN BGR KO MKD EST SRB UKR LVA KAZ HUN MDA ARM ALB BIH PO in the region. The accuracy of targeting is also Source: World Bank, SPeeD. Note: A. Information on ECA countries is based on standardized methodology for reflected in the fact that leakage incidence to the assessment of social assistance programs using different household surveys and harmonized ECAPOV aggregates of consumption per capita. Households are ranked into richest quintile is low in all the social assistance quintiles before transfers. programs except disability benefits. Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable 115 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 6.3.3  Contribution to Consumption On average, social protection transfers (with social insurance, including pensions) account for 30 percent of the cumulative consumption of Macedonians  (Table 6.3). For households in the poorest quintile, social transfers account for 51 percent of consumption, and for those in the richest quintile 23 percent. Old-age pensions constitutes 28 percent of the total consumption of the targeted population. The data also suggest that pension transfers play a much more important role in the consumption for the bottom quintile. Finally, although it contributes just 1.7 percent, social assistance is progressive in that it constitutes a higher share of total consumption for those in the poorest quintile. SFA recipients may, however, qualify for other Figure 6.11. Distribution of Benefit Amounts per Household, MKD types of social assistance benefits that increase Percent the total amount received, and their consumption. 25 According to the data obtained from the CBMIS, 78 percent of SFA beneficiary households receive 20 two different types of benefits (usually the SFA 15 and the energy subsidy), and 0.1 percent of them receive three (SFA, energy subsidy, and CA). 10 Among the households that receive two benefits, 5 22 percent of them receive the SFA and a CCT 0 while 2.1 percent receive the SFA and the CA. The 1.4] 6.8 ] 2.2] 7.6] 343 ] 8.4 ] 3.8 ] 9.2 ] 4.6 ] 120 ] -152 .4-2,47 .8-3,43 2-4,38 7.6-5, 3-6,29 .4-7,25 8-8,20 .2-9,16 .6-10, maximum amount a single household can receive 6 6 [5 1,521 [2,4 76 32. [4,38 [5,34 6,298 ,253. 09 [9,164 [ [3,4 [ [7 [8,2 is MKD 10,120, but only 27 households entitled to Source: CBMIS (MLSP). SFA and CCT reach this amount. Most recipient households receive MKD 2,500–5,500 (Figure 6.11). For those that receive three benefits (SFA, energy subsidy, and CA) the maximum amount is MKD 5,259 and the minimum MKD 3,515. Nevertheless, the overall benefit amounts may increase significantly for recipients of certain categorical programs like the PA. This cumulation of benefits may result in disincentives for beneficiaries to work. Table 6.3. Benefit Incidence (All Households) Percent Quintiles of Aggregate Expenditure (AE) consumption, Poverty status net of all SA transfers Total Q1 Q2 Q3 Q4 Q5 Poor Not Poor All social protection 30.1 51.3 38.5 34.3 31.9 23.0 47.8 28.7 All social insurance 28.1 34.0 35.6 31.9 31.3 22.7 32.6 27.7 Pension and disability insurance 28.0 34.0 35.1 31.7 31.3 22.7 32.6 27.6 All social assistance 1.7 17.0 2.1 2.1 0.2 0.1 14.9 0.6 Social assistance and related 0.6 6.7 0.9 0.4 0.1 0.0 5.7 0.2 allowances One-off monetary assistance for a newborn child, parental allowance 1.0 10.0 1.2 1.6 0.1 0.0 8.9 0.4 for children Personal, family, or disability 0.0 0.3 0.0 0.1 0.0 0.0 0.3 0.0 allowances for war veterans Source: World Bank staff estimates based on HBS data. 116 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable SOWING THE SEEDS OF A SUSTAINABLE FUTURE Despite its low coverage and small amounts, the SFA alone is very important for beneficiary households  (Table 6.4).127 The SFA contributes an average of 26.5 percent of total household consumption, and almost 35 percent of the consumption of the poorest 20 percent of beneficiary households. It also represents 11 percent of the consumption of beneficiary households in the richest quintile. Again, the pension and other contributory programs have the highest shares in beneficiary household consumption. The contribution of categorical social assistance (one-off and PA) is also important, particularly for the poorest households. Table 6.4. Adequacy of Benefits, Direct and Indirect Beneficiaries Percent Quintiles of AE consumption, net of all SA transfers Poverty status Total Q1 Q2 Q3 Q4 Q5 Poor Not Poor All social protection 67.9 84.9 94.9 74.0 69.1 55.5 82.1 66.3 All social insurance 70.7 95.7 107.2 80.2 70.3 57.6 91.4 69.2 Pension and disability insurance 70.9 95.7 110.4 80.6 70.3 57.6 91.4 69.4 All social assistance 30.1 43.7 24.3 23.5 10.1 7.2 41.7 19.1 Social assistance and related 26.5 34.7 18.6 21.0 9.2 11.4 32.0 18.4 allowances One-off monetary assistance for a newborn child, parental allowance 35.8 50.9 31.3 23.7 10.7 10.6 49.8 22.8 for children Personal, family or disability 28.0 22.0 n.a. 42.6 n.a. 30.7 22.0 36.7 allowances for war veterans Scholarships 5.6 16.4 9.8 8.7 9.3 4.3 14.1 5.3 Source: World Bank staff estimates based on HBS data. 6.3.4  Impact of Social Transfers on Poverty Without social transfers, North Macedonia’s poverty level would be much higher.128 Table 6.5 shows the simulated impact of discontinuing a particular program or combination of social transfer programs. It assumes that household welfare would diminish by the full value of a given transfer. For comparison purposes, it set the poverty line at the 20th percentile of the population per capita, using 2016 HBS consumption data129 The poverty gap is 6 percent and the Gini inequality index is 39 percent.130 In the absence of the social transfers listed below, the poverty rate using this line would more than double, reaching 41 percent. Even though pensions are not designed as anti-poverty measures, if the old-age pensions were discontinued, the poverty rate would grow to 38 percent. On the other hand, the absence of any of the social assistance programs would not impact poverty significantly. The termination of government social and child protection benefits would only have a limited impact. This is due to the budget allocated to these programs. However, the benefit-cost ratio shows that one MKD 127 There is no evidence that the SFA contributes to overall consumption of Macedonians; it has low coverage and small benefit levels. 128 This analysis draws data from the HBS for 2010, 2012, and 2016 and covers social transfers for which information is available in the datasets. The poverty rate is set at the 20th percentile of the population distribution, which allows for comparison of results over time and across countries. The data include both direct and indirect beneficiaries; and the rates refer to the share of the population living in households where at least one member receives a transfer. In the analysis, we use the ADePT software developed by the World Bank. 129 North Macedonia’s poverty rate at US$1.90 a day line in North Macedonia was 2.35 percent . 130 The poverty gap measures the extent to which individuals fall below the poverty line as a proportion of the poverty line (6 percent in this case). The sum of the poverty gaps gives the minimum cost of eliminating poverty. Therefore, assuming perfect targeting, this measure indicates what resources would be needed to lift all the poor above the poverty line. The Gini index measures the extent to which the distribution of income consumption spending among individuals or households deviates from a perfectly equal distribution. Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable 117 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 spent on the means-tested SFA program yields higher benefits in terms of poverty reduction than if it is spent on categorical programs (Figure 6.12). Table 6.5. Impact of Various Programs on Poverty Measures, Simulating the Absence of the Program Poverty rate Poverty gap index GINI index Post-transfer Indicators 0.200 0.058 0.392 Indicators without listed transfer All social protection 0.408 0.243 0.514 All social insurance 0.387 0.219 0.497 Pension and disability insurance 0.384 0.219 0.497 All social assistance 0.229 0.080 0.406 SFA, PFA, and related allowances 0.210 0.067 0.397 One-off assistance for a newborn, parental allowance 0.217 0.071 0.400 Personal, family or disability allowances for war veterans 0.202 0.059 0.392 Scholarships 0.200 0.058 0.392 Source: HBS data and World Bank staff estimates. Among the social transfers analyzed, the SFA Figure 6.12. Benefit-Cost Ratios program seems to be the most cost-efficient. SFA, PFA, and related allowances For each MKD spent on SFA, the poverty gap All social assistance decreases by 0.6 MKD. Child and family protection One-off assistance for a newborn child and parental allowance allowances (excluding disability) are next; one Personal, family or disability allowances for war veterans MKD spent on these programs reduces the poverty All social protection gap by 0.5 MKD. However, it is unlikely that the SFA Pension and disability insurance alone could lift beneficiaries out of poverty. In the ECA region, last-resort programs hardly ever lift All social insurance people out of poverty, and they are increasingly Scholarships accompanied by work requirements. In North 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Macedonia, the social protection legislation Reduction in poverty gap per MKD spent Source: HBS data and World Bank staff estimates. provides incentives for transitioning, to work but also creates disincentives for accepting a job because taking one means a 100 percent benefit withdrawal.131 In its current form, the SFA has virtually no impact on poverty because the poverty threshold is set very low and only a small portion of Macedonians receive it. Therefore, the analysis explored several scenarios with different eligibility thresholds, their financial implications, and their impact on absolute and relative poverty (Table 6.6). The thresholds are set at 5,700 MKD per equivalent adult,132 two thirds of 5,700 MKD per equivalent adult, 40 percent of the minimum wage, and 60 percent of the minimum wage. We applied the modified OECD equivalence scale, which assigns a value of 1 to the 131 Next to the provision related to a 50 percent decline in the benefit amount after three years of receipt, the social protection law also envisages an automatic return to SFA for those beneficiaries who take up seasonal, public works, or other temporary jobs. Both measures are designed to provide work incentives to SFA beneficiaries. 132 This corresponds to MKD 12,000 for a family of four (two adults and two children); it approximates the minimum wage amount. 118 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable SOWING THE SEEDS OF A SUSTAINABLE FUTURE household head, 0.5 to each additional adult member, and 0.3 to each child. The SFA eligibility criteria in the current Social Protection Law are retained.133 Table 6.6. Budget Estimates and Poverty Rates for Simulated SFA Transfers Relative poverty rate Absolute poverty rate Euro millions per year (60% US$1.90 per day of median income) (US$5.50 per day) Sim. SA (SFA+PFA) now 20.7 24.17 22.02 2.35 Sim. GMI 40% min wage 29.7 23.92 21.5 3.01 Sim. GMI 60% min wage 46.9 23.86 21.24 1.66 Sim. GMI MKD 5,700/unit 136.3 22.87 18.76 0.03 Source: HBS data; World Bank staff estimates. At the current maximum rate of assistance, the SFA program does not reduce poverty. If, for instance, the SFA benefit were raised to 40 percent of the minimum wage, its impact would be slightly more noticeable—the absolute poverty rate would decrease from 22.02 percent to 21.5 percent. However, it still would not affect poverty dramatically because the threshold would remain low. If the benefit were raised to 60 percent of the minimum wage, the outcome would improve but the impact on poverty would still be meager. On the other hand, if the maximum benefit were raised to MKD 5,700 per equivalent adult, the situation would radically change. Poverty would drop to 18.76 percent. Although, if well- targeted, this option would obviously increase SFA impact, but it would also be fiscally unsustainable and require a budget allocation of €136 million annually. 6.4 Conclusions  Current social assistance transfers suffer from some well-known problems:  oncerns raise the need to reconsider the shift of spending from means- •• Allocative social efficiency c tested to non-means-tested programs when poverty is still relatively high and income-generation opportunities for the poor are few. •• Cost-effectiveness is questionable, particularly for programs like the PA, which has spurred a significant rise in spending that was not matched by the intended impact on population growth. The PA has some positive impacts in terms of coverage and benefit adequacy, but at 0.4 percent of GDP—twice the share spent on the SFA—the results do not justify the costs. •• The SFA benefit is very low, less than 30 percent of the net minimum wage, which then gets halved after three years of collection, but there are few if any financial incentives for social assistance beneficiaries to take jobs. The rationale for providing such a benefit is to incite self-reliance and encourage people to seek work. However, when there are no job opportunities, SFA recipients fall in the poverty trap, and the system generates social exclusion. 133 According to the social protection law people are eligible for SFA if (1) at least one household member is of working age and able to work; (2) household income is below the established threshold; (3) household members do not have special assets (second house, car insurance, receive remittances larger than 70,000 MKD, have savings of more than 70,000 MKD); and (4) household members who are unemployed have registered with the employment office and eventually accept job offers. For the purpose of simulation, child allowances are determined as 1,200 MKD and social pensions at 30 percent of the minimum pension for those older than 65 and not entitled to pensions. Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable 119 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 •• Gaps in program coverage —including the high and increasing exclusion error in the CA program, and the restrictive SFA income eligibility threshold—need to be narrowed. The requirement that parents must be working or receiving unemployment benefits to receive CA conflicts directly with the declared purpose of the program, which is to support families raising children, particularly the relatively worse-off. •• The current interaction of the means-tested SFA, PFA, and CA raises several equity problems. The structure of poverty benefits suggests a framework of personal benefits (able to work, unable to work, and has children), whereby needs are assessed for the whole household or family through means-testing. This creates inefficiencies when members of the household fall between categories, specially between the SFA and the PFA. The reference group to assess needs is the household for SFA and the family for PFA and CA, but neither household nor family is well-defined, and equivalence scales are used inconsistently both between and within individual programs. 6.5  Reform Options The Government of the Republic of North Macedonia is aware of the inconsistencies in the social assistance system and seems committed to comprehensive reform. This might include reforming both targeted and categorical social assistance benefits to make the system more effective and efficient; here, the government could consider the following: 1. Consider discontinuing the Parental Allowance program, which could result in long-term savings and release resources for programs that target the poor. If it is not politically feasible to abolish it, the government might transform it into a small birth grant benefit, which would track the usual benefit architecture in EU countries. 2. Expand coverage of the poor through the main social assistance program in a fiscally sustainable way. A new guaranteed minimum income scheme based on reformed income and assets-tests could reach most current SFA and PFA beneficiaries. The benefit threshold and amount would need to be increased, and a new equivalence scale applied. This benefit would be given whenever the household income is below a set threshold, and it would equal the difference between the household income and the threshold. The GMI threshold for a four-member family should not be lower than 40 percent of the minimum wage to compare well with similar EU programs (Annex 10). Benefit levels would need to be regularly adjusted to reflect the costs of living. This would offer a comprehensive social safety net for the poor, irrespective of their labor market status. The government could consider incorporating current programs like the CCT or any future education- related allowances, and utility subsidies into benefit calculations. The new GMI formula should not discourage people from entering the labor market but rather reduce inactivity and incentivize transition to work. A mechanism disregarding income could be introduced in the new GMI scheme as a direct financial incentive to work. Finally, active labor market measures should be carefully tailored to the needs of this difficult-to-place group. Introducing the new GMI program should be part of a comprehensive social assistance reform that would tackle the inherent shortcomings of the current system: 120 Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable SOWING THE SEEDS OF A SUSTAINABLE FUTURE a. Eligibility for the reformed CA should follow the same rule as in the GMI program and there should be a single application procedure. Eligibility should be delinked from the employment requirement. The program should cover all households with children where the income falls below the established threshold (though that threshold should be higher than in the current program). Some economies of scale in terms of entitlement would be necessary. Also needed is a sustainable resolution of the current CCT program. b. A new social pension program (noncontributory), viewed as part of the social assistance reform, should be carefully designed. It too could build on the current PFA design. c. Social benefits and service provision should be consolidated to the extent possible and their administration should be simpler, more efficient, and integrated into a new social welfare information system for cash benefits and services that builds on the achievements of the CBMIS. The focus should be on human-centered design that facilitates inclusion by identifying bottlenecks, reducing bureaucratic hurdles, and making processes more efficient for the ultimate beneficiaries of social protection programs. Coordinated activities for outreach; identification of potential beneficiaries; prioritization, referral, monitoring, and tracking of beneficiaries; and better targeting (supported by integrating current cash benefit and services information systems) would bring efficiency gains in the administration and targeting of benefits and services and reinforce the coverage of social assistance and its poverty-reducing impact, thus making spending more efficient and effective. Chapter 6: Improving Social Assistance to Meet the Needs of the Most Vulnerable 121 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Chapter 7: Making the Grade on Education North Macedonia’s highly decentralized education system—with its poor accountability, few quality assurance mechanisms, and low and inefficient spending—undermines the equity of access to education and the quality of learning outcomes. Preprimary enrollment is low and tilted heavily toward wealthier children and those in urban areas. The quality of primary and secondary education quality has been declined for over a decade. Differences in learning outcomes are associated with disparities in language of instruction, teacher qualification, number of school shifts, teaching practices, disciplinary climate, and early childhood education and care access. Since 2006 education spending and efficiency have deteriorated till they are now near the bottom among international peers. Effective reform options might be to explore the feasibility of school optimization, revisit how schools are funded by introducing per-student formulas in preschool and higher education, and change the regressive preschool financing mechanism. Options to enhance quality of education might be to open up more access to preschool education, introduce a learning assessment system, and improve recruitment and teaching conditions for teachers. 7.1  Sector Overview North Macedonia has much room for improvement in terms of access to all levels of pre-university education, particularly pre-primary. Gross enrollment rates, which have not substantially changed since 2000, are below those of comparator countries, as are rates for each level. At 36 percent, enrollment in pre-primary education is the lowest for any level of education (and much lower than in peer countries), followed by tertiary at 42 percent (Table 7.1). And although secondary education is compulsory, more could be done to reach the coverage of peer countries (WB6, STEE7, UMCs and EU countries). Inequities in access to education persist by socioeconomic groups and location. According to the Multiple Indicator Cluster Survey, net attendance for 3–5-year-old children is 56 percent for those who come from advantaged backgrounds but almost zero for the poorest (Figure 7.1). Furthermore, in 2016, the share of 0–5-year-old children in kindergartens was about 32 percent for the East and Pelagonia regions but only 8 percent for the Northeast. All parents, regardless of household income, must cover about 21 percent of the cost of preschool programs (€25/month), an additional barrier for low-income families. Thus, socioeconomically advantaged households that can afford preschool are more likely to benefit from public resources. And more rich students are enrolled in primary and secondary school than the poorest children. From 2006 to 2016, secondary enrollment went up for students from all socioeconomic backgrounds, and increased the most among the poorest students, by approximately 18 pp.134 In contrast, middle-income students experienced the highest increase in tertiary enrollment, by about 12 pp. However, preschool and higher education are still mostly inaccessible to poor students, especially those in rural areas (Figures 7.1 and 7.2). 134 Enrollments rose because the Secondary Education Conditional Cash Transfers Program was introduced for children in families that receive SFA; about 20 000 students benefited. 122 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 7.1. Gross Enrollment by Level of Education, 2015 Percent Country Pre-primary Primary Lower secondary Upper secondary Tertiary North Macedonia 36 93 83 76 42 Montenegro 55 94 95 86 55 Serbia 58 101 102 91 58 Georgia 59 117 114 95 43 Croatia 63 98 104 93 69 Bulgaria 83 97 90 108 74 Latvia 88 100 116 123 67 Estonia 88 98 112 118 70 Albania 89 114 101 89 58 Lithuania 91 103 106 114 69 Slovenia 93 99 100 118 83 Slovakia 94 100 98 86 53 Averages Western Balkans (excl. MKD, BIH, KOS) 67 103 100 89 57 STEE7s 86 99 104 109 69 EU28 92 102 110 119 70 Source: ??? Note: North Macedonia’s education system consists of primary (9 years), secondary (general lasts 4 years and there are a 3 and a 4 TVET option). For benchmarking purposes, international definitions are adopted: Primary education covers only the first five grades, lower secondary covers the final four grades of primary education; and upper secondary corresponds to secondary education - general or vocational. Figure 7.1. Net Attendance Rates by Economic Quintile, Figure 7.2. Net Attendance Rates by Location, 2011 2011 Percent Percent 100 100 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Pre-primary Primary Lower secondary Upper secondary Pre-primary Primary Lower secondary Upper secondary education education education education education education education education JJPoorest JJSecond JJMiddle JJFourth JJRichest JJRural JJUrban JJTotal Source: UNESCO Institute for Statistics (UIS) and Multiple Indicator Cluster Survey. Source: UIS and Multiple Indicator Cluster Survey, 2011. Reasons for not enrolling in school are mainly demand-side. Many parents and guardians consider children aged 3–5 years to be too young for preschool; that is the explanation for why about 96 percent of 3–5-year-olds do not attend preschools. This behavior is consistent over time in both rural and urban areas. Cultural considerations, such as low expectations about the highest level of education a person can reach, are the main reasons given when 15–18-year-olds are not in secondary school. Poor learning conditions and the high cost of education also represent serious barriers to enrollment in both rural and urban areas. Supply-side factors, such as “the school is too far away” also matter but are rarely used to justify failure to enroll in secondary school. Chapter 7: Making the Grade on Education 123 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 North Macedonia’s primary school completion rates are high, but they are much lower for secondary school, which means that the working population has less education than in regional peers. At 1.4 percent, the proportion of out of school children in primary education is low. However, the completion rates fall drastically to 47 percent in secondary education, though this figure is similar to some regional peers (Figures 7.3 and 7.4). As a result, the educational attainment of Macedonians trails that of many EU countries (Table 7.2).135 However, the difference in educational attainment between younger (ages 25–34) and older (25–64) adults is one of the highest among peer countries, which attests to the country’s efforts in recent years to improve schooling. In general, younger population groups are also more likely to have a tertiary education. Figure 7.3. Completion Rates by Level of Education, Figure 7.4. Rate of Students Not Attending by Level of 2011 Education, 2011 Percent Percent 100 30 90 80 25 70 20 60 50 15 40 30 10 20 5 10 0 0 Primary Lower secondary Upper secondary Primary Lower secondary Upper secondary education education education education education education JJAlbania JJMontenegro JJSerbia JJNorth Macedonia JJAlbania JJMontenegro JJSerbia JJNorth Macedonia Source: UIS based on household surveys. Table 7.2. Educational Attainment of Macedonians Aged 25–64 and 25–34, 2016 Percent Population aged 25–64 years old Population 25–34 years old Less than primary, Upper secondary Less than primary, Upper secondary and post- and post- primary and Tertiary primary and Tertiary lower secondary secondary non- lower secondary secondary non- tertiary tertiary Lithuania 5 55 40 6 39 55 Slovakia 8 70 22 7 60 33 Latvia 9 57 33 12 46 42 Estonia 11 50 39 12 47 41 Slovenia 13 57 31 6 51 43 Montenegro 15 60 25 9 57 34 Croatia 17 60 23 6 61 33 Bulgaria 18 55 28 18 50 33 North Macedonia 32 48 21 19 49 32 Averages STEE7s 12 58 31 10 50 40 EU(28) 21 47 32 15 45 40 Source: Eurostat 2018. 135 North Macedonia has significantly fewer people with upper secondary and tertiary qualifications and far more with only basic education than other countries in the region. 124 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE Students who graduate from TVET institutions are more likely than those who graduate from gymnasia to pursue higher education. Overall, 56 percent of upper secondary students are in vocational schools—a proportion comparable to that of other WB6 countries and higher than STEE7 countries by 8 pp and EU countries by 13 pp. The proportion has also been quite steady over time. In 2016, 79 percent of TVET students enrolled in tertiary education, compared to 53 percent of students from general schools (Figure 7.5). This is problematic, given that TVET should predominantly prepare students for the world of work and gymnasia prepare them for higher education. The low transition of students from gymnasia to higher education means that almost half of those who leave the education system do not acquire the foundational skills to participate in the labor market. This situation is also suboptimal from a financial perspective, because the per-student cost of TVET is much higher than the cost of gymnasia. However, Macedonian parents and students have traditionally been more inclined to enroll in vocational schools. It also appears that students with lower achievement in primary education tend to enroll in TVET. Figure 7.5. Enrollees in Higher Education by Previous Figure 7.6. Student PISA Performance in Math, 2015 Secondary Program, 2011–16 Percent Percent 90 100 80 90 70 80 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 2011 2012 2013 2014 2015 2016 EST SVN LVA EU28 STEE7 LTU SVK HRV BGR MNE ALB WB6 MKD KOS JJVET JJGymnasia JJBelow basic JJBasic performance JJHigh performance Source: State Statistical Office. Source: World Bank staff calculations. Note: The Western Balkan average excludes Bosnia and Herzegovina and Serbia. The quality of education in North Macedonia is low and has been worsening. The 2015 scores for the Program for International Student Assessment (PISA) show that Macedonian students fall behind the EU average by more than 5 years of schooling in mathematics, and about 70 percent of the country’s 15-year-old students do not have even basic proficiency in mathematics and reading (Figure 7.6).136 Above only Kosovo, North Macedonia’s proficiency levels are the lowest among peer countries. The only national assessment in North Macedonia is Matura, the exit examination after secondary school, and student learning outcomes are no monitored to identify problems before that point. Moveover, in the Matura foundational subjects like mathematics are optional. Only 13 percent of Macedonian students take on the Matura subject of Math; 92 percent take the English subject. Preschool education, better interaction of traditional teacher-directed instruction and modern pedagogies, and an effective learning environment are crucial ways to enhance student performance. An analysis of PISA data shows that better learning outcomes in North Macedonia are associated with more years of preschool education, which holds true even controlling for student socioeconomic backgrounds (Figures 7.8 and 7.9). Simulations suggest that reaching a 100 percent preschool coverage 136 Proficiency refers to the ability to use everyday content and procedural knowledge to recognize or identify explanations of simple scientific phenomena; and with support, students can undertake structured scientific inquiries with no more than two variables (FISCAT). Chapter 7: Making the Grade on Education 125 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 would only raise North Macedonia PISA performance by 8 points. However, the effects of increasing the quality of preschool programs—such as through better teacher training and development —could be higher. Beyond preschool, promoting an effective learning environment through better engagement of teachers and students may stimulate student performance. Finally, students in schools that score in the top quintile for teaching practices outperform those in schools in the bottom quintile by 22 percent. Figure 7.7. Change in PISA Reading Scores, 2000–15 70 60 50 40 30 20 10 0 -10 -20 -30 Albania Spain Hong Kong (China) Switzerland Japan Canada Australia Poland Ireland Finland Thailand Iceland Peru Chile Russia Portugal Israel Indonesia Germany Brazil Norway Argentina* Romania Denmark Mexico Bulgaria OECD average-28 OECD average-24 Czech Republic France Greece Austria Korea United States Belgium Hungary Sweden New Zealand Latvia Italy North Macedonia Source: World Bank staff calculations using 2015 PISA data. Figure 7.8. Difference in PISA 2015 Science Scores Figure 7.9. Simulation of North Macedonia’s PISA Between Students with less than One Year of Score Gains from Increasing Preschool Attendance Preschool and Students with more than One Year PISA 2015 science points PISA 2015 science points 50 10 40 8 30 6 20 4 10 2 0 0 Zero year Up to 1 year 40% 60% 80% 100% JJAfter accounting for ESCS JJBefore accounting for ECS Source: World Bank staff calculations using PISA data, 2015. Additionally, language of instruction, teacher qualifications, the poverty rate in the municipality, and the number of shifts offered in a school are closely correlated with Matura performance. More teachers with a university degree is associated with better school examination scores (Annex 4 Table A4.1). Also, as expected, schools with two shifts trail schools with a single shift; however, students in double-shift schools are more likely to successfully finish the school year.137 Larger schools and one- shift schools outperform in Matura, which may reflect the differential effects of school instructional time—the number of hours of instruction in compulsory general education is low and uneven (as discussed below). In addition, qualified teachers are not unevenly distributed across municipalities, which needs to be addressed. For instance, municipalities that are predominantly Albanian have fewer teachers with a bachelor’s degree or higher. 137 To remove all possible unobserved confounding variables, the models included municipality fixed effects. 126 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 7.10. Teachers with Tertiary or Higher Education Figure 7.11. Average Score in State Matura Examination across Municipalities, 2016/17 by Municipality, 2016/17 Percent 100 90 80 70 60 50 0 20 40 60 80 100 Percent of Macedonian students Source: MOES 2016/17 data and World Bank staff calculations. One strength of the education system in North Macedonia is its relative equity in terms of the learning outcomes of both rural and urban students from different socioeconomic backgrounds—though this may be due to poor education quality across the board. The difference in PISA performance between students in the top and in the bottom quintiles is much less in North Macedonia than in some peer countries (Figures 7.12, 7.13 and 7.14). The country has little school segregation by socioeconomic background, which means all students have more opportunity to learn. This is notable because breaking the influence of socioeconomic background on learning is a central issue for education systems today. Another indicator of North Macedonia’s relative parity is the minimal achievement gap between rural and urban schools (Figure 7.15). Urban students scores are only 5 percent higher than those of rural students. Finally, in science North Macedonia’s equity is above the EU average. Figure 7.12. Math Scores by Student Economic Status Figure 7.13. Socioeconomic Index for Students and Schools Correlated (weighted) Percent Percent 18 60 16 50 14 12 40 10 30 8 6 20 4 10 2 0 0 8 E LVA S LVA EU28 S SVK BG R EU2 V HR 7STE LTU SVN EST MKD KO MNE SVK BG R SVN 7STEE LTU EST V HR MK D KO MNE Source: World Bank staff calculations using PISA data 2015. Source: World Bank staff calculations using PISA data 2015. Chapter 7: Making the Grade on Education 127 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 7.14. Difference in Science Scores, Top and Figure 7.15. Difference in Science Scores, Urban and Bottom Quintile Students, 2015 Rural Students, 2015 150 1.4 60 1.20 1.3 50 1.15 120 1.2 40 1.10 90 1.1 30 1.05 60 1.0 20 1.00 30 0.9 0.95 10 0 0.8 0 0.90 8 LVA EST KOS LTU KOS 7STEE LVA MKD MNE EU28 HRV BG R SVK EU2 7STEE SVN HRV LTU MKD MNE BG R SVK EST SVN JJDifference, lhs QQRatio, rhs JJDifference, lhs QQRatio, rhs Source: World Bank staff calculations using PISA data 2015. Source: World Bank staff calculations using PISA data 2015. Note: For the ratio, mean scores in urban areas is divided by mean scores in rural ones. Youth unemployment in North Macedonia is very high, and indicates a skills mismatch in the labor market. In 2016, the country had the highest general unemployment rate among peer countries, 24 percent, against the EU average of 8.8 percent. And the unemployment rate for those aged 15–24 was double that of the rate for job-seekers over 25. Recent World Bank research on labor and skills demand in North Macedonia shows a mismatch between the skills students acquire in the education system, particularly TVET, and private sector needs, with companies complaining about the quality and availability of skills despite high unemployment.138 Although labor demand is relatively subdued in North Macedonia in general, the firms that do wish to expand report they cannot fill vacant positions because they cannot find workers with the right skill profiles. The majority of firms also report that their current workforce lacks the skills to do their jobs effectively. Hence, the lack of skilled labor inhibits both individual opportunities and productivity growth. The skills gap refers not just to job-specific Figure 7.16. PISA Math Scores and Youth Unemployment by Country, 2015 technical skills but also generic skills. The jobs PISA mathematics 2015 score and occupations for which employers emphasize 575 suffering skills gaps generally require a range of higher-order cognitive and socio-emotional as 525 EST well as technical skills. This suggests that the skills SVN problem is not primarily sectoral or occupational SVK 475 LTU LVA HRV one, but that the quality of the skills imparted 425 BGR through the education and training systems is MNE GEO inadequate. 375 325 MKD 0 10 20 30 40 50 Youth unemployment, percent total labor force aged 15–24, 2014 or latest Source: WB based on ILO and PISA. 138 Koettl-Brodmann, Johannes; Johansson De Silva, Sara; Kupets, Olga; Naceva, Bojana. 2017. Looking for skills in the Former Yugoslav Republic Macedonia (English). Washington, D.C: World Bank Group. 128 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE 7.2  North Macedonia’s Spending on Education in International Context 7.2.1  Public Spending Trends and Composition North Macedonia’s educational system is partially decentralized, with its two levels of government sharing responsibilities for it. Decentralization was designed in two phases: first, assets were transferred to municipalities—along with the associated maintenance costs (e.g. heating, energy, materials and services. In education, property rights in school buildings, land, and equipment were transferred. Second, the management of human resources and their costs were also transferred to LGs (municipalities). At that point, municipalities became free to adjust human resources and maintenance costs to their needs. The management of preschool, primary, and secondary schools is now the responsibility of municipalities, and the central government focuses on tertiary education,139 though it retains responsibility for capital investments in primary and secondary schools. Education spending is mainly handled by LGs. In 2016, the central government provided 90 percent of all the resources allocated to education, of which 58 percent went to municipalities as block or earmarked grants (see Chapter 3 for full details on the intergovernmental transfers system). Municipalities add another 5 percent from their own resources and thus manage 62 percent of all public spending on education—mostly pre-university (Figure 7.17).140 CG spending accounts for the remaining 38 percent, and more than half that amount goes to higher education. For preschool, municipalities contribute about 29 percent of total preschool spending. Mostly, these resources go to capital and maintenance costs, since salaries are absorbed by the CG MLSP through transfers to municipalities. Although municipalities finance only 3–4 percent of the cost of primary and secondary education, they handle 89 percent of the spending at these levels, suggesting that they depend heavily on centrally-provided resources. Figure 7.17. Public Spending on Education, 2016 A. By government level B. CG spending by functional C. LG spending by functional classification classification Higher education Local Auxiliary services 62% provided to education Pre-school and general elementary Pre-school and general elementary Upper secondary education Education not classified by levels Upper secondary Research and development education in education Central Education (not belonging 38% to other classes) 0 20 40 60 0 20 40 60 80 Source: World Bank Boost database. 139 Since decentralization municipalities financing of preschool education through local budgets except for salaries, which are funded directly by the Ministry of Education and Science. In 2016, for instance, central government financed 99.7 percent of preschool personnel costs and municipalities financed 76 percent of non-personnel recurrent costs and 95 percent of capital expenses. Expanding preschool education will imply a higher fiscal effort for less developed municipalities. 140 Pre-university education comprises preschool, primary, and secondary education. Chapter 7: Making the Grade on Education 129 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table 7.3. Entrance Age and Duration of Compulsory Education in North Macedonia is largely public, Education, 2016 with universal access to primary and secondary Country Entrance Age Duration education. The education system serves about Jamaica 6 6 370,000 students (about 17.7 percent of the total Croatia 7 8 population, of which 29 percent attend primary Serbia 7 8 schools, 45 percent secondary schools, and 17 Albania 6 9 percent tertiary schools. About 8 percent attend Bosnia 6 9 preschool. All students in primary education in Estonia 7 9 North Macedonia are enrolled in public school; Georgia 6 9 about 97 percent of preprimary students are in Lithuania 7 9 public schools, as are 98 percent of secondary Montenegro 6 9 school students.140 Moreover, the duration of Slovenia 6 9 compulsory education in North Macedonia Jordan 6 10 (13 years) is the highest among peer countries, Slovakia 6 10 with only Paraguay allocating the same amount Bulgaria 5 11 of time (Table 7.3). Enrollment in private schools Latvia 5 11 started in 2002 for upper secondary and in 2012 for pre-primary but is still minimal. The Mauritius 5 11 importance of the private sector is most apparent Paraguay 5 13 most in tertiary education, where between 2006 North Macedonia 6 13 Source: UIS, 2017. and 2015 its share of enrollment grew from 4 to 14 percent. Comparator group countries have similarly low private sector participation in all levels of education (Table 7.4). Table 7.4. Enrollment in Public Institutions by Level of Education, 2016 Percent Lower Upper Country Pre-primary Primary Secondary Tertiary Secondary Secondary Bosnia and Herzegovina 84 98 99 97 98 80 Croatia 85 100 99 96 98 93 Latvia 93 98 98 97 97 8 Albania 94 94 95 90 92 85 Slovakia 95 93 92 85 89 83 Slovenia 97 99 100 97 98 87 Estonia 97 94 95 96 96 15 Lithuania 97 98 97 98 97 90 Montenegro 97 100 100 100 100 North Macedonia 98 100 100 97 99 85 Serbia 98 100 100 99 100 87 Bulgaria 99 99 97 96 96 85 Georgia 100 90 91 88 90 70 141 Compulsory education includes 9 years of primary, and 3-4 years of secondary. 130 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 7.4. Enrollment in Public Institutions by Level of Education, 2016 (continued) Lower Upper Country Pre-primary Primary Secondary Tertiary Secondary Secondary Averages EU average (22) 75 89 87 86 87 76 Western Balkans (excl. Kosovo) 93 98 98 96 97 84 STEE7s 94 97 97 95 96 66 Source: UIS, 2017. North Macedonia has steadily reduced its spending on education (Figure 7.18), which is currently than the EU and peer country averages. Between 2011 and 2016, public spending on education as a share of GDP slipped down from 4.62 to 3.77 percent (0.85 pp), and as a share of total government spending from 13.30 to 11.60 percent (1.7 pp). A comparative analysis of average spending on education as a share of GDP for 2013–2015 reveals that North Macedonia spends more than most WB6 countries by a modest margin, but lags behind the STEE7 by an average of 0.6 pp and behind the EU average 1.1 pp (Figure 7.19). Figure 7.18. Spending on Education, 2011–16 Figure 7.19. Spending on Education Compared, 2013–15 Percent of GDP Total spending Percent of GDP 6 14.0 6 5 13.5 5 13.0 4 4 12.5 3 3 12.0 2 2 11.5 1 11.0 1 0 10.5 0 2011 2012 2013 2014 2015 2016 SVN EST LVA EU28 STEE7 LTU V HR MK D SVK SRB BG R ALB ÂÂPercent of GDP ---- Total government expenditure, rhs Source: GDP data: Examination center of North Macedonia; LG spending: Boost database. Source: UIS and Examination center of North Macedonia. Note: For spending as a percent of total government expenditure only CG spending is Note: Georgia and Paraguay data are for 2012. included. North Macedonia spends less on pre-university education than most peer countries  (Figure 7.20). While its per student preschool spending is at the same level as the EU, North Macedonia’s preschool coverage is much lower. The per student spending for primary education, as a proportion of GDP per capita, is about 4 pp lower than both STEE7 and EU averages, despite comprising about 52 percent of total students (Table 7.5). Moreover, there are vast disparities between municipalities, many of which spend far less than the national per student average. In primary education, per-student spending may range from MKD 30,000 to MKD 120,000. This in part arises from the funding formula, which seeks to ensure that less-populated municipalities receive more funding to cover all their expenses. Furthermore, among peer countries North Macedonia has one of the highest proportions of spending that is not allocated to a specific level of education. These expenditures go to services auxiliary to education. The CG provides these services; since they account for almost 5 percent of total spending on education, their effectiveness and efficiency deserve review.142 142 Among them are subsidies and transfers in the form of scholarships, lodging for upper secondary students, and payment to student homes; about 15 percent of the auxiliary services goes to strengthen infrastructure. Chapter 7: Making the Grade on Education 131 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 7.20. Spending on Public Education Spending North Macedonia also spends less on tertiary by Level Compared, 2016 Percent education than peer countries. Its public and 100 private spending on tertiary education was 90 1.32 percent of GDP in 2011, below the OECD 80 average. Estimates based on data provided by 70 MOES put public investment in tertiary education at 0.4 percent of GDP in 2015—significantly lower 60 50 40 than the OECD average of 1.1 percent, and the 30 1.3 percent invested by Slovenia or even the 20 0.7 percent invested by Croatia (Salmi 2016). As a 10 result, public universities lack the infrastructure 0 R LVA 7 GEO LTU SVK SVN 7 EST D and resources they need (buildings, laboratories, BG STE E EU1 MK JJPreprimary JJPrimary JJSecondary JJTertiary JJNot specified by level qualified personnel) to improve their quality. Source: UIS and Boost data. Note: Expenditures “not specified” go to auxiliary services, such as scholarships, payments to student families, room and board, and infrastructure. Table 7.5. Per-Student Public Spending Compared, 2013 or Latest, Percent of GDP per Capita Pre-primary Primary Lower Secondary Upper Secondary Tertiary North Macedonia 18.18 18.36 18.36 23.55 29.48 Lithuania 18.52 16.35 16.54 18.64 Slovak Republic 15.98 19.42 18.21 20.20 21.52 Estonia 19.71 21.50 18.43 30.68 Bulgaria 31.82 22.36 23.86 19.98 16.08 Latvia 20.50 26.37 25.88 22.78 21.67 Slovenia 21.52 28.87 30.87 22.41 21.15 Albania 7.36 4.01 12.99 Croatia 23.21 25.92 Averages STEE7 22.45 22.54 22.78 20.51 22.24 EU-28 excluding GRC 16.93 21.82 25.34 24.27 28.96 Source: UIS and World Bank staff calculations for North Macedonia. Note: In North Macedonia lower secondary is part of primary education. Figure 7.21. Public Spending on Education by Spending in public institutions is skewed toward Economic Classification, 2016 Percent current expenses, to the detriment of capital 100 expenses. The CG is responsible for capital 90 spending for schools. Though a third of school 80 buildings need major repairs, reconstruction, 70 adaptation, or rehabilitation, only 5 percent of total education spending is capital. The average for 60 50 40 peer countries is 8 percent (Figure 7.21). Also, the 30 composition of current spending favors salaries 20 rather than nonpersonnel current expenses. 10 Salaries and other staff compensation accounts 0 LVA SVK EST STEE7 D SVN V R 0 ALB SRB for 71 percent of total spending on pre-university MK HR BG EU2 JJAll staff compensation JJCurrent other than staff compensation JJCapital education but about 50 percent for tertiary. About Source: World Bank staff calculations based on UIS and BOOST data, 2016. 40 percent of recurrent costs other than salaries 132 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE is for food products in preschool education, 25 percent of nonpersonnel primary recurrent expenses goes to heating and transportation, and more than 50 percent of secondary recurrent costs other than salaries goes to transportation services, which almost all municipalities outsource. 7.3  Efficiency of Spending North Macedonia’s education system is undermined by the inefficiency of public spending. Two data envelopment analyses143 were performed for North Macedonia and peer countries to measure the relative efficiency of their spending on education as a share of GDP between 2011 and 2015 using lower secondary enrollment and 2015 PISA test performance as outcomes (Figures 7.22 and 7.23). The DEAs found that North Macedonia had the lowest efficiency in terms of ensuring access to lower secondary education—countries spending similar amounts had on average 20 percent higher enrollment. North Macedonia is also the least efficient in providing quality education—countries with similar spending average nearly 34 percent (112 points) higher in PISA tests. All EU countries were considered as part of the benchmarking exercise, due to North Macedonia’s commitment to meet the EU membership requirements. However, the results are consistent only when selected peer countries are part of the analysis. The tertiary education system is also inefficient. In 2010 only 38.8 percent of all tertiary students graduated on time. High dropout rates and long average times to completion characterize North Macedonia’s tertiary education.144 Figure 7.22. Efficiency Scores Using Net Lower Secondary Education Enrollment as Output, 2015 1.0 0.9 0.8 0.7 0.6 0.5 Estonia Croatia Albania Spain Slovenia Ireland Netherlands Poland Finland North Macedonia Bulgaria Moldova Mauritius Belgium Portugal Luxembourg Romania Hungary Denmark Cyprus France United Kingdom Malta Lithuania Serbia Georgia Sweden Latvia Italy Source: World Bank staff calculations. 143 A data envelopment analysis (DEA) is a nonparametric method for measuring the relative efficiency of decision-making units (DMUs). It is based on how close or far a DMU) is from the “efficiency frontier” and is built from a combination of the inputs and outputs of best-performing DMUs. Thus, DEA provides a relative measure of efficiency, since best-performing could have room for improvement. An output maximization approach is taken in that the analysis focuses on how much the output can be expanded without changing the input used. One important limitation of DEA is that it does not take into account a DMU’s unique circumstances, which may demand different intensities in the use of resources. 144 See Salmi 2016. Chapter 7: Making the Grade on Education 133 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 7.23. Efficiency Scores Using PISA 2015 Math Scores as Output 1.0 0.9 0.8 0.7 0.6 0.5 Albania Croatia Spain Estonia Slovenia Ireland Poland Finland Netherlands North Macedonia Kosovo Moldova Cyprus Georgia Malta Lithuania Bulgaria Portugal United Kingdom France Hungary Sweden Austria Belgium Slovak Republic Germany Denmark Luxembourg Czech Republic Romania Latvia Italy Source: World Bank staff calculations. The system has become less efficient over time Figure 7.24. PISA Performance and GDP per Capita, as measured by performance on the 2000 and 2000 and 2015 PISA score 2015 PISA tests ( Figure 7.24). North Macedonia 575 already spends very little on education compared to peer countries, which represents a challenge 525 EST for sustaining improvements in education. HRV LVA SVN However, even at such low investment levels, 475 LVA LTU SVK peer countries like Bulgaria and Georgia achieved BGR MNE BGR 425 much better outcomes. JOR GEO 375 MKD 2000 KOS MKD 2015 325 0 10 20 30 40 50 60 70 GDP per capita, thousands US$, 2011 PPP QQ2000 QQ2015 QQPeers 2000 QQPeers 2015 7.3.1  Efficiency of Resource Use Source: PISA database and World Bank staff calculations. The scarcity of capital resources in North Macedonia means that school facilities are used very intensively. Among primary schools, 47 percent have double shifts, and 1 percent have triple shifts. Among secondary schools, 72 percent have double shifts, with little variation between school types. Furthermore, 69 percent of North Macedonia’s secondary schools have 400 or more students. While doing so may be desirable in the long run, the costs of moving to a single-shift system are very high. It is not recommended that the country use its scarce resources to eliminate double-shift schools—other investments would do more to improve the quality of education. However, it is recommended that at least the small number of triple-shift schools, which currently 5 percent of secondary students attend, be converted into double-shift schools: triple-shift schools fail to provide sufficient instructional time in any shift. As elsewhere in the region, however, double-shift schools limit the scope for increasing instructional time, which is already relatively minimal in North Macedonia (below). A drop in the number of Macedonian students has been accompanied by a rise in the number of teachers, which has reduced the student-teacher ratio (STR), though it is still about the same as in many peers  (Table 7.6). The pre-university population (aged 3–18) has been falling at a rate of 1.8 percent annually since 2000 which represents a total drop of 23.6 percent. As expected, enrollment has declined proportionately. However, the numbers of primary and lower secondary schools, classes, 134 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE and teachers have not declined (Figures 7.25 and 7.26). In fact, since 2002–03 the number of primary and lower secondary students has gone down 19 percent and the number of schools has stayed the same, the number of teachers has gone up by 32 percent. Moreover, municipal STRs vary widely. For example, Vraneshtica’s primary STR is 3:1 and Shuto Orizari’s is 18:1. This is partly due to demographic diversity, which means that classes are held in different languages, and because of the number of satellite schools. The differences represent an opportunity to explore the feasibility of optimizing the school network. Table 7.6. Students-Teacher Ratios by Country, 2014–15 Primary Lower Secondary Upper Secondary Secondary Tertiary Georgia 9.0 7.2 7.3 7.2 7.1 Latvia 11.1 7.3 8.9 8.0 12.6 Estonia 11.5 7.9 8.3 8.1 Lithuania 12.9 7.5 8.9 7.9 10.8 Croatia 13.7 7.8 6.4 7.0 10.1 North Macedonia 15.2 8.1 11.7 9.6 18.1 Serbia 15.2 8.4 8.6 8.5 22.5 Slovakia 15.2 11.0 11.4 11.2 15.5 Slovenia 16.9 7.3 13.0 10.1 13.6 Jordan 16.9 18.9 9.4 14.6 22.5 Bosnia and Herzegovina 17.2 9.2 11.3 10.1 11.3 Bulgaria 17.7 13.4 13.1 13.2 12.3 Albania 18.6 11.9 16.2 13.5 23.6 Mauritius 18.8 12.9 Jamaica 21.9 16.4 EU(23) 12.6 10.1 11.4 10.7 14.7 Source: UIS. Figure 7.25. Evolution of the Primary and Lower Figure 7.26. Evolution of the Upper Secondary School Secondary School Network, 2007–16 Network, 2007–16 120 130 110 120 110 100 100 90 90 80 80 70 70 60 8 9 0 1 2 3 4 5 6 7 60 8 9 0 1 2 3 4 5 6 7 7/0 08/0 009/1 010/1 2011/1 012/1 2013/1 014/1 015/1 016/1 7/0 08/0 009/1 010/1 2011/1 012/1 2013/1 014/1 015/1 016/1 200 20 2 2 2 2 2 2 200 20 2 2 2 2 2 2 ÂÂSchools ÂÂStudents ÂÂTeachers ÂÂClasses ÂÂSchools ÂÂStudents ÂÂTeachers ÂÂClasses Source: State Statistical Office. Note: Data are presented as index numbers: each value represents the magnitude of change compared to the baseline (year=2007/08). Subtracting 100 from each value shows the percentage change. 2007/08=100 Chapter 7: Making the Grade on Education 135 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 The continuous increase in teachers despite declining student numbers is not due to changes in curriculum; the causes appear to be structural. In elementary education, only 9 municipalities have cut the number of teachers in response to lower enrollment, but even in these, for every four classes reduced (to about 18 students each), only one teaching position was eliminated. In fact, in most municipalities, the loss of 15 students resulted in the addition of one teacher. Even though Macedonian STRs are still comparable to those of peer countries, North Macedonia must monitor this trend to avoid over-hiring. Despite comparable STRs, North Macedonia Figure 7.27. Hours of Instruction Time in Compulsory General Education Compared averages fewer total hours of instruction in EU28 compulsory general education than peer SVK countries. Though the length of primary and LTU lower secondary education (9 years) is the same EST SVN in North Macedonia and most peer countries, MNE its accumulated instruction time is about LVA 1,000 hours less than in the EU—equivalent to MKD SRB almost a year of schooling145—and is the lowest BIH among peer countries (Figure 7.27). These BGR significant differences make it likely that lack of HRV instructional time partly explains the relatively 0 2,000 4,000 6,000 8,000 Hours poor performances of Macedonian students in JJPrimary education JJLower secondary education international learning assessments. Source: Eurydice and OECD. North Macedonia teachers’ salaries are competitive with peer countries: The minimum and maximum annual basic gross salaries for primary education , as a percentage of GDP per capita, are the fourth Figure 7.28).146 Salaries are also high for lower and upper secondary teachers, but highest in Europe ( preschool teacher salaries are the exception—they are much lower, at about 60 percent of primary salaries and 55 percent of secondary school salaries. This reinforces the already significant gap between the salary of teachers in basic and higher education. Salaries in education are competitive with those of other sectors in North Macedonia, except for Skopje, where they are below the mean ( Annex 6 Table A6.1). This finding is consistent with a recent survey of Macedonian teachers, who mostly reported high levels of satisfaction with their possibilities for career advancement in the profession. However, many also expressed frustrations with the expectation that they adapt to frequent changes in the national curriculum without proper training or the flexibility to modify the curriculum based on differentiated learning needs.147 145 Assuming that the time of instruction per day is 6 hours in both North Macedonia and the EU, 1,000 hours is equivalent to 33 weeks of instruction, which is about one year of schooling. 146 Comparisons of wage rates relative to officially estimated GDP need to be made with care, however, because North Macedonia’s nominal GDP may be understated. The country makes a modest adjustment for the non-observed economy; adjustments in comparator countries are more realistic. 147 See Mickovska-Raleva et. al. 2017. 136 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 7.28. Annual Basic Gross Teacher Salaries Compared Percent of GDP per Capita 300 250 200 150 100 50 0 Scotland Estonia Croatia Slovenia Spain (a) Spain (b) Poland Ireland Iceland Northern Ireland Belgium (Flemish) Finland Turkey Lithuania Romania Slovakia Czech Republic Hungary Bulgaria Sweden Norway (a) Norway (b) England and Wales Greece France (teachers in Lycées) Belgium (French) Austria (a) Belgium (German) Austria (b) Luxembourg Europe The Netherlands Malta Denmark Serbia Cyprus Portugal Latvia Montenegro Germany North Macedonia Bosnia and Herzegovina Italy (a) JJMin JJMax Source: Eurydice. 7.4  Financing Mechanisms Municipal spending on education is largely financed by transfers from the central government. Earmarked and block grants cover salaries and maintenance costs, and municipalities are automatically allocated 3 percent of PIT and 3 percent of VAT revenues, which are both collected at the national level. About 95 percent of LG financing thus comes from intergovernmental transfers. Municipalities can top off resources from the CG, which could create inequalities in the amounts allocated to education, particularly in subsectors like preschool, which rely more on LG funds (Figure 7.29). Figure 7.29. Government Education Spending by Level Figure 7.30. Per-student Spending and Class Size by and Source of Financing, 2016 Education Level and Municipality, 2016–17 Per student cost 140 Total 120 Higher education 100 Upper secondary education 80 Primary education 60 40 Preschool education 20 0 20 40 60 80 100 5 10 15 20 25 30 35 Percent Class size JJCentral JJTransfers from central to LG QQPrimary and lower secondary education QQUpper secondary education JJLocal governments JJOther LG sources, donations and loans Source: BOOST database. Source: Boost database and MOES. Note: University revenues are part of the CG budget. There are several formulas for allocating intergovernmental transfers in pre-university education  (Annex 5). There are per-student formulas, including two sets each for block and earmark grants— one for primary and one for secondary education. Transfers to primary and secondary schools for recurrent spending are set based on a per-student formula. According to the ordinance governing the distribution of block grants, the formulas consist of a lump sum for each municipality plus a standard Chapter 7: Making the Grade on Education 137 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 amount per pupil, with adjustments for population density, number of students in subject classes (primary), number in vocational training classes (secondary) and total students with special needs.148 The formula gives a higher amount per student to small municipalities to compensate for the higher costs arising from their lack of economies of scale (Figure 7.30). The result is a relatively large number of small schools—about 44 percent of primary schools have fewer than 100 students. The choice of factors favored in the funding formula is the prerogative of the government. However, problems arise when the MOES, which handles capital investments for primary and secondary schools—does not have enough funds to keep the many small schools in good condition. Tertiary and preschool are still financed through inputs and line-item budgets. The input-based formula in North Macedonia is linked to the number of students enrolled in both preschool and tertiary. In higher education, funding is not linked to university performance. Moreover, the amount of public money that each university receives depends on historical trends and the ability of each rector to negotiate with the MOES and the MOF, not on objective criteria. Thus, there is no incentive for preschools to deliver quality or for universities to invest in research—0.4 percent of the education budget goes to research compared to 2.4 percent in the OECD (Salmi 2016). While North Macedonia has made some advances in financing mechanisms of primary and secondary education to make transfers more transparent, preschool and tertiary are stuck in the traditional financing scheme. In North Macedonia, the per-student allocation that each university receives is not uniform. Public higher education is financed by a combination of state budget and other sources. State financing depends on the number of approved study programs, study groups, departments, and courses; number of full-time students enrolled for the first time under the state quota in each academic year; the structure of employees; the numbers of students who graduate and of administrative staff, etc. (Blazevski, n.d.)—although depending on the available state funds, funding usually tracks historical allocations. Additional sources of funding for universities outside of the state budget include industry, and research., which is treated as self-financing, and donor programs (Salmi 2016). As the data reveal, the Universities of Skopje and Ohrid get significantly more than the national average (Table 7.7). Moreover, the per-student allocation is almost seven times higher at Ohrid, which receives the largest per-student allocation. As a result, only half of the students qualify for public subsidies as a result of their high school grades. Table 7.7. Evolution of Government Per-Student Allocations, 2010–15 University 2010 2011 2012 2013 2014 2015 St. Kliment Ohridski (Bitola) 26,052 26,537 28,857 30,927 33,667 35,345 State University in Tetovo 13,443 13,134 12,411 15,215 13,409 18,261 GoceDelcev University (Stip) 26,215 23,224 22,313 23,535 23,613 25,789 Sts. Cyril and Methodius (Skopje) 49,757 53,070 42,787 42,608 45,387 40,907 University of Information Sciences & Technology 531,010 224,617 168,141 178,611 165,127 122,465 (Ohrid) South-Eastern European University n.a. n.a. n.a. n.a. n.a. 10,989 Source: Salmi 2016. 148 While the current funding formulas protect schools with low density and special needs students, they do not specifically factor in socioeconomic weight. Usually though, students in low density areas are relatively poorer. 138 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE The current decentralized system lacks sufficient quality and accountability measures. Though it represents a cumbersome process, decentralization seeks to increase education performance and transparency and promote equity and efficiency. However, it can have negative effects if political dynamics undermine accountability, or if LGs have inadequate capacity or weak incentives to adhere to the goals and guidelines set forth in the decentralization procedures (Smoke 2015). In North Macedonia, funding for schools varies widely. The MOES establishes upper and lower buffers annually to restrict variation in per student spending, but greater transparency is required to ensure the total pool of funds is within the MOF-approved limits to ensure equity across municipalities. Finally, North Macedonia has minimal accountability and few quality assurance (QA) mechanisms, which, combined with ineffective teaching equates to poor education quality and considerable inefficiency. 7.5 Intra-national Assessment of North Macedonia's Education Spending and Outcomes The poorest municipalities tend to perform worse in the Matura examinations, but they do not spend more per student and are highly dependent on block grants. A DEA was used to measure the relative efficiency of LG education spending. Though the poorest municipalities underperform in the Matura examination, their per student spending does not differ from those of other LGs. This is not surprising, because it is more expensive to provide good-quality education in schools that serve poorer communities. Moreover, the proportion of funding from block grants tends to be higher in poorer municipalities, because they find it harder to raise their own funds, which could allow them to bridge the achievement gap. Figure 7.31. Matura Results and Poverty by Level of Figure 7.32. Poverty and Proportion of Block Grant Education and Municipality, 2016–17 Funds by Municipality, 2016–17 Performance in Matura Poverty 4.5 0.06 4.0 0.04 3.5 0.02 3.0 2.5 0 0 0.01 0.02 0.03 0.04 0.05 0.06 85 90 95 100 Poverty Proportion of funds from block grants Source: World Bank staff calculations with data from the State Examination Center, State Statistical Office, and BOOST. The structure of spending appears to matter for efficiency in completion. Four of the five least efficient municipalities in primary and lower secondary education are in the Southeast region (Figures 7.33 and 7.34). Vasilevo, the least efficient, would have to raise its completion rate by 6 pp, from 92 to 98 percent, to perform as well as other municipalities with similar class sizes and per student spending. A regression of the determinants of efficiency scores on municipality-specific characteristics found that municipalities that spend a larger share on personnel are the most efficient, suggesting that they either have more personnel or more qualified personnel. Chapter 7: Making the Grade on Education 139 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 7.33. Efficiency Scores for Primary Completion Figure 7.34. Efficiency Scores for Secondary Rates, 2016 Completion Rates, 2016 Source: World Bank staff calculations using BOOST data. Note: As outliers, Plasnica and Centar Zupa were not included. With regard to Matura scores, there is considerable heterogeneity in municipal efficiency levels. Municipalities with more students benefit from some economies of scale and are most efficient. Other municipalities must put in more effort to increase their efficiency. Dolneni, the least efficient, would need to improve its performance by 34 pp to catch up to municipalities whose spending and average class size are similar (Figure 7.35).149 Controlling for regional fixed effects, municipalities with a higher STR tend to perform better in terms of their per student expenditure spending. Because the STR correlates with class size, economies of scale might allow for cost reductions—though this should not be done at the expense of learning. Figure 7.35. Efficiency Scores for Matura Examinations Source: MOES data and World Bank staff calculations, 2018. 7.6  Reform Options North Macedonia’s highly decentralized system—with its minimal accountability, low QA mechanisms, and low and inefficient spending—operates to limit access and education quality. Since 2006 education spending and efficiency in North Macedonia have deteriorated; they are now very low compared to peer countries. Education quality, as measured by standardized learning assessments like the PISA 149 Of the seven most efficient municipalities by Matura scores, three are in the East region and two in the Southeast. 140 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE and Matura tests, is inadequate, and student performance has been declining.150 A major problem is the lack of quality mechanisms throughout the entire education system, from preschool to university. Education resources are inefficiently allocated: The numbers of schools, class sizes, and teachers have not decreased to compensate for falling enrollment. Meanwhile, infrastructure investment falls short of the need. These inefficiencies should be addressed before more money is injected into the system. Some reform options follow. Short term  To improve efficiency: 1. Conduct a feasibility study on optimizing the school network that looks for better ways of using facilities and the extent to which different optimization options are possible. The secondary school STR ratio is especially low. Some primary schools have an excessive number of classrooms available; others are overcrowded, some requiring more than two shifts. Optimizing the school network offers an opportunity to make education spending more efficient. Students in larger schools outperform in Matura. Optimizing small schools could release facilities for preschool and secondary programs and perhaps reduce the need for some auxiliary services, such as lodging and transportation. However, it is important to be aware of the equity issues that may arise. 2. Re-evaluate current primary and secondary per-student funding formulas to improve efficiency. Both formulas aim to protect low-density schools—which usually have low-income students and are likely to be allowed higher per student amounts. However, this creates smaller class sizes. Meanwhile, education funds do not adequately cover the capital investments required to keep facilities in good condition. In other words, the current formulas create an unsustainable financial situation. 3. Introduce per-student funding for preschool and higher education. Per-student formulas provide a transparent and more equitable way of distributing resources. Moreover, the current input-based financing for preschool and tertiary education does not incentivize quality. A review of the preschool and higher education financing is highly advisable to overcome issues not only of efficiency but also of equity and quality. The current governance model of universities is overburdened with provisions that are often inconsistent; as a result, resources are not managed efficiently.  To improve quality: 4. Use the Matura exam more effectively. Make mathematics a required subject to increase student and teacher efforts, which will help to promote education quality. To further increase returns on investment in the Matura exam, the MOES could analyze education outcomes in each school to link school resources (inputs) with learning scores (outputs) and better inform education policymaking. 5. Increase instructional time by lengthening the school day, not by adding more subjects, but by increasing the time spent on quality learning and on remedial or extra-curricular activities. Also, convert triple-shift to double-shift schools to ensure that students get enough instructional time. 150 About two-thirds of Macedonian students performed below basic proficiency in all three PISA subjects. Chapter 7: Making the Grade on Education 141 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 It would also be advisable for teachers to stay in schools to prepare for classes, provide mentoring, and engage in other activities related to education. Medium term  To improve teaching practices and teacher effectiveness: 6. Improve the learning environment through better teaching practices. Countries with better PISA outcomes and lower income levels than North Macedonia have in common a learning environment that uses blended learning (mostly teacher-directed instruction, complemented with problem solving). However, in North Macedonia the teaching method is almost universally lecture and oral instruction, with students as passive participants; teaching that stimulates greater student engagement and critical thinking is rare. 7. It is important to raise the qualification and motivation of teachers at all levels of education, beginning with competitive entrance into the teaching profession and introducing quality criteria into teacher selection, professional development, and career enhancement. Teachers not performing well should be given more support. Lastly, teachers have asked for adequate training in the constant changes to curricula and school policies and have expressed frustration about not being able to modify the curricula to differentiated learning needs. Teachers should therefore be systematically included when curricula guidelines are drafted and allowed some flexibility to adjust a curriculum according to students’ needs to ensure that learning goals and practices are truly student-centered. Learning outcomes and subject standards should be integral to all curricula. 8. Broaden access to preschool. Ensuring that children are on an equal footing must start early in life. Access to quality preschooling allows them to acquire social skills and build the foundations for lifelong learning. Unfortunately, preschool coverage is very low in North Macedonia.  To strengthen accountability and quality assurance : 9. Promote transparency throughout the decentralized education system. Although the MOES establishes upper and lower bands to restrict variations in per-student annual spending, treating municipalities equitably depends on spending transparency to ensure that funds spent per- student are within the bands. It is recommended that transparent criteria be set for allocation of resources, perhaps with some priorities, by LGs and schools as well as in the allocation of capital spending and ancillary services from the CG. Clarity of priorities and access to information on local processes and decisions and mechanisms for reporting, including financial reports, performance assessments, and external audits, can promote transparency. 10. Reinforce QA mechanisms to promote university transparency and competitiveness. In North Macedonia both external and internal QA systems for higher education are not yet fully developed. Although there is an accreditation board, it has a shortage of technical capacity and full-time staff and appears to be governed by political forces and input-based evaluation processes. Significant capacity building and resources are needed to establish a culture of quality and institutionalize it in all tertiary institutions, public and private. 142 Chapter 7: Making the Grade on Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE Chapter 8: Strengthening Public Spending to Improve Health Outcomes In recent decades North Macedonia has undertaken ambitious reforms in the health sector, among them cost-control measures that helped curb spending on health. Both total health spending and public health spending have been slipping for the past decade and are now lower than in peer countries. As a result, North Macedonia’s health system is relatively efficient compared to its peers: Its outcomes are better than expected, since its spending on health is relatively low. However, while the country fares well with regard to maternal mortality, its health outcomes are not as good for child health, noncommunicable diseases (NCDs), and financial risk protection. And recent trends in child health and NCDs have been worrisome. To catch up with peers in all health outcomes will require that North Macedonia give health a higher priority in public spending and allocate public resources for health more efficiently. The most promising efficiency gains are concentrated in the areas of payments to providers of primary health care, the planning of specialist services, and pharmaceuticals. 8.1 Introduction North Macedonia’s population is aging, with implications for the costs of its health system. The country’s total fertility rate is currently below replacement level (1.5), which means the population will both age and decline in coming years. By 2050, the number of Macedonians is projected to be only 1.8 million, and 35 percent will be over 60 (Figure 8.1). The combination of an aging population and lifestyle changes has meant a change in the burden of diseases (Global Burden of Disease Collaborative Network 2017). In 2016, cerebrovascular diseases were the top cause of disability-adjusted life years (DALYs)—they were responsible for 12.3 percent of the total, followed by ischemic heart disease (11.6 percent). These demographic and epidemiologic trends have significant implications for health care demand. However, in a constrained fiscal situation (see chapter 1), there are few opportunities to raise resources to meet the demand. Thus, the Government needs to optimize the use of resources allocated to health. This chapter identifies the strengths and weaknesses of the current health system and proposes reforms to overcome the shortcomings and optimize resource utilization while minimizing disruptions in service delivery. The first section analyzes and benchmarks health outcomes in North Macedonia and peer countries. The second explores health spending and the overall efficiency of the system. The third section analyzes public health spending in terms of allocative efficiency and current cost-control reforms in the areas of provider payments and pharmaceuticals. The fourth proposes short- and medium-term reform options. Chapter 8: Strengthening Public Spending to Improve Health Outcomes 143 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 8.1. Population Age Distribution, 2015 and 2050 Percent 85+ 80–84 75–79 70–74 65–69 60–64 55–59 50–54 45–49 40–44 35–39 30–34 25–29 20–24 15–19 10–14 5–9 0–4 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 «Males Females» 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 JJ2050 JJ2012 Source: UN data and World Bank staff calculations. 8.2  Health Outcomes and Services 8.2.1  Life Expectancy Although life expectancy in North Macedonia has been rising, its progress is slow compared to peer countries. In the past 20 years, life expectancy at birth has gone up by 3.3 years—a slower rate of growth than in peer countries (Figure 8.2). In 2015, it reached 75.5 years, higher than the average of 75 for UMI countries and 74.8 for structural peers,151 but below the averages of 75.8 for the WB6 countries. 81.4 for the EU, and 76.3 for the STEE7 (Figure 8.3). Figure 8.2. Life Expectancy at Birth Compared, 1995– Figure 8.3. Life Expectancy at Birth Compared, 2015 2015 Years Years 84 82 82 80 80 78 78 76 74 76 72 70 74 68 72 66 64 70 Structural 5 7 9 1 3 5 7 9 1 3 5 199 199 199 200 200 200 200 200 201 201 201 EU 7STEE WB6 MKD UMI peers ÂÂSTEE7 ÂÂEU28 ÂÂMKD ---- SP ---- UMI ÂÂWB6 Source: WDI data and World Bank staff calculations. 151 Albania, Bosnia and Herzegovina, North Macedonia, Georgia, Jamaica, Jordan, Mauritius, Paraguay, and Serbia. 144 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE 8.2.2  Maternal and Child Health While North Macedonia fares well with regard to maternal health, its child health outcomes do not compare favorably with peer countries, and recent trends are worrisome. The legacy of the socialist era—a large and well-distributed public health care infrastructure delivering preventive care for mothers and children without co-payment—has been preserved by governments of North Macedonia over time regardless of their political orientation. This contributed to favorable outcomes for access to, and use of maternal and child health services. According to the latest estimates, 98.6 percent of pregnant women receive prenatal care and 99.9 percent of births are attended by skilled health personnel (World Bank 2017). However, some outcome indicators do not match regional averages. Although maternal mortality decreased from 13 per 100,000 live births in 1995 to 8 in 2015, on par with the EU average between 1995 and 2016, the neonatal mortality rates went down only from 12.6 per 1,000 live births to 8.3, infant mortality from 22.1 to 10.7, and under -5 morality from 24.7 to 12.2. North Macedonia’s neonatal mortality rate is now the highest among peers. Its infant and under-5 mortality rates are on par with UMI countries and structural peers but higher than in STEE7, WB6 countries, and EU countries. In addition, since 2010 began, positive trends in neonatal, infant, and child mortality have reversed (Figure 8.4).152 Moreover, in recent years, there has been a significant and worrisome drop in vaccination rates in North Macedonia and all WB6 countries. The measles vaccination rate for 1–2-year-olds fell from 96 percent in 2005 to 82 percent in 2016—in EU and UMI countries the trend was the opposite (Figure 8.5). As discussed below, among reasons for the downward trends in WB6 countries might be fewer resources for health, prioritizing specialized health care over preventive and primary care services, and facility inefficiencies. Figure 8.4. Child Mortality Rates Compared, 1995–2016 Figure 8.5. Measles Immunization Rate, 2005 and 2016 Number of deaths of children under 5 years per 1,000 live births Percent of 12–23-month-olds 45 100 40 95 35 30 90 25 85 20 15 80 10 75 5 0 70 Structural 5 7 9 1 3 5 7 9 1 3 5 199 199 199 200 200 200 200 200 201 201 201 UMI EU 7STEE peers WB6 MKD ÂÂSTEE7 ÂÂEU28 ÂÂMKD ---- SP ---- UMI ÂÂWB6 JJ2005 JJ2016 Source: WDI data and World Bank staff calculations. 8.2.3  Noncommunicable Diseases NCDs are prevalent, and their mortality rates are high. At 10.2 percent the prevalence of diabetes is high, though in line with averages for the WB6 (10.2 percent), UMI (10.1 percent), and structural peers (10.6 percent). However, it is significantly higher than the EU and STEE7 averages of 6.1 percent. Moreover, since 2010 while diabetes has become less prevalent among EU and STEE7 countries, it has 152 The data available do not allow for an analysis of variations between population groups with regard to neonatal, infant, and child mortality. Further analysis based on administrative data is needed. Chapter 8: Strengthening Public Spending to Improve Health Outcomes 145 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 become considerably more so in North Macedonia and all WB6 countries. North Macedonia's cancer and mortality rates are also higher than in peer countries, especially lung cancer among males (Figure 8.6); and in 2015 the mortality rate among 30–70-year-olds from cardiovascular disease, cancer, and chronic respiratory diseases was 22 percent—far above the EU average of 12.5 percent and also higher than the averages of 18.8 percent for UMI countries, 19 percent in the WB6, 19.1 percent for structural peers, and 19.7 percent for STEE7 countries. The data thus suggest that shortcomings in NCD prevention and treatment in North Macedonia may explain its higher rates of sickness and death, especially for those who might otherwise have been healthy and productive. Not surprisingly, risk factors associated with Figure 8.6. Lung Cancer Incidence and Mortality Compared, 2012, per 100,000 Men NCDs have worsened considerably in recent Number of cases years. Over half of North Macedonia’s adults are 120 estimated to be overweight, with 22.6 percent of men and 22.1 percent of women over 18 considered 100 obese (WHO 2017b). The number of overweight 80 and obese children is of particular concern: 60 about 39.4 percent of boys and 32.9 percent of girls are overweight, and 20.3 percent of boys 40 and 16 percent of girls are obese (Institute of 20 Public Health 2014). In 2013 an estimated 40 to 50 percent of Macedonians aged 15 or over were 0 N MKD SRB E 6 CZE EU ALB HU MN V HR WB BIH SVN SVK smokers (Institute of Public Health, 2014)—more JJIncidence JJMortality than double the 2014 averages of 18.4 percent Source: European Cancer Observatory. in the EU and 23.7 percent in STEE7 countries for 2014 (Eurostat 2017). Figure 8.7. Cervical Cancer Screening Compared, Figure 8.8. Excise Tax, 2016 Latest Data Percent of target population Percent of cigarette price 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 0 R LVA SRB D ALB SVN CZE SVK MKD SRB BGR BIH BG SVK EST SVN MNE LTU V HR MK JJTaxes, percent of price of the most sold brand (ad valorem excise) JJTaxes, percent of price of the most sold brand (specific excise) Source: Eurostat. Source: WHO. Note: Data for Slovenia: 2014; Czech Republic: 2014; Slovak Republic: 47.7; North Macedonia: 2015; Serbia: 2013; and Bulgaria: 2015. It thus appears that North Macedonia’s health services, particularly prevention services, have not yet adapted to the emerging problems. For example, is it estimated that cervical cancer screening has reached only 18.7 percent of the target population (Figure 8.7). Smoking is prohibited in public spaces, health warnings on tobacco packages are mandatory, and most direct and indirect tobacco 146 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE advertising is prohibited. Yet smoking cessation services are available only in some health facilities, there is no hotline to support smokers trying to quit, and nicotine replacement therapies are not reimbursed. Moreover, taxes on tobacco—the most efficient policy to reduce consumption, are lower than in neighboring countries (Figure 8.8 and Box 8.1).153 Box 8.1. North Macedonia's Tobacco Production and Taxation North Macedonia is the only country in the WB6 where tobacco production is intensively subsidized. In 2016, of 1,267,000 ha of agriculture area, tobacco was grown on 16,379 ha (1.3 percent). In 2016, the subsidy was 60 MKD per kilogram. Since January 1, 2010, the EU has not specified subsidies for tobacco production. After dropping from 2005 to 2008, since 2009 the volume of tobacco produced annually has held steady at about 25,000 to 30,000 tons and the area allocated at about 16,000–20,000 ha. As in EU countries where tobacco production used to be subsidized, North Macedonia's subsidies for tobacco growing could support healthier and more economically beneficial programs. Low taxation in North Macedonia helps keep cigarette prices low—in 2016, the price of a 20-cigarette pack of the most popular brand was the lowest among peer countries  (Figure B8.1.1). Excise rates have been stable since 2013, with minimal rate raises and a constant ad valorem tax rate. North Macedonia is also the only country among peers where between 2008 and 2016 cigarettes became more affordable (Table B8.1.1). Figure B8.1.1. Price of a 20-cigarette Pack, Most Table B8.1.1. Share of GDP per Capita to Buy 100 Popular Brand Compared, 2016 Packs, Most Popular Brand Compared, 2008–16 Current international $ ($ PPP) 8 2008 2010 2012 2014 2016 7 6 ALB 3.5 3.5 3.7 4.2 4.4 5 BGR 3 4.1 4.1 4 3.8 4 BIH 3.1 3.5 4.2 5.2 5.9 3 2 GRC 1.4 1.6 2.1 2.5 2.5 1 MKD 2.6 2.5 2.4 2.4 2.3 0 MK D ALB SRB BIH GRC BG R SRB 1.9 1.8 2.2 3.1 3.7 Source: WHO 2017a. (continued on next page) 153 In addition, for an overview of air pollution and its impact on health, see Annex 7. Chapter 8: Strengthening Public Spending to Improve Health Outcomes 147 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 (Box 8.1 continued) Fewer domestic cigarettes sales would have little impact on North Macedonia's tobacco production. There is no conflict between public health policies to reduce consumption and the economic interests of farmers who grow raw tobacco because (1) 90 percent is exported; (2) about 50 percent of cigarettes sold in the country are imported; and (3) less than 40 percent of the tobacco in cigarettes produced in North Macedonia is domestically grown. Therefore, policies to control tobacco consumption in North Macedonia, including higher taxes, would have little economic impact. There is compelling evidence that raising tobacco prices substantially through taxation is the single most effective way to reduce tobacco use. For example,between 1993 and 2003 higher tobacco taxes in South Africa helped to reduce consumption by about 40 percent. Similarly, alcohol taxation has been found to be cost-effective in bringing down alcohol consumption and harm, and there is growing evidence that sugar taxes are also an efficient way to promote health and nutrition. When Finland reduced taxes on alcohol in 2003, alcohol-related mortality went up by 16 percent among men and 31 percent among women. Mexico’s sugar tax reduced sugar-sweetened beverage sales by 5 percent in the first year and almost 10 percent more in the second year.1 1 Sugar, Tobacco, and Alcohol Taxes (STAX) Group. Sugar, Tobacco, and alcohol taxes to achieve the SDGs. The Lancet. Published online May 29, 2018. 8.2.4  Quality of Care Waiting times have recently been cut significantly. In 2009, a sophisticated electronic health data system, based on electronic patient records and with a module for online appointments, was introduced in three tertiary care facilities. The system has since been rolled out to 3,500 health care facilities and 15,000 users and, in 2016, 1.4 million patients were registered on it. Very quickly, this reform cut average wait times steeply for a number of specialist services—wait times for radiology scans and specialist visits decreased from about 15 months to less than one week (Milevska et al. 2017). Based on patient organization survey results, wait times for a number of services are shorter than in peer countries, among them computerized tomography (CT) scans, elective surgery, cancer treatment, and visits to emergency departments (Figures 8.9 and 8.10). However, short wait times despite the relatively limited availability of CT scans in North Macedonia suggest that utilization of health care services is limited.154 However, low birth numbers and high rates of cesarean sections (C-sections) suggest that in North Macedonia some health care is inappropriate, with consequences both for the cost of the services and for patient safety. C-section rates in particular suggest inappropriate admissions and care. C-sections are contraindicated among low-risk women and should not exceed about 15 percent of births. In North Macedonia, they now reach an estimated 28.9 percent of births (WHO 2017b). Furthermore, 154 In North Macedonia, the number of CT scans per 100,000 inhabitants was 2,560 in 2015, compared to 4,383 in Serbia (in 2014), 5,137 in Bulgaria, 6,184 in Slovenia, 7,742 in Croatia, 9,476 in Lithuania, 15,190 in Estonia, 15,620 in Slovakia, and 16,852 in Latvia (Source: Eurostat 2017). Utilization of both outpatient and inpatient services tend to confirm low utilization. While outpatient contacts per capita and per year (7 in 2013) are on par with comparators (7.5 in STEE7 countries, 7 in EU countries, and 6.1 in WB6 countries in 2014), inpatient care discharges per 100 inhabitants (11.2 in 2013) is lower than in comparator statistics (23.5 in STEE7 countries, 18 in EU countries, and 12.8 in WB6 countries). 148 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE the C-section rate varies from 0 to 51 percent of births among facilities. In several facilities, the total number of deliveries annually is less than 400, which raises further concerns about quality of care and patient safety and suggests a need to concentrate some services in specific facilities to allow for economies of scale (Figure 8.11). Figure 8.9. Wait Time for Non-acute CT Scan Figure 8.10. Wait Time in Emergency Departments Compared, 2016 Compared, 2016 3.0 3.0 2.5 2.5 2.0 2.0 1.5 1.5 1.0 1.0 0.5 0.5 0 0 HRV SVN SRB MKD CZE SVK BGR SVK SVN BGR SRB CZE MKD HRV Source: Health Consumer Powerhouse 2016. Source: Health Consumer Powerhouse 2016. Note: 1=Typically less than 7 days; 2=Typically more than 7 days and less than 21 days; Note: 1=Less than 1 hour; 3=More than 3 hours. 3=Typically more than 21 days. Figure 8.11. Deliveries and C-Section Rate by Facility, Figure 8.12. Appendectomy and Abdominal and Other 2016 Hernia Procedures by Facility, 2016 Percent Percent 6,000 100 500 100 5,000 80 400 80 4,000 60 300 60 3,000 40 200 40 2,000 1,000 20 100 20 0 0 0 0 GH Kocani GH Debar & obstetrics Skopje Skopje UC for gynecology GH Prilep SH Majka Tereza GH Gegvelija CH Tetovo CH Bitola CH Stip GH Kumanovo GH Strumica GH Struga GH Gostivar GH Veles GH Ohrid GH Kavadarci GH Kicevo Ohridsky Skopje UC for surgery St Naum GH Kocani GH Debar surgery Skopje 8 Skopje GH Prilep CH Tetovo UC for digestive CH Bitola CH Stip GH September GH Struga GH Gostivar GH Ohrid GH Kumanovo GH Veles GH Gevgelija GH Kavadarci GH Kicevo GH Strumica JJNumber of deliveries ÂÂShare of C-sections, rhs JJNumber of appendectomy, abdominal and other hernia procedures ÂÂShare of complications or age >59, rhs Source: Health Insurance Fund of North Macedonia. Source: Health Insurance Fund of North Macedonia. Notes: UC=University Center; SH=Specialist Hospital; CH=Clinical Hospitals; GH=General Notes: UC=University Center; CH=Clinical Hospitals; GH=General Hospital. Hospital. Similarly, hospital appendectomies and abdominal and other hernia procedures raise questions about the quality of care. Most general hospitals see fewer than 100 cases a year. Furthermore, the complexity of cases is somewhat similar in general, university, clinical, and specialized hospitals but should typically be lower in general hospitals. This lack of differentiation suggests admissions at inappropriate levels of care or miscoding of services delivered at the facility levels (Figure 8.12). Chapter 8: Strengthening Public Spending to Improve Health Outcomes 149 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 8.2.5  Financial Risk Protection Fewer health needs are now unmet due to the Figure 8.13. Self-Reported Unmet Needs for Medical Care due to Cost by Income Quintile, 2015 high cost of care but unmet needs are still high Percent among households at the bottom of the income 5 distribution. In 2012, 6.1 percent of the population reported an unmet need for care because it 4 was too expensive. This proportion dropped nationally to 2.1 percent in 2015, close to the EU 3 and STEE7 averages, but still remains 4.7 percent, 2 more than twice as high, for the bottom income quintile (Figure 8.13). 1 Though out-of-pocket expenditures as a share 0 Total First Second Third Fourth Fifth of total health expenditures have been reduced JJMKD JJEU JJSTEE7 in recent years, they are still exceedingly high. Source: Eurostat data and World Bank staff calculations. Though down from 43 percent in 2003, in 2014 OOP expenditures accounted for 36.7 percent of total health spending (Figure 8.14). This figure is above the EU, STEE7, and UMI averages but similar to that of the WB6 (Figure 8.15). Note, however, that North Macedonia fails to meet one of the WHO criteria for financial protection, as OOP payments as a share of total health spending exceeds the recommended limit of 15-20 percent.155 Figure 8.14. OOP and Public Health Expenditure as a Figure 8.15. OOP Health Expenditures as a Share of Share of Total Health Spending, 2002–14 Total Health Spending, 2014 Percent of THE Percent of total expenditure on health 80 40 70 35 60 30 50 25 40 20 30 15 20 10 10 5 0 0 2 4 6 8 0 2 4 200 200 200 200 201 201 201 SP MKD WB6 UMI STEE7 EU ÂÂOOP expenditure ÂÂPublic health expenditure Source: WDI data and World Bank staff calculations. The risk of impoverishing and catastrophic health expenditures is also exceedingly high for Macedonians. In 2014, more than a third (38.1 percent) were at risk of being impoverished by the need for surgical care, a much higher proportion than in EU, STEE7, WB6, and structural peer countries. In addition, more than two-thirds (68.6 percent) were at risk of incurring catastrophic expenditures for surgical care—a higher proportion that in any of the comparator groups (Figure 8.16).156 Comprehensive international comparisons of health expenditures are included in Annex 8. 155 OOP spending may be underestimated due to an increase in non-responses since 2008 and in wealthy regions, to the HBS, on which OOP estimates are based. Collecting quality data on OOP spending on health regularly would ensure that estimates are accurate. Also, data on how OOP spending is distributed among various health goods and services is necessary to analyze what is driving OOP spending on health. 156 It would be important to collect quality data on OOP spending on health regularly, including data on the distribution of OOP spending among various health goods and services, to allow for further analyses on impoverishing and catastrophic spending on health. 150 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 8.16. Financial Risks from OOP Expenditures for Surgical Care Compared, 2014 A. Risk of Impoverishment B. Risk of Catastrophic Expenditures Percent of people at risk Percent of people at risk 70 40 60 35 30 50 25 40 20 30 15 20 10 10 5 0 0 MKD UMI SP WB6 STEE7 EU UMI MKD SP WB6 STEE7 EU Source: WDI data and World Bank staff calculations. 8.3  Total Health Spending: Trends and Relative Health System Efficiency 8.3.1 Trends157 Total health spending as a share of GDP has been trending down for several years, in sharp contrast with regional and EU trends. In 2001 it totaled about 9.2 percent but by 2013 had dropped to 6.1 percent, though in 2014 it went up slightly, to 6.5 percent. While total health expenditures for 2014 were in line with the UMI average, they were significantly lower than the averages of the STEE7, WB6, and EU countries (Figure 8.17). Between 2005 and 2014 North Macedonia’s health spending per capita went up 30.3 percent; meanwhile in STEE7 countries it went up on average by 76.8 percent, in UMI countries by 71 percent, in the WB6 by 63.3 percent, and in structural peers by 62 percent. As a result, health spending per capita is now lower in North Macedonia than in any peer group (Figure 8.18).158 Spending on other health inputs confirm that the resources North Macedonia allocates to health are less than in comparator countries. The country has relatively few health professionals per capita, which contributes to its relatively low spending on health. Its 421 nurses per 100,000 inhabitants ranks far below the WB6, EU, and STEE7 averages (Figure 8.19). And though its ratio of 280 physicians per 100,000 inhabitants is higher than the WB6 average of 234, it is below the STEE7 and EU averages (Figure 8.20).159 The number of hospital beds per capita, which has gone down slightly over the past decade, is below regional averages. With 443 beds per 100,000 people in 2013, North Macedonia also ranks lower than EU and STEE7 countries (Figure 8.21). 157 This section includes analyses of inputs recorded in international databases. It would be useful to have National Health Accounts for North Macedonia or other tools with data on additional inputs, such as wages and drug consumption, to allow for further analysis of country trends in total health expenditures and further international comparisons. Such data are not yet available for North Macedonia. 158 The drop between 2003 and 2008 in total health spending was because GDP grew faster than health spending. On the other hand, the magnitude of the drop since 2008 in total health spending as a share of GDP should be interpreted with caution. Public health expenditures as a share of GDP, which is a driver of total health expenditures as a share of GDP, continued to decrease during this period (section 8.4 of this chapter). It is estimated that OOP spending on health as a share of GDP, which is the other driver of total health expenditures as a share of GDP, remained relatively constant. However, as previously mentioned, OOP spending may be underestimated (section 8.2.5 of this chapter) 159 The ratio of physicians per 100,000 inhabitants increased from 255 in 2006 to 280 in 2013. Chapter 8: Strengthening Public Spending to Improve Health Outcomes 151 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Figure 8.17. Total Health Spending as a Share of GDP Figure 8.18. Health Spending per Capita, PPP, Compared, 1995–2015 Compared, 1995–2015 Percent Constant 2011 international $, in hundreds 12 45 40 10 35 8 30 25 6 20 4 15 10 2 5 0 0 5 7 9 1 3 5 7 9 1 3 5 7 9 1 3 5 7 9 1 3 199 199 199 200 200 200 200 200 201 201 199 199 199 200 200 200 200 200 201 201 Year ÂÂSTEE7 ÂÂEU28 ÂÂMKD ---- SP ---- UMI ÂÂWB6 ÂÂSTEE7 ÂÂEU28 ÂÂMKD ---- SP ---- UMI ÂÂWB6 Source: WDI data and World Bank Staff calculations. Figure 8.19. Nurses per 100,000 Inhabitants Compared, Figure 8.20. Physicians per 100,000 Inhabitants Latest Available Data Compared, Latest Available Data Number of nurses Number of physicians 1,000 4.0 3.5 800 3.0 600 2.5 2.0 400 1.5 1.0 200 0.5 0 0 EU STEE7 WB6 MKD STEE7 EU MKD WB6 Source: WHO data and World Bank staff calculations. Note: Latest data: North Macedonia: 2013; Western Balkans: 2013; European Union: 2013; STEE7s: 2014. Figure 8.21. Hospital beds per 100,000, Compared, 1995–2015 Number of beds 1,000 800 600 400 200 0 5 7 9 1 3 5 7 9 1 3 199 199 199 200 200 200 200 200 201 201 ÂÂSTEE7 ÂÂMKD ---- SP ÂÂWB6 Source: WHO data and World Bank staff calculations. 152 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE 8.3.2  Relative Health System Efficiency Comparing the relative efficiency of national health systems across countries can be done using amenable mortality, healthy life expectancy or life expectancy (depending on the data available), and health spending per capita.160 Countries toward the bottom left in Figure 8.22 spend Figure 8.22. Healthy Life Expectancy and Health Spending per Capita, 2014–15 little on health care and have a short healthy Healthy life expectancy (HALE) at birth life expectancy, and countries toward the top 74 right have both high spending on health and a 72 long healthy life expectancy. Countries at the EU28 top left, North Macedonia among them, spend 70 little on health care but have a long healthy life 68 WB expectancy, and thus appear to have the most UMI 66 MKD SP STEE7 efficient spending. However, a close analysis 64 of public health spending in North Macedonia 62 reveals room to improve the allocation of public health resources. 60 0 2,000 4,000 6,000 8,000 Health expenditure per capita, PPP, constant 2011 international $ Source: WHO and WDI data and World Bank staff calculations. 8.4  Public Health Spending 8.4.1 Overview Over the past decade, North Macedonia’s public health spending has fallen as a share of both GDP and government spending. In fact, it is the only country among its peers where public health spending as a share of GDP has almost constantly decreased over the past 15 years (Figure 8.23). At 4.1 percent in 2014, public health spending as a share of GDP ranked well below the averages for structural peers, STEE7, WB6, and the EU. Between 2006 and 2014 spending on health as a share of general government expenditures went down from 15.5 to 12.9 percent (Figure 8.24). North Macedonia’s public health spending accounted for 63 percent of total health spending, which is lower than the STEE7 and EU averages (Figure 8.25). The Health Insurance Fund (HIF) is largely responsible for North Macedonia's public health spending. Since 2009, all Macedonians who could show proof of citizenship have been covered by Social Health Insurance (SHI), which is now almost universal.161 Whereas the HIF is responsible for purchasing health services, the MOH finances capital investment in public health facilities (infrastructure and medical equipment), and conducts preventive and public health measures through programs directly financed by the central budget. Between 2011 and 2016 the HIF accounted for 85–90 percent of total health spending. HIF resources for health care come mainly from payroll- and pension-based contributions. In 2016, 87.6 percent of the HIF’s resources came from contributions earmarked for health care. Other revenues include transfers from the MLSP for maternity leave allowances, which are managed by the HIF on 160 For a discussion of the limits of such an analysis and other approaches, see Cylus et al. 2017. 161 The HIF estimates that in 2016, 90.3 percent of the population were enrolled (Health Insurance Fund of Macedonia 2017). Chapter 8: Strengthening Public Spending to Improve Health Outcomes 153 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 behalf of the MLSP,162 surpluses carried over from previous years, and revenues from copayments by pharmacies, orthopedic appliance producers, and patients for treatment abroad, and from conventions with health insurance funds from other countries (Table 8.1). Payroll contributions represent 61.5 percent of total insurance contributions, and salaried and self-employed workers make up 50.8 percent of policyholders. Pensioners contribute 27.2 percent of total insurance contributions and represent 26.1 percent of policyholders. The Employment Agency subsidizes contributions for unemployed people, the MLSP social assistance beneficiaries, and the MoH and other groups outside these categories. These contributions represent 10.5 percent of total HIF contributions, and 21.3 percent of policyholders belong to these groups. In other words, salaried and self-employed workers, and pensioners subsidize the population categories that are also subsidized by the Employment Agency, the MLSP, and the MoH. Mirroring international best practices—and in contrast with most countries in the region—these cross- subsidies are transparent and predictable because they are set out in the Law on Health Insurance. The distribution of policyholders by category is quite stable—in 2014–16 the share of salaried and self- employed workers increased by 1.4 percent and the share of pensioners and subsidized policyholders both decreased by 0.7 percent. Figure 8.23. Public Health Spending, 2014 Figure 8.24. General Government Health Spending, 2014 Percent of GDP Percent of total general government spending 9 25 8 7 20 6 15 5 4 10 3 2 5 1 0 0 5 7 9 1 3 5 7 9 1 3 5 7 9 1 3 5 7 9 1 3 199 199 199 200 200 200 200 200 201 201 199 199 199 200 200 200 200 200 201 201 ÂÂSTEE7 ÂÂEU28 ÂÂMKD ---- SP ---- UMI ÂÂWB6 ÂÂSTEE7 ÂÂEU28 ÂÂMKD ---- SP ÂÂWB6 Source: WDI data and World Bank staff calculations. Figure 8.25. Public Health Spending, 2014 Payroll contributions are relatively low compared Percent of total health spending to regional peers. Contribution rates vary by 80 category: 7.3 percent of gross salary for active 70 employees (compared to 10.2 percent in Serbia, 12 percent in Republika Srpska, 12.8 percent in 60 50 Montenegro, 15 percent in Croatia, and 16.5 percent 40 in the Federation of Bosnia and Herzegovina), 30 7.3 percent of remuneration for independent 20 workers, 7.3 percent of 20 percent of the average 10 gross salary for farmers, 13 percent of the pension 0 for pensioners, and 5.4 percent of 50 percent EU STEE7 MKD WB6 SP UMI of the average gross salary for institutions Source: WDI data and World Bank staff calculations. cross-subsidizing vulnerable categories (Health Insurance Fund of Macedonia 2017). 162 Based on international and regional experience, a review of maternity benefits is recommended. 154 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 8.1. HIF Revenues by Category, 2016 Amount Share of Share of health Category (MKD thousand) total revenues insurance contributions Health insurance contributions 23,332,141 87.6% 100.0% Payroll contributions 14,346,674 53.9% 61.5% (salaried and self-employed workers) Contributions from the pension fund 6,335,803 23.8% 27.2% Contributions from the employment agency 84,054 0.3% 0.4% (for the unemployed) Contributions from MLSP 125,610 0.5% 0.5% (for social assistance beneficiaries) Contributions from MoH 2,440,000 9.2% 10.5% (for other subsidized populations) Transfers and donations 2,755,980 10.4% n/a Transfers from MLSP (for maternity leave) 2,203,250 8.3% n/a Surpluses carried over from the previous year 552,730 2.1% n/a Other Revenues 531,615 2.0% n/a Revenues from copayments 415,020 1.6% n/a Other revenues 116,595 0.4% n/a Total revenues 26,619,736 100.0% n/a Source: Health Insurance Fund of North Macedonia 2017. 8.4.2  Allocative Efficiency The HIF purchases health care services from public and private providers.163 Services delivered by public providers represented 60.6 percent of the HIF’s health care expenditures, compared to 35 percent for services by private providers (Table 8.2). However, in nominal terms, the former increased by 11 percent and the latter by 31 percent between 2013 and 2016. Consequently, the services delivered by private providers were the main drivers of cost increases during this period. Among the services delivered by private providers, the main cost drivers were dialysis services, outpatient drugs delivered by pharmacies, and consultations delivered by primary health care physicians. Expenditures on dialysis services grew 122-fold between 2013 and 2016 due to a shift in dialysis service delivery from public providers to private providers. These services represented 13 percent of expenditures on services delivered by private providers in 2016, up from 1 percent in 2013. Expenditures on outpatient drugs delivered by private pharmacies increased by 22 percent between 2013 and 2016. These represented 32 percent of expenditures on services delivered by private providers in 2016, down from 34 percent in 2013. Expenditures on consultations delivered by primary health care physicians increased by 14 percent from 2013 to 2016. These represented 36 percent of expenditures on services delivered by private providers in 2016, up from 41 percent in 2013. The share of primary health care in public health spending has been relatively constant although there is clearly a need to give it more attention. Primary health care is the most efficient level for preventing and treating NCDs. However, between 2012 and 2016 its share of HIF expenditures declined slightly, from 30.1 to 29.3 percent, despite the need to further develop this level of care in response 163 Health care services represented 88 percent of the total HIF spending in 2016, and 11 percent went to cash benefits (sick leave, maternity leave, and travel costs). Chapter 8: Strengthening Public Spending to Improve Health Outcomes 155 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 to the rising prevalence of NCDs and the associated risks factors. Meanwhile, the share of specialist consultative services rose from 28.6 percent in 2012 to 31.3 percent in 2016, and the share of hospital care fell from 38.1 to 35.1 percent over that period (HIF 2017). Table 8.2. HIF Spending by Category, 2016 Amount Share of HIF Share of HIF Category (in MKD thousand) total spending expenditures on health care Health care services 22,787,144 87.5% 100.0% Treatment abroad 472,231 1.8% 2.1% Orthopedic appliances 459,890 1.8% 2.0% Reimbursement 52,974 0.2% 0.2% Faculty of pharmacy 18,429 0.1% 0.1% Public health care facilities 13,812,815 53.1% 60.6% Private health care facilities 7,970,805 30.6% 35.0% Primary health care physicians 2,839,458 10.9% 12.5% Pharmacies 2,515,050 9.7% 11.0% Specialists, laboratories 587,596 2.3% 2.6% Dialysis 1,024,972 3.9% 4.5% IVF 104,490 0.4% 0.5% Heart surgery 750,424 2.9% 3.3% Eye surgery 71,546 0.3% 0.3% Balneotherapy 77,268 0.3% 0.3% Cash benefits 2,808,351 10.8% n/a Sick leave 592,949 2.3% n/a Travel costs 1,943 0.0% n/a Maternity leave 2,213,458 8.5% n/a Overhead 435,077 1.7% n/a Total Expenditures 26,030,572 100.0% n/a Source: HIF 2017. Figure 8.26. Capital Spending as a Share of Public Figure 8.27. Allocated, Adjusted, and Executed Budget Health Spending, 2012–16 for Capital, 2012–16 MKD billion Percent MKD billion Percent 2.5 10 3.0 100 2.5 2.0 8 80 2.0 1.5 6 60 1.5 1.0 4 40 1.0 0.5 2 20 0.5 0 0 0 0 2012 2013 2014 2015 2016 2011 2012 2013 2014 2015 2016 JJCapital expenditure, executed budget JJApproved budget JJAdjusted budget JJExecuted budget ÂÂCapital expenditure, executed budget, as a share of public health expenditure, rhs ÂÂExecution rate, adjusted budget, rhs Source: BOOST database. 156 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE Capital spending in the health sector has decreased in recent years. Between 2012 and 2016, the average was 6.4 percent of public health spending, but it declined consistently from 9.5 to 3.9 percent over that period (Figure 8.26). It is now below the 4–5 average for OECD countries in 2007–12. The annual approved budget allocated to capital spending declined between 2012 and 2016, as did the executed budget, although the execution rate of adjusted budgets allocated to capital spending has been relatively high, varying from 94.4 percent in 2012 to 83.1 percent in 2016 (Figure 8.27). 8.4.3  Cost-Control Reforms For the past decade, North Macedonia has directed policy to controlling health care cost escalations and allocating resources more efficiently by taking better account of services actually delivered and by incentivizing services that improve health outcomes. These reforms have been successful in containing public spending on health, but there is room for improving health system efficiency, especially in the area of public provider payments systems and pharmaceuticals. When the century began, as shares of GDP both total health spending and spending on public health were high, so the government adopted cost-control policies. Starting in 2005, primary care physicians became private providers paid on the basis of the number of patients enrolled, adjusted for risk factors and performance on indicators related to preventive health care.164 Payments for specialist services are also largely based on services delivered—diagnostic-related groups (DRGs) within a certain limit for inpatient services, and fees for outpatient services up to a certain ceiling. The HIF also covers outpatient drugs delivered by private pharmacies, up to a quarterly quota. These were important reforms that neighboring countries—Albania, Croatia, Montenegro, and Serbia—have been trying to implement for years. As mentioned earlier, these reforms successfully contained public spending on health. However, there is still room from improvement in terms of health system efficiency, including in the area of public provider payment systems and pharmaceuticals. Public Provider Payments There is a need to provide incentives for better primary care services. Primary care physicians (general practitioners [GPs], gynecologists, and dentists) receive capitation payments from the HIF. The amount per patient GPs receive varies with patient age, with higher coefficients for younger and older patients—an international best practice. These amounts represent 70 percent of the possible total; the other 30 percent is based on whether they achieve their targets for preventive health care. In 2016, performance payments for GPs were based on a set of indicators aiming to incentivize the delivery of preventive health care services such as training and early detection of diabetes, kidney diseases, cardiovascular diseases, spinal deformities, and vision and hearing impairments. However, the indicators are largely process-based, and their impact on health service utilization and outcomes is not yet clear. For example, performance payments are disbursed on completion of training rather than an assessment of knowledge, or upon the invitation of patients for screenings rather than on effective screening rates. As a consequence, primary care physicians typically receive 100 percent of the possible payment and there is little or no evidence that the system effectively influences outcomes and benefits patients. 164 In this system, provider payments are controlled and predictable (based on population estimates), by contrast with a fee-for-service payment system where the total cost, which depends on the volume of services delivered, is uncertain. Chapter 8: Strengthening Public Spending to Improve Health Outcomes 157 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 With regard to specialist services, the system Figure 8.28. Bed Occupancy Rate Compared, Latest Available Year tends to favor the least efficient facilities over the Percent more efficient. The budget for each health facility 80 is largely based on (1) the expected number of 70 cases for each DRG (based on the number actually 60 seen in the previous year) multiplied by the unit 50 cost of each DRG as defined by the HIF; and (2) the 40 number of expected outpatient consultations 30 multiplied by the unit cost of a consultation. However, in some public facilities, actual costs are 20 higher than the unit costs defined by the HIF. This 10 is mainly due to the fact that provider payment 0 EU LTU HRV LVA EST SVK SRB SVN MKD reform was not accompanied by an adaptation Source: WHO. on the supply side—consequently, some facilities Note: Latest available for Lithuania: 2014; Croatia: 2014; Latvia: 2014; Estonia: 2014; Slovak Republic: 2014; Slovenia: 2014; North Macedonia: 2013. have costly and underused services. This is illustrated by the low bed occupancy rate, which has been constant for the past 15 years (Figure 8.28).165 The HIF takes this challenge into account, and increases the budget for facilities that would operate at a deficit, should its defined unit costs be paid, while decreasing the budget for facilities that would have surpluses because their actual costs were lower than defined unit costs. As a result, the least efficient facilities are favored at the expense of the most efficient. However, the number of inefficient facilities that receive budgets beyond the cost of services actually delivered has dropped from 52 in 2014 to 27 in 2017 (for a total of 103 facilities). Figure 8.29. Acute Respiratory Admissions by Facility, 2016 Number of admissions Percent 2,500 1.2 2,000 1.0 0.8 1,500 0.6 1,000 0.4 500 0.2 0 0 and allergology Skopje Institute for lung pulmonary diseases SH for lung diseases SH for lung diseases SH for lung diseases GH Kocani GH Debar Rehabilitation Skopje GH Prilep Institute for and TB Decata Kozle and TB Yasenovo CH Bitola and TB Leshok CH Tetovo CH Stip diseases and TB Skopje Institute Otesevo GH Kumanovo GH Gostivar GH Veles GH Struga GH September 8 GH Kavadarci GH Strumica GH Ohrid GH Kicevo GH Gevgelija JJRespiratory infections and symptoms JJChronic obstructive pulmonary diseases JJBronchitis and asthma ÂÂShare of complications or advanced age, rhs Source: HIF. Notes: UC=University Center; SH=Specialist Hospital; CH=Clinical Hospitals; GH=General Hospital. Hospital admissions for acute respiratory conditions, diabetes, and hypertension, also suggest unnecessary admissions and a lack of uniform admissions procedures. Most acute respiratory conditions (respiratory infections and symptoms, bronchitis, and asthma) and chronic pulmonary diseases are not associated with complications and should not require hospital admissions—they could be treated more efficiently through outpatient services. Variations in the ratios of respiratory infections and symptoms, bronchitis, asthma, and chronic pulmonary diseases to total number of 165 Even though North Macedonia has relatively few beds, the occupancy rate is low. This suggests that the country has a poor distribution of inpatient resources rather than an excessive supply. 158 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE acute respiratory conditions also imply that admissions procedures need to be harmonized (Figure 8.29). Similarly, most hospital admissions related to diabetes and hypertension are rarely associated with complications and would be preventable with an effective primary care system for managing them (Figure 8.30). Figure 8.30. Diabetes and Hypertension Admissions by Facility, 2016 Number of admissions 500 400 300 200 100 0 GH Kocani GH Debar Skopje Skopje Skopje UC for Cardiology GH Prilep hospital of 8th CH Bitola UC for Endocrinology CH Stip CH Tetovo General municipal September in GH Gostivar GH Struga GH Kavadarci GH Kicevo GH Veles GH Kumanovo GH Strumica GH Ohrid GH Gevgelija JJHypertension with complications JJDiabetes with severe complications JJHypertension without complications JJDiabetes without severe complications Source: HIF. Notes: UC=University Center; SH=Specialist Hospital; CH=Clinical Hospitals; GH=General Hospital Hard annual ceilings have led public specialist services to accumulate arrears. Based on its budgets, the HIF contracts with public facilities and transfers payments based on the number of cases or services actually delivered, up to a limit; HIF does not pay for those that exceed the targets. Therefore, facilities that exceed the targets due to unexpected increases in demand for services, which they have to fulfill, have deficits. This, in turn, leads to a lower quality of care. For example, shortages of drugs and consumables can occur if suppliers that have not been paid stop delivering them. In 2008, the Government decided to clear most health facility arrears. However, they have again accumulated. The total amount varies by year, from MKD 138 million in 2016 (0.55 percent of HIF spending that year) to MKD 672 million in 2012 (3.11 percent). Accumulated arrears now amount to MKD 3,310 million.166 Pharmaceuticals Data on pharmaceuticals is very limited in North Macedonia. Total spending on pharmaceuticals is not monitored, because there are no national health accounts or other tools to monitor health financing. Only aggregate figures for total public spending on drugs are available. More detail is not available as data are not reported from specialist services such as hospitals that procure and dispense drugs on their own.167 The number of drugs covered by the HIF is rather restricted. There are two lists of drugs covered by the HIF: list A is prescription drugs, which are mainly delivered by private pharmacies, and list B is drugs used only in hospitals. Very few drugs have been added to the lists since 2008. Inadequate procedures 166 In 2016, 95 of 114 facilities had arrears. 167 For outpatient drugs, the HIF closely monitors financial data, prescriptions, and consumption in terms of defined daily doses. Chapter 8: Strengthening Public Spending to Improve Health Outcomes 159 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 contribute to this lack of updates of the HIF’s lists.168 As a result, in 2016, only 427 international non- proprietary name (INN) forms and strengths, corresponding to 1,210 branded medicinal products, were on list A for reimbursement.169 Another 355 INN forms and strengths, corresponding to 763 branded products, were on list B. In other words, HIF will reimburse only 1,973 branded medicinal products. For comparison, the Croatian HIF reimburses over 4,300 and the Serbian HIF about 2,930 branded medical products. Failure to introduce new drugs to the lists reduces patient access to cost-effective therapies. These restrictions have driven stakeholders to adopt ad hoc financing systems for innovative drugs. For example, the MOH directly purchases innovative drugs for a limited number of patients affected by certain conditions, among them rare diseases, diabetes, HIV-AIDS, and hemophilia. MALMED, the National Drug Agency, directly purchases innovative drugs for a limited number of patients suffering from melanoma. Some hospitals directly purchase innovative drugs through small scale programs and with their own resources, which require approval from the HIF. None of these ad hoc financing schemes are based on health technology assessments that systematically take into account such factors as whether a drug is cost-effective, the fiscal impact, or public health priorities. The MOH and HIF have used reference pricing to control public spending on drugs. The MoH sets the maximum price for each prescription drug used in North Macedonia, including those covered by the HIF, using reference prices from a basket of countries that include Bulgaria, Croatia, France, Germany, Greece, the Netherlands, Poland, Russia, Serbia, Slovenia, Turkey, and the UK. The HIF then defines the reference reimbursement price for drugs on its lists, using mainly Bulgaria, Croatia, Serbia and Slovenia as reference countries.170 As a result, between 2011 and 2013, the reference price fell for 415 generic drugs and 337 patented drugs. Comparative studies have also found that HIF reference prices are lower than in neighboring countries; a review of 310 INN forms and strengths conducted under the public finance review found that 73.5 percent of drugs had lower reference prices in North Macedonia than in Serbia. International best practices suggest that reference pricing could be streamlined. Instead of the MOH and the HIF repeating external reference pricing, the MOH could continue to define the maximum price for each INN based on a limited number of comparators for which prices can be verified. The HIF would then regularly conduct a tendering process for each INN, through which wholesalers would offer their products at a price which should be lower than the maximum price set by the MOH.171 The lowest price tendered would then become the reference price. Such reforms would likely bring reference prices down further. Decreases in the HIF reference price have been offset by constant increases in the volume of drugs covered by the HIF. The HIF reimburses private pharmacies up to a quarterly ceiling, which is revised annually. Between 2008 and 2016 the total amount reimbursed to private pharmacies went up by 99.2 percent, and the number of prescriptions went up by 116.4 percent due to decreases in reference prices (Figure 8.31). These increases involve an improved coverage of prescription drugs. However, it seems that some prescriptions could be rationalized. In particular, the number of antibiotics 168 The procedure to update HIF lists was defined in 2012. The 14 commissions of the National Drug Agency (MALMED) are supposed to keep them current, but so far only 2 have been established. 169 An INN is a unique nonbranded name that is globally recognized and is public property. A nonproprietary name is also known as a generic name. 170 The HIF also takes into account previous prices and the maximum price set by the MOH to ensure that the reference price of each drug is below the maximum and the pricein the previous year. 171 This system is particularly effective for off-patent drugs for which there are several suppliers; managed entry agreements could be considered for medicines still under patent. 160 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE prescriptions increased by 1.7 percent between Figure 8.31. Number of Prescriptions and Total Outpatient Drug Reimbursements, 2008–16 2015 and 2016, while in neighboring Serbia and Million MKD million Montenegro, antimicrobial resistance campaigns 25 3.0 helped reduce the consumption of antibiotics. 2.5 20 The cost of hospital medicines procured by health 2.0 facilities is also increasing, though more slowly. 15 1.5 Hospital budgets, which are mainly based on DRGs, 10 cover the cost of drugs, and hospital pharmacies 1.0 procure the medicines they dispense, so little 172 5 0.5 information is available about the consumption of hospital drugs. However data on hospitals from 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 0 the Treasury information system suggests that JJNumber of prescriptions ÂÂTotal Value, rhs their spending on pharmaceuticals and medical Source: HIF 2017. devices rose slowly by 5.5 percent between 2014 and 2016. It is nevertheless important to consider reforms in this area, since in 2016 hospital spending on pharmaceuticals and medical devices constituted 28 percent of total HIF spending on public health care services, and 17 percent of HIF spending health care services generally (HIF 2017). Regional and international experience demonstrates that centralizing drug procurement for off-patent drugs and introducing managed entry agreements for costly on-patent medicines can have major efficiency gains by reducing unit drug prices. Centralized procurement would also allow for better control of hospital drug quality, which has recently been put into question.173 8.5  Reform Options Given the current constrained fiscal environment, the Government’s short-term focus should be on reforms that improve efficiency, quality, and fiscal sustainability without compromising health care services. Health outcomes for Macedonians are already lower than in comparator countries, particularly with regard to child health, NCDs and financial risk protection, and demographic and epidemiologic trends will push up the needs and demand for health care. Meanwhile, there is limited space for households to pay more for health care given the already high share of OOPs in total health spending, and the fiscal environment is not conducive to immediate expansion of public resources for health. In this context, in the short term the government should give priority to reforms that do not compromise access to, and the quality of, health care. Short Term 1. Prepare a health facility master plan for specialist health care, taking the population’s needs into account, and formulate recommendations to optimize the supply. In North Macedonia, where the number of hospital beds is in line with international standards, rather than closing facilities, this is likely to lead to the conversion of curative -into long-term care beds and services. International 172 Most hospital drugs are procured by individual hospital pharmacies; procurement of only a few costly drugs is centralized. 173 It seems that hospitals have recently boughtcounterfeit drugs through “parallel imports” from countries whose authorization procedures are recognized by North Macedonia (EU countries, Russia and Turkey), but without adequate distribution practice and authenticity guarantees. These transactions are currently being prosecuted. Chapter 8: Strengthening Public Spending to Improve Health Outcomes 161 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 experience shows that putting hospital master plans in place is a medium- to long-term effort. It would be advisable to accompany such a policy with more flexible caps on annual facility budgets— particularly for the most efficient facilities—to avoid hospital deficits and arrears. In parallel, the distribution of personnel among medical specialties should be reviewed. 2. To help reduce unit prices for hospital drugs, gradually centralize procurement procedures while ensuring closer quality controls for off-patent hospital drugs, including use of centralized tendering for those drugs. In parallel, managed-entry agreements to lower drugs prices for high-cost patented medicines would be advisable. Rational prescription practices should also be promoted by establishing clinical protocols and pharmacotherapeutic guidelines and through campaigns about antimicrobial resistance. 3. Add to the lists of drugs covered by the HIF new but cost-effective therapies. This will require changes in how the HIF lists of drugs are set, with clear criteria for adding new drugs. Reference pricing could be simplified to avoid overlap in MOH and HIF functions: The MOH would focus on external reference pricing to set maximum prices for prescription drugs, based on comparison cases with transparent reference prices. This would require revising the list of countries used for reference pricing. The HIF would then conduct regular tendering processes to lower the reference prices. Short to Medium Term 4. Implement priority measures to improve the cost-effectiveness of provider payments. For primary care, for instance, the process indicators on which performance payments are based should be replaced with outcome indicators. Specifically, the government could consider disbursing performance payments based on assessments of provider knowledge and competencies, and the delivery of effective services related to maternal and child health and NCDs. The HIF information system would need to be upgraded to improve HIF capacity to verify performance. 5. Increase specific taxes to ensure mobilization of adequate resources for health. An option would be to raise taxes on products that have negative effects on health outcomes—such as tobacco— which would help raise revenues while encouraging tobacco users to quit, and therefore reduce risk factors. The government should also consider raising taxes on alcohol and sugary drinks to deter their consumption and again decrease risk factors and improve health and nutrition. 6. Make health sector data more available. Data on health outputs and outcomes allowing for in- country comparisons, such as multiple indicator cluster surveys, are necessary to adequately monitor health results, including variations among different population groups. More data on total health spending, private and public health, such as national health accounts, are crucial to fully understand cost drivers and areas of potential efficiency gains when funding is constrained. This, in turn, requires quality data on (1) household OOP payments on health, such as data per income group and data on the distribution of resources between health goods and services; and (2) public spending on specialist services, including pharmaceuticals. Finally, data on quality of care is important to identify inefficiencies in the health system, such as inappropriate and costly procedures. 162 Chapter 8: Strengthening Public Spending to Improve Health Outcomes SOWING THE SEEDS OF A SUSTAINABLE FUTURE Chapter 9: Rebalancing Government Support to Improve Agriculture To unlock its full potential, agriculture needs to significantly improve its performance. Farm productivity is low, with decreasing returns to scale and limited use of technology and too many workers, many caught in a low- productivity trap. Current agricultural support measures fail to address the problem. While government support for agriculture is far more generous in North Macedonia than in regional peers, much of it targets low-value products, so that productive resources are misallocated. This chapter proposes a rebalancing, from market support and direct payments to rural development more generally. This can lead to more efficient farming and better use of public resources to heighten productivity and sector competitiveness by modernizing farms and giving priority to high-value production. 9.1 Introduction 9.1.1  Rationale and Objectives Agriculture is an important sector in North Macedonia. It accounts for about 10 percent of GDP, 18 percent of formal employment (2016), 12 percent of total exports, and 40 percent of household consumption. But its contribution to economic growth is slipping, with other sectors generating greater value added. This situation is characteristic of structural transformation (World Bank 2015a). Despite its productive potential, agriculture in North Macedonia has significant structural weaknesses that impede its development. Its worst problems are unfavorable farming structures (small and fragmented land holdings), inefficiency and low productivity, minimal use of technology, high labor intensity, low financial with capital availability for investment (especially for smallholders), and outdated production management practices. The importance of agriculture to North Macedonia’s economy, together with the structural challenges it poses, and the country’s EU accession ambitions, have led to active public intervention in the primary sector. Between 2010 and 2015, budgetary transfers to farmers represented 1.14 percent of GDP and were on average double those of other WB6 countries and almost 60 percent higher than the EU28 average. Moreover, the fact that market support and direct payments (per output and area or animal) constitute on average 80 percent of agriculture budget outlays has raised questions about the effectiveness of this type of support and whether it can facilitate the necessary structural adjustment for agriculture. The impact of these interventions is not clear and there is little empirical analysis linking spending to results. Chapter 9: Rebalancing Government Support to Improve Agriculture 163 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 9.1.2  Methodological Approach This chapter addresses the efficiency of public expenditure from both a spending and a technical perspective. Spending efficiency is examined by economic and functional category. Technical efficiency is analyzed carried out using data envelopment analysis, which is based on estimation of a production frontier defined by the most technically efficient farms in the 2014 Farm Accounting Data Network (FADN) individual farm database. Technical efficiency scores for all farm types are derived based on the distance of each farm from this “optimal” frontier (for details see Section 9.4). The effectiveness of public spending is assessed in terms of three factors:(1) the links between farm productivity for different farm types and subsidy intensity (defined as the ratio between subsidies received and the value of agricultural output); (2) the relationship between changes in MSDP levels and changes in municipal poverty 174; and (3) the impact of agricultural and rural development support on economy-wide output, value-added, and on employment through the use of input-output (IO) analysis.  quity of public spending on agriculture is addressed by estimating the distribution of MSDP by farm E type and production specialization, as well as by farm size, both in terms of the economic value of production and area. 9.2  Agricultural Structure, Challenges, and Strategic Priorities 9.2.1  Structural Characteristics The share of agriculture in GDP (10.2 percent in 2016) and employment (17.9 percent in 2015) in North Macedonia is higher than in most WB6, STEE7, and UMI countries  (Figure 9.1). Figure 9.1. Agricultural Value Added, Percent of GDP, Its contribution to economic growth, however, 2016, and Percent of Employment, 2015 Agriculture employment share, percent 2015 has been shrinking. Between 2000 and 2016, 45 agricultural GDP fluctuated more than total 40 ALB GDP as the economy experienced a structural 35 transformation, causing a reallocation of 30 economic factors in rural areas and a slower 25 UMI adjustment in productive capacity (Figure 9.2). 20 BIH SRB MKD However, while the country’s average annual 15 economic growth between 2007 and 2016 was 10 MNE moderately high compared to regional peers, 5 KOS its growth in agricultural value-added was low 0 0 5 10 15 20 25 (Figure 9.3). Agriculture value added, percent of GDP, 2016 Source: WDI data and World Bank staff calculations. 174 With municipalities aggregated on the basis of the 33 AFSARD payment centers. 164 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 9.2. Agricultural Value–Added and GDP Growth, Figure 9.3. Agricultural Value-Added and GDP Growth, 2010–16 2007–16 Average Annual percent growth, 2000–16 Annual average percent growth, 2007–16 25 7 20 6 5 15 4 10 3 5 2 0 1 -5 0 -1 -10 -2 -15 -3 -20 -4 0 2 4 6 8 0 2 4 6 200 200 200 200 200 201 201 201 201 ALB BIH BGR HRV EST KOS LVA LTU MKD MNE SVK SVN SRB UMC ÂÂAgriculture value added ÂÂGDP JJAgriculture value added JJGDP Source: WDI. Within North Macedonia, there are major regional differences in the contribution of agriculture to the economy  (Figure 9.4). Unlike the Southwest and Northeast regions, agriculture contributes a large share of regional GDP in the Southeast, Pelagonia, and Vardar regions, which are also important contributors to national agricultural value-added. In the Skopje region, though agriculture represents a small share of the economy, its contribution to national agricultural GDP is comparatively high, which suggests a concentration of higher value addition in the primary sector. Figure 9.4. Contribution of Agriculture to the Economy, Figure 9.5. Food, Beverages, and Tobacco, 2011, by Region, 2014 Percent of Value Added in Manufacturing Compared Percent Percent 35 30 30 25 25 20 20 15 15 10 10 5 0 5 on ion ion ion ion ion ion ion egi reg reg reg reg reg reg reg ar r Eas t e s t a s t ia o g ea s t j e Var d thw the ago n Pol rth Sko p 0 Sou Sou Pel No MKD SRB BIH LTU BGR LVA ALB EST SVK SVN JJPercent of agriculture in regional GDP JJPercent of regional agriculture GDP in national agriculture GDP Source: State Statistical Office. Source: WDI. Food processing has a major role in both manufacturing and GDP but here, too, structural problems undermine its competitiveness. In 2011,175 food processing accounted for 29.5 percent of manufacturing value added (Figure 9.5) and about 4 percent of GDP—a higher share than that of comparable economies for which data are available. North Macedonia has about 8,000 registered food-processing companies, mostly SMEs, but their output is very low (€2,000 per employee) compared to the EU average (€40,000) and there is minimal, if any, vertical integration. Limited institutional and production capacity and 175 Latest year for which comparable data are available. Chapter 9: Rebalancing Government Support to Improve Agriculture 165 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 the slow pace of alignment to EU standards make it hard for the sector to compete in large markets. Subsectors like fruit and vegetables, wine, and meat are export-oriented and use modern equipment, but their production capacity is low. Most production units do not have the financial resources to modernize and meet EU hygiene and food safety standards. Greater investment in food processing is needed, especially with regard to quality control and compliance with EU food safety standards. Keeping food trade balanced will become increasingly difficult for North Macedonia. Import substitution policies, which preserve high import tariffs, and direct producer subsidies for products in which North Macedonia has no comparative advantage, such as cereals, meat, and milk, limit both the competitiveness of its agriculture and its trade performance. Despite the decline in the share of exports since 2013 (Figure 9.6), food represented about 12 percent of trade in 2015, which is relatively high compared to peer countries (Figure 9.7). Most of North Macedonia’s trade is with EU member states and WB6 countries, with exports targeting high-income consumers. To increase food exports will require structural reforms in primary production (scale of production, inputs, diversification) and modernization of the food industry. North Macedonia exports vegetables and fruit, tobacco, and beverages (primarily wine) (Figure 9.8). It is a net importer of all other food categories, and the trade deficit is largest for meat, dairy, coffee, tea, and processed goods. Figure 9.6. Food Trade, 2011–16 Figure 9.7. Food Trade Compared, 2015 Percent of merchandise trade value Percent of merchandise trade value 16 18 14 16 12 14 12 10 10 8 8 6 6 4 4 2 2 0 0 2011 2012 2013 2014 2015 2016 ALB BIH BGR HRV EST LTU MKD SVK SVN UMI JJExports JJImports JJExports JJImports Source: WDI. Figure 9.8. Agricultural Trade Balance, 2016 4 Animal and vegetable oils, fats, etc. 22 Oil-seeds and oleaginous fruits 12 Tobacco and tobacco manufactures 11 Beverages 09 Miscellaneous edible products and preparations 08 Feeding stuff for animals 07 Coffee, tea, cocoa, spices 06 Sugars, sugar preparations and honey 05 Vegetables and fruit 04 Cereals and cereal preparations 03 Fish 02 Dairy products and bird's eggs 01 Meat and meat preparations Total agri-food -100 -80 -60 -40 -20 0 20 40 60 80 100 € million Source: State Statistical Office. 166 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE Agricultural land use does not appear to be driven by output value ( Table 9.1). Agriculture occupies half of the country’s area. About a third of arable land is left fallow, and this share has been relatively stable in the last few decades. Crop production accounts for 76 percent of total farm output, and livestock represents the remaining 24 percent (Annex 9 Table A9.1). Table 9.1. Farmland Allocation and Land Productivity, 2015 Share of sown area, percent Output value per hectare, MKD thousands Cereals 51.8 37.6 Industrial crops 7.3 204.6 Fodder crops 10.7 180.1 Fruits 7.7 406.1 Vegetables 14.8 715.8 Wine 7.7 172.6 Source: State Statistical Office and Eurostat data and World Bank staff calculations. Farm structures are barriers to technological Figure 9.9. Distribution of Farms by Value in Euros investment and, together with climate factors, Compared, 2013 Percent affect agricultural productivity. Small, fragmented 100 holdings dominate North Macedonia’s agriculture: 90 58 percent of farms have less than one ha 80 and 95 percent have less than 5 ha (Table 9.2). 70 These shares are slightly higher than in Kosovo, 60 Montenegro, and Serbia, but much higher than 50 40 in the STEE7 countries. Moreover, 58 percent of 30 farms have standard output of less than €2,000, 20 which confirms the significance of subsistence 10 and part-time farming. Among regional peers, 0 MKD BGR EST HRV LVA LTU SVN SVK MNE only Montenegro and Bulgaria have a similar JJ>€49,999 JJ€25,000–49,999 JJ€15,000–24,999 JJ€8,000–14,999 situation (Figure 9.9). Furthermore, structural JJ€4,000–7,999 JJ€2,000–3,999 JJ<€2,000 Source: State Statistical Office and Eurostat data and World Bank staff calculations. problems such as lack of irrigation, poor public infrastructure, poor advisory services, low access to credit, and a poorly functioning land market also make North Macedonia’s agriculture sector far from competitive. Table 9.2. Share of Holdings by Size of Farmed Area, 2013 Percent < 2 ha 2 to 5 ha 5 to 10 ha 10 to 20 ha > 20 ha Bulgaria 72.2 10.9 4.3 2.7 9.9 Estonia 9.2 21.6 20.7 17.4 31.1 Croatia 38.6 30.6 15.7 8.0 7.1 Latvia 21.6 19.7 19.7 19.3 19.7 Lithuania 14.1 39.1 22.4 11.7 12.8 Slovenia 25.4 34.3 23.8 11.3 5.2 Slovakia 25.1 27.4 12.1 9.4 26.0 Chapter 9: Rebalancing Government Support to Improve Agriculture 167 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table 9.2. Share of Holdings by Size of Farmed Area, 2013 (continued) < 2 ha 2 to 5 ha 5 to 10 ha 10 to 20 ha > 20 ha Kosovo (2014) 63.2 23.3 8.3 3.2 2.0 Montenegro (2010) 73.4 15.6 5.6 2.4 3.0 Serbia (2012) 48.8 28.9 14.1 5.1 3.1 < 1 ha 1 to 3 ha 3 to 5 ha 5 to 8 ha > 8 ha North Macedonia 58.2 29.4 7.3 2.9 2.2 Source: State Statistical Office and Eurostat data and World Bank staff calculations. Farm labor is unproductive and underemployment is high. Small farm size and low capital utilization have made agriculture relatively labor-intensive (Figure 9.10) with persistently low labor productivity and underemployment. Youth make up a significantly higher share of farmers in North Macedonia than in STEE7 countries (Figure 9.11). Structural constraints in North Macedonia's agriculture prevent younger farmers from realizing their potential on off the farm by acquiring more skills and participating in the food value chain, and limited education makes it hard for them to adopt new technology. Figure 9.10. Annual Work Units per Agricultural Figure 9.11. Farm Household Members by Age Holding Compared, 2013 Compared, 2013 Annual Work Units Percent 2.5 100 90 2.0 80 70 1.5 60 50 1.0 40 30 0.5 20 10 0 0 SVK MKD BGR HRV EST SVN LVA LTU MKD BGR EST HRV LVA LTU SVN SVK JJUp to 25 years JJ25–34 JJ35–44 JJ45–54 JJ55–64 JJ65+ Source: State Statistical Office, Eurostat. Source: State Statistical Office and Eurostat data and World Bank staff calculations. 9.2.2  Policy Priorities and Institutions North Macedonia is aligning its agricultural and rural development policies with the EU Common Agricultural Policy (CAP). Changes in policy objectives, design, and implementation, and in legislation and institutional capacities, have been taking place since 2005, when the country was identified as a candidate EU member state. The main policy strategic goal, as defined in the National Strategy for Agriculture and Rural Development 2014–20 (NSARD), is “Increasing the international competitiveness of North Macedonia’s agricultural production and agro-food industry and securing sustainable development of rural areas, with four priority areas specified: improvement of technological and market infrastructure; strengthening integration in the agri-food sector; providing access to production factors; and improving rural infrastructure. Agricultural support structures are slowly evolving. Farm support is based on direct payments to producers, which are based on output, agricultural area, and number of livestock. Eligibility criteria are limited and payment thresholds are applied based on size. Certain types of direct payment have been 168 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE in place since the late 2000s, but others change annually. Rural development support is better aligned with the CAP and has three thematic priorities that correspond to the 2007–13 CAP. Rural development is further supported by Instrument for Pre-Accession Assistance for Rural Development (IPARD) I and II.176 These, however, represent a small share of total public spending on agriculture. Support for specific measures goes primarily to veterinary services and food safety, followed by knowledge transfer and extension services. The institutional landscape for agriculture and rural development is complex, and capacity is rather low. The dominant institution in this area is the Ministry of Agriculture, Forestry and Water Economy (MAFWE), which is responsible for design, management, and coordination of policy. Conduct of relevant programs lies with the Agency for Financial Support in Agriculture and Rural Development (AFSARD), which is responsible for direct payments, rural development measures (national and IPARD), and other agricultural activities funded by the government. Advisory services are provided by the National Extension Agency, which operates through 33 local units. The Food Safety and Veterinary Agency does as its name implies, the Phytosanitary Directorate of MAFWE deals with phytosanitary issues and problem prevention. Since 2014, Inspection Services for Agriculture and Forestry have been separate. Finally, LGs have no explicit agricultural mandates, but most municipalities employ agricultural and environmental service staffs to advise local farmers. 9.3  Public Spending on Agriculture and Rural Development 9.3.1  Total Public Spending Public spending on agriculture in North Macedonia is high. Between 2011 and 2016, agriculture spending averaged 4.7 percent of total government spending (Figures 9.12 and 9.13), and its average annual rate of growth (6.9 percent) was more than four times faster than that of total public spending (1.5 percent). In 2014 the share of public spending allocated to agriculture in North Macedonia was higher than that in Albania, Montenegro, and several STEE7s (Figure 9.13). Figure 9.12. Total Public and Agriculture Spending, Figure 9.13. Agricultural Value Added and Spending, 2011–16 2014 MKD billion in 2011 values Percent ARD spending, percent total public spending 180 6 6 160 MKD 5 5 140 120 4 4 100 BGR 3 3 80 LTU 60 2 2 ALB 40 LVA EST MNE 1 1 SVN 20 SVK 0 0 0 2011 2012 2013 2014 2015 2016 0 5 10 15 20 25 ÂÂTotal government spending ÂÂAgriculture spending Agriculture value added, percent of GDP ÂÂPercent share of agriculture spending, rhs Source: Ministry of Finance; AFSARD. Source: WDI. 176 IPARD I had a very low absorption rate, 7 percent. Chapter 9: Rebalancing Government Support to Improve Agriculture 169 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Between 2010 and 2015, total budgetary transfers to agriculture according to the OECD Producer Support Estimate (APM Database for Western Balkan Countries, 2016) reached 1.14 percent of GDP in North Macedonia. That is more than double transfers in four other WB6 countries and almost 60 percent higher than in Serbia and the EU28 average (Figure 9.14). Agricultural budget absorption has improved in recent years (Figure 9.15). For 2011–16 public spending on agriculture and rural development (on average 4.70 percent of the total, in 2011 prices) was smaller than spending on health care (15.75 percent) and education (12.70 percent), but much higher than spending on recreation and culture (2.45 percent), residential construction (1 percent), and environmental protection (0.47 percent). Figure 9.14. Total Budgetary Transfers to Agriculture Figure 9.15. Budget Planning, Execution and Compared, 2010–15 Average Absorption, 2011–16 Percent of GDP Current MKD billion Absorption rate 1.2 12 1.0 0.9 1.0 10 0.8 0.8 8 0.7 0.6 0.6 6 0.5 0.4 0.4 4 0.3 0.2 2 0.2 0.1 0 0 0 MKD SRB EU28 BIH KOS MNE ALB 2011 2012 2013 2014 2015 2016 JJBudget JJExecution ÂÂAbsorption Source: APM database; OECD (TSEb transfers). Source: MOF and AFSARD data and World Bank staff calculations. 9.3.2  Spending by Economic Category Subsidies and transfers dominate agricultural spending (2011–16) and on average account for 75 percent of it. The 9.3 percent average share of wages and salaries in the agricultural budget is lower less than 14.3 percent in the national budget (Figures 9.16 and 9.17). The average shares of capital expenditure are almost identical—10.1 and 10.7 percent—due to higher agricultural capital expenditure since 2014. Figure 9.16. Wages & Salaries and Capital Spending in Figure 9.17. Agriculture Spending by Category, 2011–16 Total Public and Agriculture Spending, 2011–16 Percent Percent 20 18 16 2.9 2.5 14 12 9.3 10 8 74.6 6 10.7 4 2 0 Government Agriculture Government Agriculture Wages & Salaries Wages & Salaries Capital Expenditure Capital Expenditure JJ2011 JJ2012 JJ2013 JJ2014 JJ2015 JJ2016 JJWages and salaries JJCapital expenditure JJSubsidies, transfers JJLoans, etc JJOperational expenses Source: MOF and; AFSARD data and World Bank staff calculations. 170 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE 9.3.3  Spending by Function By far, the most important spending category Figure 9.18. Agriculture Spending by Function, 2011–16 is market support and direct payments  (Figure MKD billion at current prices 9 9.18), which for 2011–16 averaged 81 percent of 8 total spending, with a small decrease in 2016. 7 Rural development accounted for 10.5 percent of 6 it, with significant increases in 2014 and 2016.177 5 Food safety and veterinary services accounted 4 for 5.7 percent of total spending and spending 3 on extension services was low (0.9 percent) 2 and stable. Spending on hydro-meteorological 1 services was also low and stable (1.5 percent). 0 Inspection services (recorded since 2014) 2011 JJExtension services 2012 2013 2014 2015 JJHydro-meteorological services 2016 accounted for 0.6 percent of spending. JJFood safety and veterinary services JJInspection services JJRural development JJMarket support and direct payments Source: MOF and; AFSARD data and World Bank staff calculations. 9.3.4  Spending by Category of Support Market Support and Direct Payments Between 2011 and 2016 MSDP absorption rates improved consistently (Figure 9.19),from 62 to 94 percent. Within this category, direct payments for crops dominate, with 63.3 percent of total MSDP, 31.3 percent for livestock payments and 5.4 percent for other farm support. Since 2015, crop payments have fallen by about 25 percent and livestock payments have gone up by 13 percent (Figure 9.20). Figure 9.19. MSDP: Budget Planning, Execution and Figure 9.20. MSDP: Types, 2012–16 Absorption, 2011–16 Current prices MKD million, percent Absorption rate Current prices MKD million, percent of shares 8,000 1.0 7,000 0.9 63.3 0.8 6,000 0.7 5,000 0.6 4,000 0.5 3,000 0.4 0.3 2,000 0.2 1,000 0.1 5.3 31.3 0 0 2011 2012 2013 2014 2015 2016 JJBudget JJExecution ÂÂAbsorption JJDirect payments for crop production JJDirect payments for livestock production JJOther farm support Source: MOF and; AFSARD data and World Bank staff calculations. Source: AFSARD data. Direct support tends to encourage low-value farm production. For 2031–15, the product that got the most MDSP support was tobacco at 27.7 percent, even though in 2015 it made up just 5.1 percent of agricultural output value (Figure 9.21 and Table 9.3). It was followed by field crops (7.2 percent of output 177 For 2009–2016, MSDP accounted for 75.7 percent of CAP spending in the EU-27 and 80.5 percent in the EU-15. Rural development spending accounted for 24.3 percent in the former and 19.5 percent in the latter (EC 2017). Chapter 9: Rebalancing Government Support to Improve Agriculture 171 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 value; 14.9 percent of MSDP), sheep and goats (1.8; 14.5), vineyards, grapes and wine (4.9; 12.2), cattle (4.2; 9.3), milk (11.6; 8.4), and fruits and vegetables (50; 5.2). Figure 9.21. MSDP by Sector, 2013–16 Percent of total spending 0.2 1.1 0.3 8.4 0.0 13.4 1.8 14.9 9.3 5.2 12.2 27.7 5.3 JJField crops JJVegetables JJVineyards, grapes, wine JJFruit JJTobacco JJCattle JJSheep JJGoats JJMilk JJPigs and sows JJBroilers JJBeekeeping JJOther livestock sectors Source: AFSARD. Table 9.3. Per Hectare and Per Head Direct Payments Payments (MKD million) Hectares/head MKD per hectare/head Field crops 760.24 186,405 4,079 Vegetables 270.24 44,236 6,112 Vineyards, grapes, wine 976.28 22,918 42,621 Fruit 289.21 22,925 12,613 Tobacco 1,497.24 16,128 92,843 Cattle 507.29 253,442 2,004 Sheep 796.22 733,510 1086 Goats 68.25 88,064 778 Pigs and sows 12.75 195,443 65 Source: State Statistical Office and AFSARD data, and World Bank staff calculations. Rural Development Spending Since 2014, there has been a substantial improvement in absorption of rural development funding, in contrast to 2011–13 period, when several measures were not activated and there were many delays in LG public infrastructure projects managed by municipalities faced considerable implementation delays. Spending on rural diversification and quality of life (48 percent of the total spending for 2012–16) and farm competitiveness (45 percent) dominate rural development support ( Figures 9.22 and 9.23). In recent years, however, North Macedonia invested most heavily in building farm competitiveness. Sustainable management of natural resources currently accounts for 7 percent of rural development support. Though its share of spending has gone up in recent years, it is still far below EU CAP shares spent on such issues. 172 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE Figure 9.22. Rural Development: Budget Planning, Figure 9.23 Rural Development Support by Priority Execution and Absorption, 2011–16 Axis, 2012–16 MKD million current prices Absorption rate Current prices MKD million and percent shares 2,000 1.2 1,800 45 1.0 1,600 1,400 0.8 1,200 1,000 0.6 800 0.4 7 600 400 0.2 200 0 48 0 2011 2012 2013 2014 2015 2016 JJBudget JJExecution ÂÂAbsorption JJAgricultural competitiveness JJSustainable management of natural resources JJRural diversification and quality of life Source: MOF and AFSARD data and World Bank staff calculations. Source: AFSARD. Figure 9.24. Rural Development Support by Measure, Between 2013 and 2016, rural diversification and 2013–16 quality of life measures took the largest share Percent of total spending of rural development spending, 46.9 percent, 22.13 0.66 0.48 followed by agricultural infrastructure, 22.1 percent, 0.01 1.66 2.04 and modernization, 17.6 percent ( Figure 9.24). 2.78 2.25 2.69 Including these measures in rural development programs is a good move. However, measures 17.61 that require private co-financing, such as support to young farmers, food processing and marketing, and quality upgrade and advisory services, 0.85 represent such a small share of expenditure that 46.85 they require more attention. These measures JJFarm modernization JJYoung farmers JJFood processing and marketing further policy objectives specified in NSARD 2014– JJInfrastructure JJQuality schemes JJEconomic value of forests 20; giving them more support could enhance the JJAdvisory services JJOrganic farming JJAgri-environment JJRural diversification JJLFA Payments JJRural innovation competitiveness of North Macedonia's agriculture. Source: AFSARD. 9.4  Analysis of the Efficiency, Effectiveness, and Equity of Spending 9.4.1  Efficiency of Spending Analysis of farm efficiency makes it possible to assess the relative output of agricultural holdings in terms of input use. The technical efficiency of each farm is computed with reference to its best- performing or best-practice counterpart in the country, which defines the production frontier. To be consistent with the resource-saving objectives set as part of North Macedonia’s economic restructuring, an input-oriented approach is adopted. The analysis is thus intended to demonstrate how much a farm can proportionately reduce inputs and still produce equivalent outputs using the available production technology. Chapter 9: Rebalancing Government Support to Improve Agriculture 173 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 A two-stage bootstrapped DEA approach, which makes it possible to assess links between public funding and farm technical efficiency, is applied in two ways.178 In the first stage, technical efficiency (TE) scores are computed using bootstrapped DEA, and farms that receive subsidies179 and those that do not are compared. Comparisons are also made across different farm groups and categories based on their economic size and specialization. In the second stage, the potential direct effects of funding on farm efficiency are estimated using a truncated regression model, with TE scores derived from the first stage used as the dependent variable. The discussion of main results is based on Tables 9.4–9.7. Farms in North Macedonia are characterized by significant technical inefficiency, both across and within farm categories. Efficiency scores are low for the full FADN sample—a TE coefficient of 0.45 indicates that the average farm in the sample can produce the same output using 55 percent fewer inputs, given the available production technology (Table 9.4). The TE of farms in North Macedonia is lower than that of farms in Kosovo—Kosovo’s TE was found to be 0.61.180 That there is room to improve farm TE in North Macedonia has been confirmed in a study of the efficiency of its grape-producing farms. Subsidized farms have lower TE. Farms receiving subsidies (TE = 0.44) apparently underperform those that do not (TE = 0.49), with average TE estimates lowest for subsidized livestock farms (TE = 0.38) (Table 9.4). Furthermore, farms producing tobacco, which is heavily supported through coupled subsidies, have a low average TE score (TE = 0.41), while farms which do not produce tobacco on average operate more efficiently (TE = 0.46). Our findings confirm those of several studies.181 Specialist crop farms perform better. Specialist crop farms182 on average are more efficient (TE = 0.49) than specialist livestock and mixed farms (Table 9.4), which is likely due to the significant presence of comparatively intensive production within specialist crop farms. According to farm size classification, small can be efficient. The economic size classification reflects the farm size distribution in North Macedonia. Very small farms, which have an economic size of up to €4,000, are more technically efficient (TE = 0.50) than larger farms with a size of more than €8,000 (TE = 0.44) and medium farms, with a size of €4,000–8,000 (TE = 0.42).183 This finding aligns with previous 178 DEA (non-parametric) is preferred to the stochastic frontier approach (parametric) because it can accommodate multiple outputs in defining the production function. This is particularly relevant for farms in North Macedonia, where, according to FADN 2014 data, the production of most farms consists of at least two outputs. Also, DEA requires fewer constraints in modeling, and needs no pre-specified functional form for the production function. To address the main limitation of DEA, which relates to its deterministic nature—in other words, it does not account for noise in the data (Simar and Wilson 1998, 2000)—this analysis used a bootstrapped DEA. 179 Subsidized farms are those that receive direct payments and coupled support, i.e., those for which the FADN variable SE605 is positive. This does not include rural development payments. 180 TE scores for farms in Kosovo (0.61), found by Sauer et al. (2015), should be compared to their equivalent technical efficiency coefficients (that is, not bias corrected) for farms in North Macedonia, which are 0.57. In our analysis we use the more robust, bias-corrected TE. Results derived from bootstrapped DEA differ considerably from those produced by standard DEA in terms of average TE values, and also distribution patterns. This indicates the significant bias inherent to conventional DEA measures. 181 Zhu and Lansink (2010) found a negative correlation between coupled subsidies and TE in Germany, and a negative effect of subsidy intensity on agricultural TFP in Germany, the Netherlands, and Sweden. Mary (2013) found that market support payments have a negative effect on TE. Finally, in an investigation of crop farms in six OECD countries, Bokusheva and Cechura (2017) concluded that farm support payments negatively influence crop efficiency, while coupled support appears to be more distorting than other forms of support. 182 The FADN classification of farms is used; called the TF8 system, it includes farms that are (1) specialized for field crops, (2) specialized for horticulture, (3) specialized for permanent crops, (4) specialized for grazing livestock, (5) specialized for granivores, (6) mixed crop production, (7) mixed livestock production, and (8) mixed crop production and livestock. For example, each farm is attached to category 1 if two-thirds of its output value is field crops. Due to the small sample we grouped farms as (a) specialist crops: categories 1,2,3,6; (b) specialist livestock: 4,5,7; and (c) mixed crops-livestock: 8. 183 Two farm economic size classifications are considered here. The first is consistent with the farm size distribution in North Macedonia's agriculture (Figure 9), where 58 percent of farms are less than -2,000 and only 1.3 percent are over €25,000. Hence, it is considered that the classification applied here accommodates comparative analysis reflecting country conditions, which is a rather common practice (indicatively see Annex 10; EC 2011). The second classification uses the Eurostat thresholds (€8,000 as a cut-off for small farms and €25,000 as a cut-off for medium farms. 174 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE studies184 and with other findings on North Macedonia.185 The main argument for small farmers being comparatively more efficient is that when conditions are stable, small farmers gravitate toward an optimal resource allocation solution through trial and error. This does not apply to medium and larger, more commercial, farmers—not because they are less efficient at allocation but because unlike small farmers, they must respond to a more rapidly changing technological and market environment, to which their allocation decisions must respond quickly. Table 9.4. Technical Efficiency for Different Farm Categories Farm groups Number of farms (%) TEs TEb 1 All farms 454 0.57 0.45 2 Total subsidies Receiving subsidies 328 (72%) 0.55 0.44 Receiving no subsidies 126 (28%) 0.62 0.49 3 Crop subsidies Receiving subsidies on crops 284 (63%) 0.55 0.44 Receiving no subsidies on crops 170 (37%) 0.60 0.47 4 Livestock subsidies Receiving subsidies on livestock 175 (39%) 0.49 0.38 Receiving no subsidies on livestock 279 (61%) 0.62 0.50 5 Type of agricultural holding Specialist Crops 268 (59%) 0.61 0.49 Specialist Livestock 123 (27%) 0.51 0.40 Mixed Crops – Livestock 63 (14%) 0.51 0.40 6 Farm size Small (less than €4000) 134 (30%) 0.62 0.50 Medium (between €4000 and €8000) 118 (26%) 0.52 0.42 Large (over €8000) 202 (44%) 0.57 0.44 Small (less than €8000) 252 (55%) 0.57 0.46 Medium (between €8000 and €25000) 149 (33%) 0.53 0.42 Large (over €25000) 53 (12%) 0.66 0.50 7 Net investment on fixed assets Non-negative 103 (23%) 0.71 0.55 Negative 351 (77%) 0.53 0.42 8 Industrial crops (mainly tobacco) Yes 59 (13%) 0.50 0.41 No 395 (87%) 0.58 0.46 Source: FADN 2014 data and World Bank staff calculations. Notes: TEs refers to the average value of farms’ technical efficiency scores estimated using standard DEA; TEb refers to the average value of farm bias-corrected TE scores estimated using bootstrapped DEA. All efficiency scores were derived with respect to a common production frontier so that they are comparable for different farm groups. 184 The literature was initiated by the seminal work of Lau and Yotopoulos (1971). Furthermore, Masterson (2007) also found that in Paraguay small farms are more technically efficient than large. Similarly, Sini (2009) found that small farms are much more efficient than medium ones, and argues that differentiated technology and an increased number of heterogeneous management strategies allow different-sized farms to coexist. Just as differentiated management strategies are used, public interventions must be differentiated to improve farm efficiency. This finding is consistent with the interpretation provided in Mellor’s seminal paper (1969), which argues that small farmers have substantial allocative control over land, labor, and capital resources, which they use to allocate their resources under a careful chain of logic. 185 See Manevska-Tasevska (2012). Chapter 9: Rebalancing Government Support to Improve Agriculture 175 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 However, very large farms are the most efficient. Using the EU thresholds for farm size, very large farms (over €25,000) are most efficient (TE = 0.50) and they outperform small farms (up to € 8,000 Euro; TE = 0.46) and especially medium-sized farms (€8,000-25,000; TE = 0.42). The medium size farms are least efficient across the board. TE scores of non-subsidized farms converge on average for all three farm size classes regardless of the classification scheme used (Table 9.5). TE scores for subsidized small farms (up to €4,000) are comparable to those of non-subsidized small farms (TE = 0.50), but the TE of very large farms (over € 25,000) is higher for subsidized (TE = 0.52) than for non-subsidized farms (TE = 0.46) (Table 9.6). In all other size categories regardless of classification, non-subsidized TE scores are higher than subsidized ones. It appears that in general, subsidized medium-sized farms have low TE scores. Unlike farm subsidies, investments in fixed assets are associated with higher efficiency. Farms with non-negative net investment on fixed assets186 score much higher (TE = 0.55) than both those with negative net investment (TE = 0.42) and the full sample (TE = 0.45). Furthermore, TE scores for farms with non-negative net investment are almost identical for farms that also receive subsidies and those that do not,187 These findings are supported by other studies on North Macedonia’s agriculture.188 TE results offer more insights when different categorizations are considered. For example, the TE of non-subsidized farms does not converge across specialization types: Specialist crop farms (TE = 0.52) considerably outperform specialist livestock (TE = 0.39) and mixed farms (TE = 0.34). This pattern is also evident for subsidized farms, but to a lesser degree (Table 9.6). A deeper investigation of the two best- performing farm types, small farms under €4,000 and specialist crop farms, produces other interesting findings. Small farms score high (TE = 0.55) when they specialize in crops and very low (TE = 0.24) when they specialize in livestock (Table 9.5). Furthermore, for farms with non-negative net investments on fixed assets (which have an average TE score of 0.55), both small and large farms outperform medium ones. Finally, specialist crop farmers who invest in fixed assets (TE = 0.57, the highest average score of the groups) outperform specialist livestock and mixed farmers (TE = 0.51 for both). Table 9.5. Within-Group Analysis for Farm Groups with the Highest Average Efficiency Farm groups Number of farms (%) TEs TEb A. Non-subsidized farms 126 0.62 0.49 Size Small (less than €4000) 46 (37%) 0.63 0.50 Medium (between €4000 and €8000) 39 (31%) 0.62 0.49 Large (over €8000) 41 (33%) 0.61 0.47 Small (less than €8000) 85 (68%) 0.62 0.50 Medium (between €8000 and €25000) 28 (22%) 0.61 0.47 Large (over €25000) 13 (10%) 0.61 0.46 Type Specialist Crops 99 (79%) 0.66 0.52 Specialist Livestock 17 (13%) 0.51 0.39 Mixed Crops – Livestock 10 (8%) 0.44 0.34 186 Net investment in fixed assets was used here as a proxy for farm investment because the current FADN database for North Macedonia does not contain information about on-farm rural development support. 187 TE scores are 0.57 for non-subsidized and 0.54 and subsidized farms. 188 Not surprisingly, Petrovska et al. (2013) find that the managerial competence positively influences farm efficiency. In a similar vein, Manevska-Tasevska (2012), find that more organized farmers with profit maximization objectives, better organization, and interest in investment, are technically more efficient. 176 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 9.5. Within-Group Analysis for Farm Groups with the Highest Average Efficiency (continued) Farm groups Number of farms (%) TEs TEb B. Small (less than €4000) 134 0.62 0.50 Type Specialist Crops 101 (75%) 0.67 0.55 Specialist Livestock 14 (10%) 0.31 0.24 Mixed Crops – Livestock 19 (14%) 0.55 0.43 Subsidized Yes 88 (66%) 0.61 0.50 No 46 (34%) 0.63 0.50 C. Specialist Crops 268 0.61 0.49 Size Small (less than €4000) 101 (38%) 0.67 0.55 Medium (between €4000 and €8000) 70 (26%) 0.57 0.46 Large (over €8000) 97 (36%) 0.57 0.45 Small (less than €8000) 171 (64%) 0.63 0.52 Medium (between €8000 and €25000) 71 (26%) 0.55 0.44 Large (over €25000) 26 (10%) 0.62 0.47 Subsidized Yes 169 (63%) 0.58 0.47 No 99 (37%) 0.66 0.52 D. Non-negative net investment 103 0.71 0.55 Size Small (less than €4000) 33 (32%) 0.73 0.59 Medium (between €4000 and €8000) 27 (26%) 0.67 0.52 Large (over €8000) 43 (42%) 0.70 0.54 Small (less than €8000) 60 (58%) 0.71 0.56 Medium (between €8000 and €25000) 32 (31%) 0.69 0.53 Large (over €25000) 11 (11%) 0.74 0.55 Type Specialist Crops 63 (61%) 0.72 0.57 Specialist Livestock 25 (24%) 0.68 0.51 Mixed Crops – Livestock 15 (15%) 0.68 0.51 Subsidized Yes 60 (58%) 0.70 0.54 No 43 (42%) 0.71 0.56 Source: 2014 FADN data and World Bank staff estimates. Notes: TEs refers to the average value of farm technical efficiency scores estimated using standard DEA; TEb refers to the average value of bias-corrected TE scores estimated using bootstrapped DEA; because all TE scores have been derived with respect to a common production frontier, they are comparable for different groups. Table 9.6. Technical Efficiency of Subsidized Farms, by Farm Size and Type Farm groups Number of farms (%) TEs TEb Subsidized farms 328 0.55 0.44 Size Small (less than €4000) 88 (27%) 0.61 0.50 Medium (between €4000 and €8000) 79 (24%) 0.47 0.38 Large (over €8000) 161 (49%) 0.56 0.43 Small (less than €8000) 167 (51%) 0.54 0.44 Medium (between €8000 and €25000) 121 (37%) 0.51 0.40 Large (over €25000) 40 (12%) 0.68 0.52 Type Specialist Crops 169 (52%) 0.58 0.47 Specialist Livestock 106 (32%) 0.51 0.40 Mixed Crops – Livestock 53 (16%) 0.52 0.41 Source: FADN 2014 data and World Bank staff estimates. Notes: TEs refers to the average value of farm technical efficiency scores estimated using standard DEA; TEb refers to the average value of farm bias-corrected TE scores estimated using bootstrapped DEA; because all TE scores were derived with respect to a common production frontier, they are comparable for different farm groups. Chapter 9: Rebalancing Government Support to Improve Agriculture 177 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Agricultural subsidies are found to reduce technical efficiency. Second-stage regression results show (Table 9.7) that, controlling for farm economic size, subsidies negatively impact TE scores (–0.056). The negative impacts are slightly lower for farms specializing in crops (–0.046) and not statistically significant for livestock farms. This indicates that the current subsidies are linked to technical inefficiencies in production for all farms examined farms, especially crop farms. TE improves with investment in agricultural assets. Truncated regression estimates confirm the findings on the positive influence of farm investments on TE. In the full sample, the estimates show a strong positive impact of net investments on TE scores (0.142). These effects are lower for crop farms (0.106), and higher for livestock farms (0.175). This confirms that farms that invest in fixed assets perform better, whatever their specialization, and suggests that investing is a good way to improve the efficiency of all farm types. The impact of size on TE depends on how size is defined. Estimates of the effects of size dummies on TE confirm the first-stage efficiency results showing that very small farms (no more than €4,000) are more efficient than both medium and large farms in the full sample and also than specialist crops farms. However, if the Eurostat classification scheme is applied, small farms (up to €8,000) outscore medium ones (€8,000–25,000) in the total sample and specialist farms, but underperform very large farms (€25,000) in specialist livestock production. Table 9.7. Effects of Subsidies and Net Investment on Farm Technical Efficiency Total sample Specialist crops Specialist livestock Total subsidies –0.056 –0.046 –0.038a Net investment 0.142 0.106 0.175 Source: FADN 2014 data and World Bank staff estimates. Notes: Figures represent estimated coefficients of total subsidies and net investment variables in truncated regression models. a Not statistically significant. Farm TE scores differ considerably by region. Both regionally and nationally, farms with higher TE scores are located in regions where agriculture is important (Table 9.8). Farms in Vardar and Skopje considerably outscore the national average, but TE scores in the Southeast and East are only slightly above average. Except for Skopje, in regions where farms are more efficient, agriculture is a large share of regional GDP and, except for the East, the same holds for the contribution of regional agriculture to national agricultural GDP (Figure 9.4). Table 9.8. Technical Efficiency of Farms, by Region Region Number of farms (%) TEs TEb East 104 (23%) 0.60 0.47 Northeast 42 (9%) 0.52 0.41 Pelagonia 77 (17%) 0.51 0.40 Polog 36 (8%) 0.53 0.43 Skopje 39 (9%) 0.64 0.51 Southeast 74 (16%) 0.59 0.48 Southwest 40 (9%) 0.46 0.37 Vardar 42 (9%) 0.68 0.54 Source: FADN 2014 data and World Bank staff estimates. Notes: TEs refers to the average value of farm technical efficiency scores estimated using standard DEA; TEb refers to the average value of bias-corrected TE scores estimated using bootstrapped DEA; because all efficiency scores were derived with respect to a common production frontier, they are comparable for different farm groups. 178 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE 9.4.2  Spending Effectiveness This assessment of the effectiveness of spending is based on the 2014 FADN database. The intent is to estimate the output elasticities of production factors and to explore possible links between elasticity of capital and subsidy intensity for different farm categories so as to address the role of subsidies in farm modernization. Results of are presented in Tables 9.9–9.11 and Figure 9.25.189 Two additional methods are applied to the assessment of policy effectiveness Data on subsidies and social financial assistance (SFA) paymentsfor 2013–16 for the 33 AFSARD regional payment centers (each one of which corresponds to one or more municipality) are used to identify the possible relationship between spending and poverty (proxied by SFA). Finally, the study assessed the impacts of support on economy-wide output, value-added, and employment in the context of IO analysis. Productivity Analysis Farms in North Macedonia are labor-intensive and operate with decreasing returns to scale. Results on partial output elasticities nationally show that the output elasticity of labor (0.36 for all farms) is about triple that of capital (0.12) for the whole 2014 FADN sample and several farm sub-groups (Table 9.9). A capital output elasticity of 0.12 indicates that, keeping all else constant, a 10 percent increase in capital usage by the average farm in the sample would raise output by 1.2 percent. Furthermore, the estimated partial output elasticities sum to less than unity (α+β+γ<1) in all cases except large livestock farms, whose coefficients sum to close to unity. This result implies that farms of all sizes and categories appear to operate under decreasing returns to scale, except for large livestock farms, which use modern technology and whose returns to scale are constant. Table 9.9. Output Elasticities of Factors of Production and Subsidy Intensity by Subsidy Status and Type Output elasticities Farm groups/categories Land Labor Capital Subsidy intensity (%) Total sample 0.28 0.36 0.12 11.11 Subsidized 0.29 0.34 0.12 14.77 Non-subsidized 0.26 0.41 0.13 0.00 Specialist crops 0.23 0.39 0.11 6.92 Specialist livestock 0.39 0.32 0.23 13.92 Source: FADN 2014 data and World Bank staff estimates. Notes: Figures in columns 2–4 are the estimated output elasticities of land, labor. and capital. Subsidy intensity in column 5 is defined as the value of subsidies divided by the value of total output for each farm group. a Not statistically significant. Subsidized farms do not use capital more productively than non-subsidized farms. The output elasticity of capital for non-subsidized farms appears to be slightly higher (0.13) than for subsidized farms (0.12) (Table 9.9), but the difference is not statistically significant (Table 9.11).190 There is no evidence that subsidies lead to productivity gains. 189 Possible simultaneity problems due to correlation of inputs with unobserved productivity shocks may bias the estimated coefficients. To address these problems, several methods have been proposed, commonly based on the use of appropriate instruments (for example Aguirregabiria 2009). However, because data limitations do not allow us to properly address the possible simultaneity bias, the results should be interpreted with caution. 190 Table 9.11 presents results of hypothesis tests for coefficient equality across regressions focusing on the coefficients of primary interest, i.e., the output elasticities of capital. Chapter 9: Rebalancing Government Support to Improve Agriculture 179 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Neither farm size nor specialization plays a role in the contribution of capital to productivity. The difference in capital output elasticity among the three size groups are not found to be statistically significant (Tables 9.10 and 9.11). Estimations and significance tests specific to farm specialization types indicate that the output elasticity of capital used by specialist livestock farms (0.23) is not statistically different from that of specialist crops farms (0.11) (Table 9.11). With regard to specialist crops, capital seems to be of comparatively higher importance for small farms (0.19) but does not seem to have much of a role for medium and large farms (the coefficient is low and statistically insignificant). On the other hand, among livestock farms, larger farms appear to invest in capital and use their assets more productively (capital elasticity=0.29), than smaller farms, whose production is found to be independent of capital usage (the coefficient is low and statistically insignificant).191 Table 9.10. Output Elasticities of Factors of Production and Subsidy Intensity by Alternative Size Classifications Farm size classification using €4000 and €8000 Farm size classification using €8000 and €25000 as thresholds as thresholds Subsidy Subsidy Land Labor Capital Land Labor Capital intensity Farm groups/categories No of obs. elasticity elasticity elasticity intensity No of obs. elasticity elasticity elasticity (%) (%) Small 134 0.18 0.17 0.12 9.40 252 0.20 0.21 0.10 9.15 Medium 118 0.09 0.06a 0.08 9.00 149 0.21 0.18 0.13 11.20 Large 202 0.21 0.38 0.15 11.81 53 -0.01 a 0.95 0.12 a 12.47 Specialist crops / 101 0.13 0.27 0.19 7.03 171 0.2 0.30 0.15 6.33 small Specialist crops / 167 0.15 0.30 0.06 a 6.90 97 0.12 a 0.40 0.02 a 7.25 medium & large Specialist livestock / 38 0.21a 0.16 0.08 a 11.53 99 0.32 0.17 0.19 13.46 small & medium Specialist livestock / 85 0.33 0.32 0.29 14.20 24 0.30 0.67 0.16 a 14.38 large Source: FADN 2014 data and World Bank staff estimates. Notes: Figures in columns 3–5 and 8–10 are estimated output elasticities of land, labor and capital. Subsidy intensity in columns 6 and 11 is defined as the value of subsidies divided by the value of total output for each farm group. a Not statistically significant. Table 9.11. Tests for the Equality of Capital Output Elasticities Across Farm Groups Hypothesis X2 statistic Probability >X2 Decision β_subsidized= β_nonsubsidized 0.03 0.87 Not rejected β_smallMK= β_mediumMK=β_largeMK 1.35 0.51 Not rejected β_smallEuro= β_mediumEuro=β_largeEuro 0.20 0.91 Not rejected β_crops= β_livestock 2.05 0.15 Not rejected Note: β denotes the output elasticity with respect to capital. Subsidy intensity in North Macedonia can be reduced. Subsidy intensity (defined as the ratio of subsidies received to the value of agricultural output) for the total sample slightly exceeds 11 percent, with variations among farm groups (the last column in Table 9). Livestock farms (Table 9.9), especially large ones (Table 9.10), have high subsidy intensity (14 percent), but farms in the sample specializing in crop production seem to be less intensively subsidized, with little variation by farm size (7 percent). 191 Estimates for specialist livestock farms are different if the Eurostat size classification is applied. In that case, capital is of comparatively higher importance for the production of small and medium farms than for very large farms, in which case there is a lower and statistically insignificant coefficient. 180 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE The relationship between subsidies and farm Figure 9.25. Difference Between Capital Output Elasticity and Subsidy intensity modernization is complex. Viewing capital output 0.16 elasticities and subsidy intensities together, there 0.14 0.12 seems to be no single pattern characterizing 0.10 0.08 farm categories (Figure 9.25). For livestock farms, 0.06 especially large ones, high subsidy intensity 0.04 0.02 (14 percent) is accompanied by high capital output -0.02 0 elasticity (23 percent for livestock farms generally -0.04 -0.06 s m arge rops all large dium large and 29 percent for the largest ones). On the other d d ll k farm sidize sidize Sma Mediu L toc All b b a l i st c lives ps/sm m & & me tock/ u u ci is t ro diu ll es S n-s hand, small crop farms with a relatively high No Spe ecial alist c s/me /sma ist liv Sp peci crop to ck c ial s e capital share (19 percent) are among the groups S list Sp live cia list Spe pecia with the lowest subsidy intensity (7 percent). Source: World Bank staff estimations. S Smaller crop farms are not heavily subsidized, and capital is important for their production activities. For large, heavily subsidized livestock farms, coupled subsidies seem to provide incentives to accumulate capital in terms of buildings, equipment, and animals, but fail to raise efficiency. This result also implies that coupled subsidies shift resources to non-competitive activities and thus reduce the competitiveness of agriculture in North Macedonia. A more active rural development policy, and less distortive coupled subsidies could lead to more efficient resource allocation decisions farmers and thus to higher TE. Regional patterns are also difficult to draw. Results on the output elasticity of capital are not statistically significant for most regions (Table 9.12). Capital output elasticity and subsidy intensity are much higher in the East region than the national average, which could imply that heavy subsidization may have induced capital investment. However (see Table 9.8), the TE gains of farms in the East are negligible compared to the national average. Table 9.12. Output Elasticities of Factors of Production and Subsidy Intensity, by Region Region Number of observations Land elasticity Labor elasticity Capital elasticity Subsidy intensity (%) East 104 0.20 0.30 0.35 19 Northeast 42 0.21a 0.27 0.07 a 11 Pelagonia 77 0.67 -0.02 a 0.07 a 10 Polog 36 0.56 0.31 0.16 a 14 Skopje 39 0.25 0.40 -0.01a 4 Southeast 74 0.17 0.77 0.00 a 4 Southwest 40 0.34 -0.01a 0.22 11 Vardar 42 0.13 a 0.77 0.10 a 6 Source: FADN 2014 data and World Bank staff estimates. Notes: Figures in columns 3–5 represent estimated output elasticities of land, labor, and capital. Subsidy intensity in column 6 is defined as the value of subsidies divided by the value of total output for each region. a Not statistically significant. Chapter 9: Rebalancing Government Support to Improve Agriculture 181 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 MSDP and SFA payments Agricultural subsidies have very little impact on poverty reduction. Controlling for regional effects,192 estimations based on an OLS regression model with robust standard errors show that MSDP has a negligible negative but statistically significant effect (–0.005) on SFA payments193 (Table 9.13).194 A 100 percent increase in MSDP leads to a 0.5 percent reduction in poverty as proxied by SFA payments.195 Furthermore, given the low impact on poverty reduction achieved by these policy funds that make up more than 1 percent of GDP, agricultural subsidies do not seem to be an effective way to pursue social policy objectives. Table 9.13. Relationship between MSDP and SFA payments: OLS Regression Results Dependent variable: SFA Coefficient Robust standard errors P-value MSDP -0.005** 0.003 0.050 Constant 6.652*** 8.631 0.003 R2 0.80 F-statistic 164.51 No of observations 132 Source: SSO and AFSARD data and World Bank staff calculations. Notes: **statistically significant at the 5% level; ***statistically significant at the 1% level; Dummies corresponding to regional centers are included. Input-output Analysis Rural development projects have more impact on the economy than subsidies. They reinforce structural links within an economy and thus induce generation of shock-specific higher impacts.196 IO analysis has shown that the economy-wide effects of rural development projects are higher than those of direct payments when productive capacity-adjustments and structural impacts are considered.197 Initially, direct support has more impact on the national economy than rural development measures—not surprising considering how much direct payments are (Table 9.15). The annual average MSDP impacts are substantial in terms of value added (+2.91 percent) and employment (+2.70 percent), though they have less impact on output (+2.26 percent). Rural development measures increase economy-wide value added and employment by 0.32 percent and output by 0.27 percent annually. Together, the two types of public spending on agriculture raise economy-wide output by 2.53 percent, value added by 192 A set of dummy variables corresponding to AFSARD regional payment centers has been constructed for the model. The estimated coefficients are omitted from Table 9.13 for presentation reasons. 193 North Macedonia's SFA goes to persons capable of working and households with a monthly income lower than the SFA monthly amount (currently about €46) that do not have means for existence. SFA targets very poor persons and households and given a lack of data on local poverty indicators, it is here considered as a proxy for poverty. 194 Alternative model specifications commonly used in panel data analysis have been also tested, despite the small number of observations. Estimating a fixed- effects model yields an MSDP coefficient of –0.001, which is being statistically significant at the 10% level, thus implying that spending has an even smaller negative effect on poverty. On the other hand, a random-effects formulation does not lead to statistically significant results (not statistically significant Wald test and MSDP coefficient). 195 This finding, however, must be treated with caution. Further investigation is needed, given the small number of observations and data limitations, which do not permit the use of alternative poverty metrics and the consideration of additional factors that may affect poverty levels. Richer LG data for a longer period, including alternative poverty indicators and detailed data on spending, would probably lead to safer inferences. The use of SFA payments as the sole poverty indicator is problematic, since this type of social support is granted to all households with very low incomes, whatever their sector of employment. Alternative poverty indicators were only available at the national and rural-urban level. 196 See, for example, Psaltopoulos et al. 2004, 2006 and Espinosa et al. 2014. 197 This assessment of public spending on agriculture extends work carried out as part of the World Bank (2017) regional study on Agriculture, Jobs and Growth in the Western Balkans. The analysis began by using 2013 public IO supply and use matrices for North Macedonia (State Statistical Office 2016), to construct a national industry-by-industry symmetric table in basic prices. Using FADN data, agriculture was disaggregated into eight subsectors reflecting the FADN TF-8 classification. AFSARD data on agricultural and rural development payments per measure for 2011–16 were converted in model baseline prices and classified by subsector (according to their output structure and final demand for investment activity (see Table 9.14). Impact analysis shocks were performed via the traditional Leontief procedure; the size of each shock corresponded to the annual average value of 2011–16 spending per measure (World Bank 2017). 182 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE 3.23 percent, and employment by 3.02 percent annually. For 2010–15, spending on agriculture averaged 1.14 percent of national value added, which implied that at least in terms of this economic variable, agricultural support is specific to a multiplier of 2.83.198 When the impacts per MKD million spent are Table 9.14. Public Spending on Agriculture, Average Annual Values, 2011–16 calculated, the inference of effectiveness is MKD million 2013 prices rather different, and the unit impacts are very Policy measure Economic flows close. MKD 1 million spent on MSDP generates Direct payments 5,430.42 extra output of MKD 5.2 million MKD, and 1 million Rural development 659.05 MKD spent on rural development generates Farm investment 160.13 5.10 million MKD. In terms of value added, the Infrastructure 145.55 unit spending impacts are 2.34 and 2.12 million Food processing 15.39 MKD. For job creation, the impacts are close, at 3.70 and 3.57 jobs. This finding contradicts the Rural diversification 295.64 effectiveness estimates because it does not Agri-environment/0rganic farming 42.34 Source: AFSARD data and World Bank staff calculations. include productive capacity-adjustment effects (i.e., economic impacts when projects launched with policy support are operational). For rural development projects these effects are much larger than for direct payments. Table 9.15. Policy Impacts, 2011–16: Average Annual Impacts Compared to 2013 MKD million to New Jobs Impacts MSDP Rural Development Change in output 28220.22 3361.69 % change in output 2.26 0.27 Change in value added 12693.56 1399.45 % change in value added 2.91 0.32 Change in employment 20101 2351 % change in employment 2.70 0.32 Change in output per million MKD spent 5.20 5.10 Change in value added per million MKD spent 2.34 2.12 Change in employment per million MKD spent 3.70 3.57 Source: World Bank staff estimates. 9.4.3  Spending Equity 199 The distribution of direct payments by farm type favors the grazing livestock type,200which accounts for only 25 percent of the sample but receives 55.2 percent of all subsidies (Table 9.16). The pattern is reversed for farms specializing in low-value field crops (23.2 percent of farms; 11.5 percent of subsidies); 198 For every 1,000 MKD provided, economy-wide links raise value-added by 2,830 MKD. 199 Due to lack of data on the population of North Macedonia's farms, 2014 FADN individual farm data used to assess the equity of public spending on agriculture by estimating the distribution of MSDP by farm type, production specialization, and farm economic and area size. 200 When interpreting these results, the on-farm pluriactive orientation of North Macedonia's farms must be kept in mind. Chapter 9: Rebalancing Government Support to Improve Agriculture 183 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 high-value horticulture (5.7 percent; 0.4 percent); and permanent crops (13 percent; 5.8 percent). Thus, subsidy intensity differs considerably by farm type, and is comparatively higher for grazing livestock (14.69), mixed crops and livestock (14.59) and mixed crops (12.05); much lower for field crops (6.40) and permanent crops (7.80); and extremely low for horticulture (0.72). Table 9.16. Distribution of Direct Payments by Farm Type, 2014 Type of farm Number of farms Subsidies (MKD million) % of farms % of Subsidies Subsidy intensity Specialist field crops 130 4.973 23.2 11.5 6.40 Specialist horticulture 32 0.159 5.7 0.4 0.72 Specialist permanent crops 73 2.527 13.0 5.8 7.80 Specialist grazing livestock 142 23.973 25.3 55.2 14.69 Specialist granivores 12 0.183 2.1 0.4 6.76b Mixed cropping 42 3.509 7.5 8.1 12,05 Mixed livestock holdings 7 0.991 1.2 2.3 Mixed crops-livestock 123 7.113 21.9 16.4 14.59 Total 561 43.428 100.0 100.0 11.11 Source: FADN 2014 data and World Bank staff calculations. Note: a. Subsidy intensity is defined as the value of subsidies divided by the value of total output; b. subsidy intensity for both specialist crops and mixed livestock holdings. The distribution of subsidies is positively correlated with economic farm size but is more equitable than in EU28 countries. In 2914 larger farms, 42.4 percent of the sample, accounted for 78.3 percent of total subsidies; medium-sized farms (25.3 percent of the sample, 13.3 percent; and smaller farms, about 33 percent of the sample, only 8.3 percent (Table 9.17). Thus, subsidy intensity is much higher for larger farms. That said, the distribution is more equitable if the Eurostat farm economic size classification scheme is adopted (Table 9.17) Table 9.17. Distribution of Direct Payments by Farm Size, 2014 Type of holding No. of farms Subsidies (MKD million) % of farms % of subsidies Subsidy intensity Small farms (up to €4000) 181 3.610 32.3 8.3 9.40 Medium farms (€4,001–8,000) 142 5.795 25.3 13.3 9.00 Large farms (over €8,000) 238 34.023 42.4 78.3 11.81 Total 561 43.428 100.0 100.0 11.11 Small farms (up to €8,000) 323 9.405 57.6 21.7 9.15 Medium farms (€8,001–25,000) 175 16.720 31.2 38.5 11.20 Large farms (over €25,000) 63 17.303 11.2 39.8 12.47 Total 561 43.428 100.0 100.0 11.11 Source: FADN 2014 data and World Bank staff calculations. Note: Subsidy intensity is defined as value of subsidies divided by value of total output. Subsidies by agricultural area are positively correlated with farm size, though not to the extent observed for economic size(Table 9.18). Farms smaller than 0.5 ha, 22 percent of the sample, receive 16 percent of subsidies. These farms appear to be more heavily subsidized than the net two groups (farms 0.5–1 ha and those 1–3 ha). In fact, the small-firm subsidy intensity of 10.11 is only exceeded by the largest farms (over 10 ha), which have a subsidy intensity of 16.28. Finally, farms up to 3 ha receive 30.3 percent of total subsidies. 184 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table 9.18. Distribution of Direct Payments by Farm Area, 2014 Type of holding No. of farms Subsidies (MKD million) % of farms % of subsidies Subsidy intensity Up to 0.5 ha 124 6.951 22.1 16.0 10.11 0.51–1 ha 38 0.986 6.8 2.3 7.41 1.01–3 ha 152 5.196 27.1 12.0 7.44 35 ha 103 5.345 18.4 12.3 10.44 5–10 ha 81 6.664 14.4 15.3 8.85 Over 10 ha 63 18.286 11.2 42.1 16.28 Total 561 43.428 100.0 100.0 11.11 Source: FADN 2014 data and World Bank staff calculations. Note: Subsidy intensity is defined as value of subsidies divided by value of total output. 9.5  Reform Options Short Term 1. Gradually reduce the size of direct farm subsidies from 0.84 percent of GDP to the EU-28 average of 0.33 percent to improve efficiency and support structural transformation. Farms in North Macedonia, particularly subsidized farms and those in heavily supported sectors like tobacco and livestock, are not technically efficient. Subsidies have a negative impact on the technical efficiency of farms and do not seem to address the structural problems of agriculture, even though government support for agriculture as a share of GDP, is much higher than in the rest of the WB6 and far above the EU-28 average, which makes reducing direct farm subsidies a high priority for the short term. This would both improve the efficiency of farms in North Macedonia and allow for reallocation of scarce public resources to areas of greater need. Within agriculture, for instance, these resources could be used to build up institutional capacity for delivering such important services as extension and knowledge transfer systems, and on sanitary and phytosanitary measures for make the sector’s more competitive. 2. To heighten the impact of public spending on farm competitiveness, move from largely coupled and unconditional subsidies to decoupled and conditional support. Current farm subsidies have led to a highly distorted allocation of production factors and thus to unproductive and inefficient farms. To stimulate a market orientation for production decisions it will also be necessary in the medium term to follow the newer CAP practices that involve a convergence of area-based support and even a flat rate of support per ha. Recent experience in the EU-28 shows that decoupling payments and focusing on cross-compliance (e.g., adoption of good agricultural practices) would give of public spending more impact on agricultural performance and farm competitiveness. Decoupled subsidies would allow farmers to respond to market demand rather than subsidy rates, which should shift land use and production toward higher-value products where North Macedonia appears to have a comparative advantage. Cross-compliance, introduced in 2013 with enhancements planned soon, needs to be implemented gradually, with time to actively sustainable farming practices, make Macedonian products more competitive, and facilitate the transfer of scarce farmland from less efficient to more efficient farmers, particularly through land rental. Chapter 9: Rebalancing Government Support to Improve Agriculture 185 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 3. Intensify spending on programs that support rural development. It has been found that farmers in North Macedonia who invest in fixed assets are more efficient, and that rural development has more impact economy-wide than farm subsidies. North Macedonia in recent years has been able to absorb significantly more rural development programs, which suggests that additional funds could be spent with good effect. Yet, the low technical efficiency of North Macedonia's farms underscores the need for more attention to the enabling environment and for investment in such public services as irrigation, agricultural extension, knowledge sharing programs, business development, and productive partnerships. Resources for these programs could be freed up by reducing current farm subsidies. Medium Term 4. Modernizing agriculture and making it more productive and competitive is paramount. The relationship between subsidies and farm modernization is complex. Heavily-subsidized larger livestock farms have been using direct payments to accumulate capital; the influence of capital on productivity is also high even for small crop farms, which receive comparatively little support. The TE gap between these two types of farms make it clear that coupled subsidies shift resources to less-competitive farm activities even through investments, which pushes agriculture in North Macedonia into an uncompetitive pattern. Farm size, specialization, and subsidization status do not affect how much capital contributes to productivity. Farms in North Macedonia are labor- intensive and except for large livestock farms, they operate with decreasing returns of scale. 5. Finding the “missing middle” is critical to transformation of the sector. No matter what the type of specialization or the type of support schemes, small and large farms use their inputs more efficiently than mid-size units. On the one hand, this indicates that in North Macedonia the scale of factors of production should be carefully considered in a context of land consolidation, where factor markets are imperfect, and where scale could be achieved through upward aggregation of agricultural products, creation of producer associations, and a focus on high-value crops. On the other hand, it raises concerns about how much potential medium-size firms have to modernize and questions about how to deal with such structural difficulties as too little access to finance, technology, and markets, that make it harder to transform small producers into more efficient larger units. Clearly, current agricultural policies cannot support such transformation; more comprehensive thinking and incentives will be necessary before the enabling environment for agriculture can be improved. This could be a real turning point, considering that the agricultural labor force in North Macedonia is relatively young. 6. Distinguish between agricultural support and social assistance to make government use of its fiscal resources more efficient. Since farm subsidies in North Macedonia have done little to reduce poverty, a rethinking of the mix of measures to reduce rural poverty is recommended. Better- targeted social assistance could be a more effective way to deal with rural poverty, agricultural support directed to building the productive. 186 Chapter 9: Rebalancing Government Support to Improve Agriculture SOWING THE SEEDS OF A SUSTAINABLE FUTURE References Aguirregabiria, V. 2009. “Econometric Issues and Methods in the Estimation of Production Functions.” Working Paper Number 15973, Munich Personal RePEc Archive, https://mpra.ub.uni-muenchen. de. APM Database for Western Balkan Countries, 2016. Agricultural Policy Measures Database compiled as part of the SWG Projects. http://app.seerural.org/agricultural-statistics/ Atoyan, R., and I. Jankulov. 2015. “Western Balkans: The Quest for Jobs.” In Regional Economic Issues, April 2015: Europe: The Western Balkans: 15 Years of Economic Transition, Washington DC: International Monetary Fund. Banker, R. D., A. Charnes, and W. W. Cooper. 1984 “Some Models for Estimating Technical and Scale Inefficiencies in Data Envelopment Analysis.” Management Science 30: 1078–92. Blazevski, N. M. n.d. Baseline Report on Access, Completion and Prevention of Dropout in Education in the SEE Countries: Implementation of SEE 2020 Strategy—Support to the Education and Competencies Dimension of the SEE 2020 Strategy. pp. 53–64. Sarajevo, Bosnia and Herzegovina: Regional Cooperation Council. Blazevski, N. M., and M. Ristovska. 2012. “Is Government Effective in Promoting Human Capital: The Case of FYR Macedonia M. In Labor Market and Skills in the Western Balkans, ed. by M Arandarenko and W. Bartlett Eds., pp. 129–155. Skopje, FYR Macedonia: Foundation for the Advancement of Economics. Boeters, S., Böhringer, C., Büttner, T., & Kraus, M. (2010). Economic effects of VAT reforms in Germany. Applied Economics, 42(17), 2165-2182. Bokusheva, R., and L. Cechura, 2017. “Evaluating Dynamics, Sources and Drivers of Productivity Growth at the Farm Level”. OECD Food, Agriculture and Fisheries Papers, No. 106. OECD Publishing, Paris. http://dx.doi.org/10.1787/5f2d0601-en. Bowlin, W.F. 1998. “Measuring Performance: An Introduction to Data Envelopment Analysis (DEA). Journal of Cost Analysis 7: 3–27. Center for Economic Analyses CEA, 2016, Benefits and Costs from Foreign Direct Investments in the Technological Industrial Development Zones: the Case of FYR Macedonia in the Period of 2007– 2014, Skopje. Cohen, A.J., M. Brauer, R. Burnett, H. R. Anderson, J. Frostad, K. Estep, K. Balakrishnan, B. Brunekreef, L. Dandona, R. Dandona, and V. Feigin. 2017. “Estimates and 25-year Trends of the Global Burden of Disease Attributable to Ambient Air Pollution: An Analysis of Data from the Global Burden of Diseases Study 2015. The Lancet, 38910082, pp.1907–18. Copenhagen Economics (2008). Study on reduced VAT applied to goods and services in the Member States of the European Union (No. 13). Directorate General Taxation and Customs Union, European Commission Cyan, Musharraf R., Jorge Martinez-Vazquez, and Andrey Timofeev. 2012. “Financing Equitable Service Delivery for All Citizens.” UNDP, Skopje, FYR Macedonia. Cylus, J., I. Papanicolas, and P. C. Smith. 2017. “Identifying the Causes of Inefficiencies in Health Systems. Eurohealth 232: 3–7. De Mooij, R. A., & Saito, I. (2014). Japan’s Corporate Income Tax: Facts, Issues and Reform Options (No. 14-138). International Monetary Fund. References 187 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Devereux, M. P., and R. Griffith. 2003.”Evaluating Tax Policy for Location Decisions. International Tax and Public Finance 102: 107–26 Dimitrievski, D., A. Kotevska, and I. Janeska-Stameskovska. 2017. “FYR Macedonia: Agricultural Policy Development and Assessment.” In Analysis of the Agricultural and Rural Development Policies of the Western Balkans, ed. by T. Volk et al. Seville: IPTS-JRC. Dimitrievski, D., A. Kotevska, I. Janeska-Stameskovska, E. Tuna, and M. Nacka. 2014. “Agriculture and Agricultural Policy in FYR Macedonia.” In Agricultural Policy and European Integration in Southeastern Europe, ed. by T. Volk et al. Rome: FAO. Eriksen, Michael, Judith Mackay, Neil Schluger, Farhad Islami, and Jeffrey Drope. 2015. The Tobacco Atlas: Fifth Edition. https://tobaccoatlas.org/. Espinosa, M., D. Psaltopoulos, F. Santini, E. Phimister, D. Roberts, S. Mary, T. Ratinger, D. Skuras, E. Balamou, M. A. Cardenete, and S. Gomez y Paloma. 2014. “Ex-Ante Analysis of the Regional Impacts of the Common Agricultural Policy: A Rural–Urban Recursive Dynamic CGE Model Approach.” European Planning Studies 22: 1342–67. European Cancer Observatory. 2017. Cancer Factsheets and Database. Lyon: EUCAN. http://eco.iarc.fr/ EUCAN/CancerSearch.aspx. European Commission. 2011. Impact Assessment: Common Agricultural Policy towards 2020. Commission Staff Working Paper. SEC2011 1153. European Commission, Brussels. European Commission. 2014.” Guidelines on Regional State Aid for 2014–2020.” Competition Policy Brief Issue 14. Brussels: EC. ——. 2015. FYR Macedonia: Agriculture, Rural Development and Food Safety, Veterinary and Phytosanitary Policy. Instrument for Pre-Accession Assistance IPAII 2014–20. Brussels: European Commission. ——. 2015. The 2015 Ageing Report, Economic and Budgetary Projections for the 28 EU Member States (2013–2060). Brussels: EC, ——. 2017. “Agriculture in the European Union and the Member States–Statistical Factsheet 2017.” DG Agriculture and Rural Development, Brussels. European Union. 2016. “Monitoring and Evaluation of the EC-IPA Rural Investment Single-Donor Trust Fund.” Brussels: EU. Eurostat. 2017. European Statistics. Database. Brussels: European Commission. http://ec.europa.eu/ eurostat/data/database. Feruglio, Nicoletta, Jorge Martinez-Vazquez, and Andrey Timofeev. 2008. “Assessment of Fiscal Decentralization in FYR Macedonia.” Skopje: UNDP FYR Macedonia. FYR Macedonia 2012. EU Instrument for Pre-Accession IPA: Rural Development Programme 2007–2013. Fourth modification. Skopje. ——. 2012 published/2014 revised. Republic of Macedonia—Monitoring the Situation of Children and Women—Multiple Indicator Cluster Survey 2011. Skopje. ——. 2015. “EU Instrument for Pre-Accession for Rural Development IPARD: National Programme for Agriculture and Rural Development 2014–2020.” Skopje. ——. n.d. The Republic of FYR Macedonia Comprehensive Education Strategy for 2016–2020: first draft. Skopje: Ministry of Education and Science of the FYR Macedonia. ——. 2018. Economic Reform Programme: 2018–2020. Skopje: Republic of FYR Macedonia Ministry of Finance. Global Burden of Disease Collaborative Network. 2017. Global Burden of Disease Study 2016 GBD 2016 Disability-Adjusted Life Years and Healthy Life Expectancy 1990–2016. Seattle, United States: Institute for Health Metrics and Evaluation (IHME). 188 References SOWING THE SEEDS OF A SUSTAINABLE FUTURE Gómez-Mera, L., T. Kenyon, Y. Margalit, J. G. Reis, and G. Varela. 2014. New Voices in Investment: A Survey of Investors from Emerging Countries. Washington, DC: World Bank Publications. Health Insurance Fund of Macedonia. 2017. Annual Report for 2016. Skopje: Health Insurance of Macedonia Hanushek, E.A. 2014. “The Economic Relevance of Early Childhood Development.” Prepare for Life! Raising Awareness for Early Literacy Education. 249-253. Herczyński, J. 2011. Financing Decentralized Education in FYR Macedonia. https://www.ssrn. com/ abstract=2524821 House, W. J. 2016. The Results of a Tracer Study of TVET/HE Graduates in FYR Macedonia. Skopje: Ministry of Education and Science of the FYR Macedonia. Institute of Public Health. 2014. Annual Report on the Health Status of the Population in the Republic of Macedonia 2013. Skopje: Institute of Public Health. ——. 2017. “World Obesity Day 2017.” Skopje, FYR Macedonia: Institute of Public Health. International Labour Force (ILO). N.d. Employment, Unemployment, and Earnings, 2014–18. http:// unstats.un.org/unsd/mbs/data_files/t11.pdf. IMF. 2015. “Export Competitiveness in FYR Macedonia.” Selected Issues Paper, IMF Country Report No. 15/243. Washington, DC: International Monetary Fund. ——. 2016. FYR Macedonia TADAT Performance Assessment Report. Washington, DC: International Monetary Fund. Kleinhanß, W., C. Murillo, C. San Juan, and S. Sperlich. 2007. “Efficiency, Subsidies, and Environmental Adaptation of Animal Farming under CAP.” Agricultural Economics 36: 49–65. Kontolaimou, A., I.Giotopoulos, and A. Tsakanikas, A. 2016. „A Typology of European Countries based on Innovation Efficiency and Technology Gaps: The Role of Early-Stage Entrepreneurship.” Economic Modelling 52: 477–84. Ladvenicová, J., and S. Miklovičová. 2015. “The Relationship between Farm Size and Productivity in Slovakia.” Visegrad Journal on Bioeconomy and Sustainable Development, 4: 46–50. Lampietti, J. A., D. G. Lugg, P. van der Celen, and A. Branczik, A. 2009. The Changing Face of Rural Space: Agriculture and Rural Development in the Western Balkans. World Bank, Washington. DC. Lau, L. J., and P. A. Yotopoulos. 1971. “A Test for Relative Efficiency and Application to Indian Agriculture.” American Economic Review 61: 94–107. Levitas, T. 2011. “Local Government Finances and the Status of Fiscal Decentralization in FYR Macedonia: A Statistical Review, 2008–2011.” IDG Working Paper No. 2011–02. Arlington, VA: International Development Group. MAFWE. 2007a. “Law of Agriculture and Rural Development.” Ministry of Agriculture, Forestry and Water Economy. Skopje: Ministry of Agriculture, Forestry and Water Economy. ——. 2007b. National Strategy for Agriculture and Rural Development for 2007–13. Skopje: Ministry of Agriculture, Forestry and Water Economy. ——. 2010a. Annual Report for Agriculture and Rural Development. Skopje: Ministry of Agriculture, Forestry and Water Economy. ——. 2010b. “Law of Agriculture and Rural Development.” Skopje: Ministry of Agriculture, Forestry and Water Economy. ——. 2014. National Agricultural and Rural Development Strategy NSARD 2014–2020. Skopje: Ministry of Agriculture, Forestry and Water Economy. ——. 2016. Standard Results for Agricultural Holdings Covered by FADN for the Accounting Year of 2014. Skopje: Ministry of Agriculture, Forestry and Water Economy. References 189 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Manevska–Tasevska, G. 2012. Efficiency Analysis of Commercial Grape-Producing Family Farms in the FYR Macedonia. Doctoral Thesis, Faculty of Natural Resources and Agricultural Sciences, Department of Economics, Swedish University of Agricultural Sciences, Uppsala. Martinez-Vazquez, J., and S. Lago-Peñas, eds. 2013. The Challenge of Local Government Size: Theoretical Perspectives, International Experience and Policy Reform. Cheltenham: Edward Elgar. Mary, S. 2013. “Assessing the Impacts of Pillar 1 and 2 Subsidies on TFP in French Crop Farms.” Journal of Agricultural Economics 64: 133–44. Masterson, T. 2007. “Productivity, Technical Efficiency and Farm Size in Paraguayan Agriculture.” Working Paper No. 490, Levy Economics Institute, New York. Mellor, J. W. 1969. “The Subsistence Farmer in Traditional Economies.” In editor. Subsistence Agriculture and Economic Development, edited by C. R. Wharton, Jr. Aldine: Chicago. Mendoza, E., A. Razin, and L. Tesar. 1994. “Effective Tax Rates in Macroeconomics: Cross-Country Estimates of Tax Rates on Factor Incomes and Consumption.” Journal of Monetary Economics 34: 297–323 Meisner, C., D. Gjorgjev, and F. Tozija. 2015. “Estimating Health Impacts and Economic Costs of Air Pollution in the Republic of Macedonia.” South Eastern European Journal of Public Health 41. Mickovska-Raleva, A., A. T. Misoska, O. H. Zaevska, S. Cherepnalkovska, and V. K. Ivanovikj. 2017. From Numbers to Practice–Identification and Analysis of the Indicators Related to the Quality of the Teaching Process in Primary Education in FYR Macedonia. Skopje, FYR Macedonia: Finance Think. Milevska, K.N., S. Chichevalieva, N. A. Ponce, E. van Ginneken, and J. Winkelmann. 2017. “The Former Yugoslav Republic of Macedonia: Health System Review.” Health Systems in Transition 193: 1. National Bank of the Republic of FYR Macedonia. 2016. Quarterly Report, August 2016. Skopje: [[insert Bank acronym]]. OECD. 2017. Tracking Special Economic Zones in the Western Balkans: Objectives, Features and Key Challenges. Paris: OECD Publishing. ——. 2016, Consumption Tax Trends 2016: VAT/GST and Excise Rates, Trends and Policy Issues. Paris: OECD Publishing. ——. 2017. OECD Pensions at a Glance 2017. Paris: OECD Publishing. Pastor, J. T. 1996. “Translation Invariance in DEA: A Generalization.” Annals of Operations Research 66: 93–102. Petrovska, M., A. Martinovska-Stojceska, B. Öhlmer, and Kotevska. 2013. “Technical Efficiency of Macedonian Pig Farms.” Agroeconomia Croatica 3: 18–25. PIOM (Pension and Disability Insurance Fund) Annual report. Pringle, R., Burlini, E. and Schiessl, M. 2014. Evaluation on the IPARD Programme and the Aid for the Agricultural Sector and Rural Development in the FYR Macedonia. Skopje. Psaltopoulos, Demetrios, Eudokia Balamou, and Kenneth J. Thomson. 2006. Rural–Urban Impacts of CAP Measures in Greece: An Inter-regional SAM Approach, Journal of Agricultural Economics, 57: 441–58. Rutkowski, J. 2010. Demand for Skills in FYR Macedonia. Washington, D.C.: World Bank. Saez, E., J. Slemrod and S. Giertz. 2012. “The Elasticity of Taxable Income with Respect to Marginal Tax Rates: A Critical Review.” Journal of Economic Literature, 50 (1): 3-50. Salmi, J. 2016. Design and Implementation of a Sustainable Funding Model for the FYR Macedonian Tertiary Education System. Skopje: Ministry of Education and Science of the FYR Macedonia. Sarkis J. 2007. “Preparing Your Data for DEA.” In Modeling Data Irregularities and Structural Complexities in Data Envelopment Analysis, edited by J. Zhu and W. D. Cook. Bosston, MA: Springer. 190 References SOWING THE SEEDS OF A SUSTAINABLE FUTURE Sauer, J., M. Gorton, and S. Davidova. 2015. “Migration and Farm Technical Efficiency: Evidence from Kosovo.” Agricultural Economics 46: 629–41. Savov, E. 2017. Fiscal Decentralization Indicators for South-East Europe: 2006–15. Skopje, FYR Macedonia: NALAS. Schneider, F., A. Buehn, and C. Montenegro. 2010. “Shadow Economies All over the World. New Estimates for 162 Countries from 1999 to 2007.” Policy Research Working Paper 5356, World Bank Group, Washington, DC. Sejdini, M. 2016. “An Overview of the Reformed Local Government in FYR Macedonia.” Business and Economic Research 61: 440–63. doi: http://dx.doi.org/ 10.5296/ ber.v6i1.9350. Simar, L., and P. W. Wilson. 1998. “Sensitivity Analysis of Efficiency Scores: How to Bootstrap in Nonparametric Frontier Models.” Management Science 44: 49–61. ——. 2000. “Statistical Inference in Nonparametric Frontier Models: The State of the Art.” Journal of Productivity Analysis 13: 49–78. Sini, M.P. 2009. “Evolution of Efficiency Factors, Different Ways by Which It Can be Reached and Competitiveness of Small Farms.” Paper presented at 111 EAAE-IAAE Seminar on Small Farms: Decline or Persistence?, University of Kent, Canterbury, UK,26th-27th June. Smoke, P. 2015. “Accountability and Service Delivery in Decentralising Environments: Understanding Context and Strategically Advancing Reform. A Governance Practitioner’s Notebook: Alternative ideas and Approaches.” OECD Working Paper 2015, OECD, Paris. Spicka, J. 2014. “The Regional Efficiency of Mixed Crop and Livestock Type of Farming and its Determinant.” Agris on-line Papers in Economics and Informatics 6: 99–109. State Statistical Office. 2016. “Supply and Use Tables, 2013.” Skopje, FYR Macedonia: State Statistical Office. STAX (Sugar, Tobacco, and Alcohol Taxes Group). 2018. “Sugar, Tobacco, and Alcohol Taxes to Achieve the SDGs.” The Lancet. Published online May 29, 2018. UNESCO. 2011. World Data on Education: The Former Yugoslav Republic of FYR Macedonia. http://www. ibe.unesco.org/fileadmin/user_upload/Publications/WDE/2010/pdf-versions/The_Former_ Yugoslav_Rep_of_FYR Macedonia.pdf. UNICEF. 2013a. Formula for Financing the System for Care and Education of Preschool Children in the Republic of FYR Macedonia, and Plan to Gradually Increase the Inclusion of the Children with ECD Programmes. Skopje, FYR Macedonia: UNICEF. ——. 2013b. Functional Analysis of the System for Care and Education of Preschool Children in the Republic of FYR Macedonia. Skopje, FYR Macedonia: UNICEF. United Nations, Department of Economic and Social Affairs, Population Division. 2017. World Population Prospects: The 2017 Revision, Volume I: Comprehensive Tables. New York: UN. World Bank. 2013. FYR Macedonia, former Yugoslav Republic of: Analysis of the Agricultural Support Programs. Report No. ACS5258. ECSAR Europe and Central Asia. Washington DC. ——. 2014. The Inverting Pyramid : Pension Systems Facing Demographic Challenges in Europe and Central Asia. Washington, DC: World Bank. ——. 2015a. FYR Macedonia Public Expenditure Review—Fiscal Policy for Growth. Macroeconomics and Fiscal Management. Washington, DC: World Bank. ——. 2015b. Options for Low Income Countries’ Effective and Efficient Use of Tax Incentives for Investment: Tools for the Assessment of Tax Incentives. Washington, DC: World Bank. ——. 2011. Putting Higher Education to Work: Skills and Research for Growth in East Asia. Washington, DC: World Bank. References 191 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 ——.2017a. World Development Indicators database. Washington, DC: World Bank. http://databank. worldbank.org/data/reports.aspx?source=world-development-indicators. ——. 2017b. Western Balkans: Agriculture for Jobs and Growth in the Western Balkans – A Regional Study. Report No. AUS19883. Agriculture Global Practice GFA03, South Eastern Europe and Balkans ECCU4, Europe and Central Asia (ECA), Washington DC. ——. 2019. Tobacco use and tobacco taxation in North Macedonia. Washington, DC: World Bank. World Bank, IMF, OECD and UN. 2015. Options for Low Income Countries’ Effective and Efficient Use of Tax Incentives for Investment: Tools for the Assessment of Tax Incentives. Washington, DC: World Bank. WHO (World Health Organization). 2017a. WHO Report on the Global Tobacco Epidemic 2017. Monitoring Tobacco Use and Prevention Policies. Geneva: WHO. ——. 2017b. European Health for All Database. Copenhagen: WHO. https://gateway.euro.who.int/en/ data-sources/european-health-for-all-database/ Zhu, X., and A. O. Lansink. 2010. “Impacts of CAP Subsidies on Technical Efficiency of Crop Farms in Germany, the Netherlands and Sweden.” Journal of Agricultural Economics, 61: 545–64. 192 References SOWING THE SEEDS OF A SUSTAINABLE FUTURE Annex 1. The Government Program, 2017–20 The goals of the Government Program for 2017–2020 are to achieve stronger economic growth, support creation of jobs, and increase living standards. Specifically, it plans to support domestic and foreign companies in in creating jobs, opening new markets, and increasing trade. It supports SMEs by working to minimize administrative procedures and facilitate access to credit; promoting tourism and its development; raising the minimum wage; direct supporting to companies to employ unemployed persons and provide above-average salaries; and improving road and utility infrastructure. health and education services. and environmental protection. Specific reforms outlined in the program are to: •• Reform direct taxes and excises  by introducing progressive tax rates for the PIT with a higher rate for high earners and separate PIT returns for citizens with very low monthly salaries. It might also raise property tax rates; eliminate f the maximum top-level salary contributions; and increase duties on luxury goods. •• Support infrastructure development b  y constructing or rehabilitating national, regional, and local roads; constructing and modernizing railway corridors and upgrading transport capacity; adding more energy capacity, including renewable energy; constructing water supply and wastewater systems; and investing in ICT to raise the quality of public services, thereby promoting digital literacy, access, and inclusiveness. •• Protect the most vulnerable t  hrough programs to promote employment and added incentives for related private sector activity; better-targeted social assistance to reduce poverty below 16 percent by 2020; broadening access to health and education services; and introducing protections from energy poverty. •• Build up agricultural production and improve farmer quality of life  by direct subsidies to profitable and market-oriented agricultural holdings; budget transparency for the National Program for Agriculture and Rural Development; consolidation of agricultural land; and horizontal integration of agricultural food entities and formation of cooperatives. •• Improve education by investing in its infrastructure; investing in science and connecting universities with private businesses to encourage and commercialize innovation; improving the quality of educational programs and their relevance to market needs; and modernizing teacher training. •• Improve the quality of health services  by constructing new hospitals and investing in equipment and related training; promoting the continued education of health workers; and improving service quality. •• Reform public administration and transparency  through transparent and efficient government and administrative operations; reinforcing the capacity of human resources management; adopting a strategy for good governance that will ensure the transparency of all state institutions; mandating consultations with citizens, business associations, and civil society organizations; and strengthening the control of Parliament. Ensuring full transparency of public spending will be a very high priority. Annex 1. The Government Program, 2017–20 193 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 EU and NATO integration efforts will lay the foundation for continued judiciary and public administration reforms responsive to EU political criteria; reforms to meet EU standards for a functioning market economy; further reforms to enhance competitiveness; and reforms necessary to accelerate NATO membership. The government is also supporting local self-government and balanced regional development. This would entail policies to increase not only the capacity of municipalities but also their transparency and accountability. A policy will be drafted to transferring new powers and more funding to municipalities so that they can successfully execute the responsibilities entrusted to them and provide better services to citizens. 194 Annex 1. The Government Program, 2017–20 SOWING THE SEEDS OF A SUSTAINABLE FUTURE Annex 2. The Fiscal Impact of Reform Options Table A2.1. Cummulative Fiscal Impact of Reform Measures, 2019–20 Cumulative Fiscal Reform Action Priority Impact (% of GDP unless otherwise indicated) Chapter 2: Public Financial Management 1. Draft and carry out a strategy for clearing arrears that is orderly, S –0.5–3.5 transparent, and credible. 2. Update the treasury information system to (a) record the date an invoice is received and its due date(s); (b) monitor and report monthly on all M Neutral accounts payable; and (c) record commitments before contracts are signed. 3. Strengthen MOF capacity for macrofiscal forecasting. M Neutral 4. Draft a comprehensive medium-term budget framework linked to the M Neutral overarching fiscal strategy. 5. Strengthen assessment of fiscal risks and their monitoring, e.g., by M Neutral integrating their oversight within a single coordinating body. 6. Improve the transparency of SOEs by publishing all SOE financial reports. M Neutral 7. Enhance the monitoring of local government financial performance. M Neutral Chapter 3: Fiscal Decentralization 1. Clarify functional assignments by better defining the responsibilities of S Neutral relevant bodies at each tier of government. 2. Consider measures to increase local own revenue capacity, such as identifying new tax sources (e.g., an annual local vehicle registration tax) S-M +0.5-3 and measures to improve collection of all taxes (property tax), charges, and communal fees. 3. Improve the effectiveness of conditional block grants by, e.g., making more transparent the calculation of transfers and selection of municipalities to S Neutral receive transfers. 4. Move to a more optimal size for municipalities by reducing the number of M Positive municipalities with fewer than 10,000 residents. 5. Improve the measurement of the spending needs related to functional M ? assignments. 6. Stabilize LG revenues by raising the amount of PIT collections they receive. M Neutral 7. Enhance the equalization capacity of general grants by reforming the M Neutral current formula. 8. Restructure the capital grants system. M Neutral Chapter 4: Tax Policy 1. Remove the tax exemption on the sales of securities; add a higher PIT tax bracket with a top marginal tax rate above 15 percent; broaden the CIT base; S 1 and increase the PIT standard deduction. 2. Expand the VAT base by removing preferential treatment for items not in S > 1.5 line with international standards. 3. Accelerate the increase in excise taxes on cigarettes and energy to meet EU S Positive requirements. 4. Review PIT, CIT, and VA tax expenditures and which tax incentives for corporations should be maintained and do a cost-benefit analysis of S-L Neutral special economic zones. 5. Review cadaster valuations and increase revenue from the property tax. M Positive 6. Move the CIT rate closer to 15 percent. L 1.2 (CIT of 13 percent) Annex 2. The Fiscal Impact of Reform Options 195 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Table A2.1. Cummulative Fiscal Impact of Reform Measures, 2019–20 (continued) Cumulative Fiscal Reform Action Priority Impact (% of GDP unless otherwise indicated) 7. For owner-occupied housing, lengthen the tenure to qualify for a capital L Positive gains exemption. owner-occupied housing. Chapter 5: Pension System 1. Adopt a consistent policy reform package that includes the following: (a) replace the current Swiss-indexation of pension benefits with CPI indexation; (b) gradually increase and equalize the retirement age for S-M 3.2 men and women; (c) declare a three-year pension indexation freeze; and (id) gradual increase the PAYG contribution rate by a total of 2 percentage points. 2. Reform the system of early retirement for hazardous occupations. S 0.14 3. Remove the accrual cap for voluntary switchers and equalize PAYG accrual S 2.09 rate differentials. 4. Consider linking retirement age to average life expectancy in North M-L 0.54 Macedonia. 5. Consider increasing the contribution rate for the second pension system M-L 1.51 pillar. Chapter 6: Social Assistance 0.1–0.2% in the medium term, 0.4% by 1. Make reform of the Parental Allowance program a priority. S-M 2028 (if the program is discontinued). 2. Introduce a new universal minimum income. In parallel, consolidate social 0.01% (from assistance benefits and services to the extent possible and integrate the S-M administrative and cash benefit and services information systems to improve targeting and targeting efficiencies) increase the coverage and poverty-reducing impact of social assistance. Chapter 7: Education 1. Study the feasibility of optimizing the school network. The study should form the basis for a publicized action plan with concrete targets and a S Positive timeline for completion. 2. Reevaluate the current primary and secondary per student funding formulas that protect low-density schools, creating inefficiently small class S Neutral sizes. 3. Introduce per student funding for preschool and higher education to S Neutral provide a transparent and more equitable way of distributing resources. 4. Make mathematics a required subject in Matura examinations. S Neutral 5. Increase instructional time in schools by lengthening the school day (with a focus on increasing the time spent on quality learning and on remedial or S ? extracurricular activities) and converting triple-shift to double-shift schools. 6. Draw up Develop a comprehensive teacher policy reform agenda encompassing strategies to raise the qualifications and motivation of M Neutral teachers. Chapter 8: Health 1. Prepare and act on a health facility master plan specialist health care level, taking the population’s needs into account, and formulate S Positive recommendations to optimize the supply side. 2. Review the distribution of human resources in health in view of the S Neutral population’s needs, with recommendations to improve staff management. 196 Annex 2. The Fiscal Impact of Reform Options SOWING THE SEEDS OF A SUSTAINABLE FUTURE Table A2.1. Cummulative Fiscal Impact of Reform Measures, 2019–20 (continued) Cumulative Fiscal Reform Action Priority Impact (% of GDP unless otherwise indicated) 3. Revise the list of drugs covered by the Health Insurance Fund to include new cost-effective therapies based on simplified procedures and clear S Positive criteria. 4. To reduce the unit prices for hospital drugs, centralize procurement of off- patent hospital drugs while ensuring closer quality controls and negotiate S Positive managed-entry agreements for high-cost patented medicines. 5. Consider increasing taxes on specific products, such as tobacco, that have S-M +0.8-2 negative health impact. Chapter 9: Agriculture 1. Reduce the size of direct farm subsidies from 0.84 percent of GDP to the EU-28 average of 0.33 percent to improve efficiency and support structural S +0.5 transformation. 2. Increase the impact of public spending on farm competitiveness by moving from largely coupled and unconditional subsidies to decoupled and S Neutral conditional support. 3. Increase spending on programs that support rural development. S Negative Note: Priorities are classified as S = Short Term; M = Medium Term; L = Long Term Annex 2. The Fiscal Impact of Reform Options 197 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Annex 3. Investment Tax Incentives in Selected Countries Albania Length of Tax Area Incentive Eligibility criteria incentives No VAT on machinery Machinery costing >€ 360 000. n/a No VAT for all foreign goods imported Being a company registered in a free n/a in a free zone. zone VAT No VAT on Albanian goods going to Being a company registered in a free n/a free zones zone No VAT on import of cement and steel Aimed at construction of hydropower n/a plants Excise No excise on fuel For oil-producing companies n/a No custom duties for all foreign Customs goods imported into a free zone 20% deduction on capital expenditure Being a company registered in a free 2 years within 2 years zone 50% reduction on 15% profit tax rate Being a company registered in a free 5 years for 5 years zone 150% deduction of wages and social contributions for the first year; new Being a company registered in a free Profit tax expenses in the next years also n/a zone deductible by 150% in subsequent years 200% deduction of training costs for Being a company registered in a free 10 years employees zone Being a company registered in a free 200% deduction of R&D spending 10 years zone Infrastructure No infrastructure taxes on new Being a company registered in a free n/a taxes projects zone Being a company registered in a free Real estate taxes No taxes on real estate 5 years zone Real estate Being a company registered in a free No transfer taxation on buildings n/a transfer taxes zone Bosnia and Herzegovina Length of Tax Area Incentive Eligibility criteria incentives Be a domestic or foreign company No limit Exemption from VAT on goods registered and operating in BIH's while VAT imported into the zone for production free zones and undertaking trading operating in activities the zone Customs-free imports of raw material, Be a domestic or foreign company semi- finished goods, and machinery registered and operating in BIH's free n/a and equipment zones. Customs-free imports of machinery Be a foreign company registered and and equipment except passenger operating in BIH. Investment to be n/a Customs duty vehicles and slot machines made by the foreign firm. Customs-free imports of all equipment and machinery to be used Be a domestic or foreign company registered and operating in a BIH free n/a for reconstruction and renovation zone. purposes 198 Annex 3. Investment Tax Incentives in Selected Countries SOWING THE SEEDS OF A SUSTAINABLE FUTURE Customs-free imports of raw Be a foreign company registered and materials, semi-finished goods, and operating in BIH. Move production n/a machinery and equipment lines from a third country. Customs duty Be a domestic or foreign company registered and operating in BIH. Customs-free imports of equipment Goods and equipment to be used n/a and goods solely for trade and promotion purposes. Excise exemption for all export goods Be a foreign company registered and Excise tax n/a operating in BIH. Be a foreign company registered 30% reduction in corporate income and operating in FBIH. Re-invest at Per fiscal tax for the pertinent fiscal year. least 50% of profit in production year equipment and machinery. Be a foreign company registered and operating in FBIH. Invest a minimum 50% reduction in CIT for 5 fiscal years of BAM 20 million (Bosnian marks) in 5 fiscal the Federation over 5 years, with an years CIT initial investment of at least 4 million BAM. Be a domestic or foreign company registered and operating in FBIH. Chance to claim tax deductions of Provide a full-time contract lasting at twice the annual gross salary of each least 12 months AND certify that the 12 months new employee. worker has not previously worked or has links with the company over the previous five years. Republika Srpska (RS) Be a foreign company registered and operating in RS. Real estate or Tax exemptions based on the property acquired to be used for the n/a acquisition of property production and other working needs of the company. CIT Be a foreign company registered and Tax exemptions based on the creation operating in RS. Employ at least 30 new employees, providing permanent, 12 months of new workplaces full time contracts for a minimum of one year. North Macedonia Length of Tax Area Incentive Eligibility criteria incentives Being a domestic or foreign company registered and operating in Total exemption from profit tax, Technological Industrial Development 10 years CIT currently 10% Zones (established to host high-tech clean-industry production that is export-oriented). Total exemption from personal PIT 10 years income tax, currently 10. The beneficiaries of TIDZs pay Lower rents significantly low rent given the area n/a of the parcels. The amount of aid for construction in TIDZ is limited to €0.5 million. Construction The beneficiaries of the zones use grants this aid in the form of a grant under conditions specified in the Law on TIDZ. Annex 3. Investment Tax Incentives in Selected Countries 199 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Exemption from local compensation, Communal taxes determined by the municipality in n/a which the TIDZ is located Zone beneficiaries can receive a grant for training employees in the amount Aid for training of 50% of the eligible training costs n/a employees for general trainings or 25% of the eligible training costs for general trainings and developments. Zone beneficiaries are exempt from paying VAT on goods and services in TIDZs (except supply intended for final consumption) and import VAT n/a of goods into TIDZs (except goods intended for final consumption), under conditions specified in the Law on TIDZ. Exemption from paying customs duties for equipment for performing Customs duties the activity in the Zone, machines n/a and spare parts, under conditions specified in the Law on TIDZ. Kosovo Length of Tax Area Incentive Eligibility criteria incentives Deferred payment of VAT on imports Up to 6 VAT Yes of selected plant machinery. months Only permitted for “heavy” equipment Additional deduction allowance of and vehicles. Applicable to new 10% on top of normal depreciation assets or assets first placed in service of the cost of acquisition of heavy CIT in Kosovo. Only allowable on assets n/a equipment (production lines for plant first used up to 31 December 2012 and machinery) and heavy transport (note, both income tax usage period vehicles. is extended to 31 December 2014). T hese incentives were provided Exemption from custom duties on Customs duty based on UNMIK Regulation NO. n/a machinery and raw material 2007/31 PART C. Business registration fees Exemption from fees of registering Manufacturing businesses n/a (Municipality of businesses involved in production Suhareka). Free use of Free use of municipal land for Up to 30 municipal land Conducting a new business business activities. years in Peja Montenegro Length of Tax Area Incentive Eligibility criteria incentives Any foreign investor establishing a business entity in Montenegro. Grants ranging from €3,000 to €10,000 Minimum investment value of per job created, depending on the One-off Grants €500,000, an assurance of at least 20 score obtained in the evaluation payment new jobs within three years from the criteria date of concluding the Use of Funds Agreement 200 Annex 3. Investment Tax Incentives in Selected Countries SOWING THE SEEDS OF A SUSTAINABLE FUTURE For the Being a domestic or foreign company duration Exemption from paying contributions registered in Monenegro. Hire of the for compulsory social insurance on PIT and SSC certain categories of disadvantaged contractual wage earning and personal income population (over 40s, Romas, Ashkalis engagement tax or Egyptians, long-term unemployed). with the employee. Customs duties are not paid for the goods entering the zone no matter Customs duty what their type or their purpose in the zone. VAT is not paid on importd goods VAT entering the zone no matter what their type or their purpose Goods entering the zone are not affectd by foreign-trade No foreign-trade restrictions. restrictions (permits, quotas etc.) No Goods stored Being a domestic or foreign company limitation in the zone registered and operating in the zone while in the may stay for Unlimited period zone unlimited period of time Goods may be temporarily taken from the zone into the rest of Montenegro or enter the zone from the rest of Montenegro to be improied, Goods may be assembled, tesed, of repaierd, or for temporarily marketing presentations etc taken in and out Goods entering Montenegrin territory of the zone. from the zone commercial purposes are subject to the payment of customs, custom duties and value added tax, bjt only for the foreign components present in them. In he zones, capital investments, and Capital investments derived transfer of profits are tax- free. Reductioin of 40% for all fiscal charges relating to construction Fiscal charges permits to be paid to be Municipality for construction of Bar for all buildings in the sites Free Zone Port of Barinendedfor manufacturing. Annex 3. Investment Tax Incentives in Selected Countries 201 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Serbia Length of Tax Area Incentive Eligibility criteria incentives 20–40% for investment projects - Investors who have investment (depending on the regional level of projects in sectors listed in the decree and apply for grants before th development) with specific limitations on the amount per new job created. project becomes operational. - Investors who can provide a Up to 30% increase (depending on minimum of 25% of eligible costs the regional level of development). from their own or other sources that Up to 20% increase of the eligible do not contain any other state grants. costs of gross salaries (for any - Investors are required to continue increase in the number of new jobs investingin and operating the same created over 200–1,000 new jobs locations for 5 years afer the initial investment is realised for a specified Grants period of time: large enterprises for n/a at least 5 years and SMEs for at least 3 yearsd. Employment Subsidies Programme - Investors cannot reduce the number (grant amount per employee: €850– of employees after the factory opens 1,700; Apprentice Programme (grant for 5 years for large enterprises, and 3 amount per employee: €170–210, years for SMEs. Retraining Programme (grant amount - Create at least 20-50 jobs and per employee: €850) invest at least €150,000–600,000 of the eligible costs of investment (depending on the regional level of development). Large investors must demonstrate incentive effects. Being a domestic or foreign company Exemption from corporate profit tax registered and operating in Serbia. CIT Invest an amount exceeding €8.5 10 years for period of 10 years million in fixed assets. Employ at least 100 additional employees. Customs-free imports of raw Being a domestic or foreign company Customs duty materials, semi-finished goods, registered and operating in Serbia's n/a andmachinery and equipment. free zone No Being a domestic or foreign company limitation Exemption from VAT on income VAT generated from commercial activities. registered and operating in Serbia's while free zone operating in the zone City building land and development lease fees exemptions or deductions, Being a domestic or foreign company Local tax registered and operating in Serbia. Be n/a ands other local fee exemptions or located in a specific municipality. deductions. Romania Length of Tax Area Incentive Eligibility criteria incentives Length of Incentives Awarded aid Eligibility criteria incentives Profits reinvested in certain assets, Asset must be held for half of its Year of such as technological equipment, CIT useful life or five years, w hichever is reinvest- software, desktops, and laptops are shorter. ment exempt from CIT. tax Reimburse 10%–50% of employer Business operating in targeted salary contributions for those in Wage subsidy economic sector and in a particular n/a targeted economic sectors, such as region R&D activities 202 Annex 3. Investment Tax Incentives in Selected Countries SOWING THE SEEDS OF A SUSTAINABLE FUTURE Size of reimbursement depends on region. Eligilbilty depends on Reimburse 10%–50% of capital continuation of business activity. (E.g,. n/a CAPEX subsidy spending for most economic sectors businesses who have discontinued a similar activity within the EEA in the past two years are not eligible) 50% of employee training costs fpr Select companies with growth Training subsidy n/a large companies and 70% for SMEs potential Lithuania Length of Tax Area Incentive Eligibility criteria incentives Double deduction of costs of Purchase of new fixed asset to qualifying asset: 100% in the year increase capacity of the firm in up to 2023 the cost were incurred and then as production, services, or sales normal depreciation 6 or 10 years, Firm must be located in a special depending economic zone. Additional on whether Exemption or 50% reduction in the requirements relate to the size the firm was CIT rate of the investment, the number of CIT established employees, and the type of business before activity. or after 1 January 2018 Firm must be located in a special economic zone. Additional requirements relate to the size Exempt from taxation of dividends n/a of the investment, the number of employees, and the type of business activity. Firm must be located in a special economic zone. Additional Real estate tax Exempt requirements relate to the size of the n/a investment, the number of employees and the type of business activity. Annex 3. Investment Tax Incentives in Selected Countries 203 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Annex 4. Education Regressions Table A4.1. North Macedonia: Determinants of Science Performance, PISA, 2015 Variables OLS Cluster school Multilevel Age –1.585 –4.641 (5.834) (5.623) Gender (1=female) 6.308** 1.940 (2.978) (2.959) Grade 11.23*** 15.10*** (3.995) (3.924) Kindergarten (More than 1 year) 11.49*** 12.06*** (3.848) (3.132) Macedonian student (D=1) 43.68*** 38.30*** (3.553) (4.356) ESCS 6.613*** 6.861*** (1.795) (1.734) ESCSS 39.50*** 18.04* (5.956) (10.53) General education (D=1) 27.08*** 40.62*** (3.894) (4.899) Public school –32.94*** –40.94** (7.943) (18.26) Rural (1=Rural) –8.227* –13.58 (4.839) (9.650) % Teacher master's degree 10.34 23.74 (7.563) (22.94) % of certified teachers –7.671 4.659 (5.972) (14.12) Teacher-directed instruction 6.937*** 5.791*** (1.615) (1.448) Disciplinary climate 6.672*** 6.130*** (1.681) (1.473) Shortage of educational material –1.705 –3.990 (2.134) (4.326) Shortage of educational staff –2.377 0.531 (2.189) (4.924) Class size 0.610*** 0.515 (0.201) (0.473) Student - teacher ratio 0.0921 0.308 (0.148) (0.382) School autonomy –6.703 2.086 (6.888) (14.65) 204 Annex 4. Education Regressions SOWING THE SEEDS OF A SUSTAINABLE FUTURE 393.2*** 426.8*** Constant (92.30) (90.89) Observations 3,555 3,555 R-squared 0.260 Source: World Bank staff calculations. Table A4.2. Factors Associated with the Matura Efficiency Score, OLS Regression (1) (2) (3) (4) Variables Efficiency Matura Efficiency Matura Efficiency Matura Efficiency Matura 0.00553 0.00432 0.00587 0.00478 Participation of block grant (0.00357) (0.00440) (0.00419) (0.00423) 0.000822 0.00328 0.00371 0.00377 % Personnel expenditure (0.00186) (0.00283) (0.00320) (0.00262) –0.000247 –0.00436 0.00924* Class size (0.00726) (0.0100) (0.00530) 0.0205 0.0262* 0.0204** Student teacher ratio (0.0128) (0.0149) (0.00852) –0.0467 –0.0865 –0.0689 –0.0820 Double-shift schools (0.0508) (0.0567) (0.0601) (0.0551) –0.00121 0.00107 0.00111 0.00138 % Teachers with university degree (0.00616) (0.00614) (0.00665) (0.00610) 0.000555 0.000803 0.00111 0.000825 % Teacher open ended contract (0.00160) (0.00206) (0.00216) (0.00203) –0.0394 0.0160 –0.0237 0.00636 Rural municipality (0.124) (0.127) (0.124) (0.120) –3.044** –2.768 –2.098 –2.594 Poverty (1.393) (1.856) (2.207) (1.811) 0.254 0.00456 -0.263 -0.147 Constant (0.721) (0.765) (0.764) (0.679) Observations 33 33 33 33 Regional fixed effects No Yes Yes Yes R-squared 0.364 0.482 0.416 0.477 Notes: Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Annex 4. Education Regressions 205 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Annex 5. Funding Formulas for Education Earmarked grants for primary education. Criteria for allocation of earmarked grants in primary education are: basic amount for the municipality, number of primary sudents in the municipality, and the population density of the municipality. EG = B + Pw * С Pw = P + (P x K) EG - Earmarked grant for primary education for the municipality B - Basic amount per municipality P - Total number of pupils in the municipality Pw - Weighted students in the municipality С - Standard (amount) per student K - coefficient (weight) for the population density of the municipality: •• up to 20 inhabitants per km2 - (1.4) •• from 20 to 35 inhabitants per km2 - (0.8) •• from 35 to 70 inhabitants per km2 - (0.6) Note: The coefficients by population density changed in 2008. Earmarked grants are no longer provided for secondary education. Block grants for primary education. Criteria for allocation of block grants for primary education by municipalities are: basic amount by municipality, number of students in the municipality, number of pupils in subject teaching, number of students with special needs, population density. BG = Bp + Pw * C Pw = P + (P * K + Pt * Kt + Ps * Ks) BG - Block grant for primary education for the municipality Bp - Basic amount per municipality Pw - Weighted students in municipality C - Standard (amount) per student P - Total number of students in the municipality Pt - Number of students in the subject teaching in the municipality Ps - Number of students with special needs in municipality Kt - coefficient (weight) for subject teaching (0.2) Ks - coefficient (weight) for students with special needs (1) K - coefficient (weight) for the population density of the municipality: •• up to 20 inhabitants per km2 - (0.6) •• from 20 to 50 inhabitants per km2 - (0.4 •• from 50 to 70 inhabitants per km2 - (0.2) 206 Annex 5. Funding Formulas for Education SOWING THE SEEDS OF A SUSTAINABLE FUTURE Block grants for secondary education. Criteria for allocating block grants for secondary education by municipalities and the City of Skopje are: basic amount, number of pupils in secondary education, number of students in vocational education, and number of secondary students with special education needs. BG = Bs + Pw * С Pw = P + (Ps* Ks + Pv x Kv) BG - Block grant for secondary education for municipality/City of Skopje Bs - Basic amount per municipality / City of Skopje Pw - Weighted students in the municipality / City of Skopje C - Standard (amount) per student P - Total students in secondary education in municipality / City of Skopje Ys - Number of students in vocational education in the City of Skopje Pv - Number of students with special needs in secondary education in the municipality / City of Skopje Ks - Coefficient (weight) for vocational education (0.1) Kv - Coefficient (weight) for students with special needs in secondary education (1) Note: Before 2018, here was no coefficient for students with special needs. Annex 5. Funding Formulas for Education 207 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Annex 6. Average Net Wages Table A6.1. Average Net Wage per Employee by Region, 2016 Republic of Vardar Southwest Southeast Pelagonia Polog Northeast Skopje Macedonia Region East Region Region Region Region Region Region Region Accommodation and 15 423 13 770 14 678 14 018 15 466 13 831 16 573 12 346 16 925 food service Administrative and 16 535 15 120 14 013 14 953 14 867 13 167 65 831 15 144 15 353 support service Agriculture, forestry and 16 873 17 736 16 130 16 600 14 900 17 698 16 760 16 370 19 187 fishing Manufacturing 17 096 16 427 13 954 17 051 14 415 16 451 13 975 12 892 24 833 Water supply; sewerage, waste management and 19 854 16 761 17 151 16 378 15 252 20 541 : 17 911 24 458 remediation activities Wholesale and retail trade; repair of motor 20 313 16 201 14 591 14 862 16 301 16 545 15 085 13 569 23 379 vehicles and motorcycles Arts, entertainment, and 21 096 18 125 19 091 16 874 18 889 20 337 25 246 18 658 22 102 recreation Construction 21 205 17 158 16 922 18 720 15 113 18 162 18 702 15 890 24 340 Education 22 190 20 022 22 281 21 379 20 367 20 886 23 755 20 887 23 189 Average net wage, paid in 22 342 17 699 16 701 20 044 17 506 20 600 21 484 16 952 26 243 denars, by region Transportation and 22 907 15 976 16 711 17 720 15 059 15 755 17 996 16 800 27 283 storage Real estate activities 23 969 22 187 17 855 21 291 : 23 347 17 495 : 25 376 Human health and social 24 571 20 893 20 814 22 111 20 892 22 847 22 499 21 070 27 940 work Other service activities 26 128 16 323 17 737 29 273 19 008 22 927 : 19 189 28 617 Mining and quarrying 26 232 15 053 29 715 19 628 : 26 294 : : 31 167 Public administration and defense; compulsory 26 734 24 225 23 825 23 231 23 896 24 889 23 665 22 593 27 623 social security Professional, scientific, 28 891 19 328 20 706 23 497 16 952 20 817 15 132 28 375 30 825 and technical Information and 36 489 25 406 22 396 19 446 45 940 26 703 : 18 610 37 850 communication Electricity, gas, steam and air conditioning 37 225 32 513 : 33 782 : : : : 37 055 supply Financial and insurance 39 701 32 131 31 086 32 129 33 857 37 213 33 086 30 032 40 641 activities 208 Annex 6. Average Net Wage SOWING THE SEEDS OF A SUSTAINABLE FUTURE Annex 7. Air Pollution in North Macedonia Population-weighted exposure to ambient particulate matter (PM2.5) pollution is defined as the average exposure of a nation's people to concentrations of suspended particles measuring less than 2.5 microns in aerodynamic diameter, which can penetrate deep into the respiratory tract and cause severe health damage. Long-term exposure to high levels of fine particles in the air contributes to a range of health problems, among them respiratory diseases, lung cancer, and heart disease. In 2015 at 4.5 micrograms per cubic meter, the mean annual exposure to PM2.5 in North Macedonia was 2.6 times higher than the EU average. Ambient PM2.5 pollution is responsible for high disability-adjusted life years (DALYs) and mortality rates in North Macedonia (Figure A7.1). Recent estimates suggest that reducing ambient PM10 or PM2.5 by 1 microgram per cubic meter could save North Macedonia an average of about €34 million a year on health costs (Meisner et al. 2015). Figure A7.1. Mortality and DALY Rates from Ambient Particulate Matter Pollution Compared, 2015 A. B. 70 1,400 60 1,200 50 1,000 40 800 30 600 20 400 10 200 0 0 UMI MKD STEE7 WB6 SP EU UMI MKD WB6 STEE7 SP EU Source: Cohen et al. 2017 and GBD data,and World Bank staff calculations. Annex 7. Air Pollution in North Macedonia 209 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Annex 8. Health Spending in North Macedonia and Comparators Health Spending, Health Spending Health Spending Health Spending, Country per Capita, PPP Total (% of GDP) Public (% of GDP) Private (% of GDP) (const. 2011 int. $) North Macedonia 6.5 4.1 2.4 851 Serbia 10.4 6.4 4.0 1,312 Montenegro 6.4 3.7 2.7 888 Georgia 7.4 1.6 5.9 628 Bosnia & Herzegovina 9.6 6.8 2.8 957 Croatia 7.8 6.4 1.4 1,652 Estonia 6.4 5.0 1.4 1,668 Slovenia 9.2 6.6 2.6 2,698 Lithuania 6.6 4.4 2.1 1,718 Bulgaria 8.4 4.6 3.8 1,399 Albania 5.9 2.9 2.9 615 Latvia 5.9 3.7 2.2 940 Slovak Republic 8.1 5.8 2.2 2,179 Kosovo Mauritius 4.8 2.4 2.4 896 Paraguay 9.8 4.5 5.3 873 Jamaica 5.4 2.8 2.6 476 Jordan 7.5 5.2 2.3 798 European Union 10.0 7.8 2.2 4,135 UMICs 6.2 3.4 2.7 1,032 Structural Peers 8.2 4.7 3.5 872 Western Balkans 8.8 5.5 3.3 1,036 STEE7s 7.7 5.3 2.4 1,731 210 Annex 8. Health Spending in North Macedonia and Comparator SOWING THE SEEDS OF A SUSTAINABLE FUTURE Annex 9. Structure of Agricultural Output, North Macedonia Table A9.1. Structure of Agricultural Output, 2005–15 2005 2010 2011 2012 2013 2014 2015 01 Cereals 12.0 7.1 10.9 8.8 9.9 10.1 7.2 02 Industrial crops 9.2 8.1 6.7 7.1 6.5 6.5 5.5 02.3 Raw tobacco 8.7 7.7 6.2 6.7 6.1 6.0 5.1 03 Forage plants 10.3 6.1 5.9 5.6 6.5 6.6 7.2 04 Vegetables 29.0 36.4 33.0 35.8 33.1 36.2 36.0 05 Potatoes 2.9 3.0 4.2 2.7 4.7 4.3 3.4 06 Fruits 11.1 10.3 10.8 11.2 11.0 9.0 11.6 07 Wine 3.9 3.1 3.3 3.3 4.0 3.4 4.9 10 Crop output (01 to 07) 78.4 74.1 74.9 74.5 75.7 76.1 75.9 11.1 Cattle 3.2 4.8 4.7 4.7 4.1 3.8 4.2 11.2 Pigs 3.4 4.4 4.5 3.9 3.3 3.2 3.8 11.3 Equines 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11.4 Sheep and goats 1.0 1.9 2.0 2.1 2.0 1.8 1.8 11.5 Poultry 1.2 0.7 0.6 0.8 1.0 0.9 0.8 12.1 Milk 9.5 11.4 11.0 11.7 11.7 12.0 11.6 12.2 Eggs 2.7 1.8 1.5 1.6 1.4 1.4 1.2 12.3 Other animal products 0.4 0.3 0.3 0.2 0.3 0.3 0.3 13 Animal output (11+12) 21.4 25.4 24.6 25.0 23.8 23.4 23.6 14 Agricultural goods output (10+13) 99.8 99.5 99.5 99.5 99.5 99.5 99.5 15 Agricultural services output 0.2 0.5 0.5 0.5 0.5 0.5 0.5 16 Agricultural output (14+15) 60,549 70,713 71,220 72,039 77,722 78,707 80,254 Source: World Bank staff calculations based on data from State Statistical Office. Annex 9. Structure of Agricultural Output, North Macedonia 211 REPUBLIC OF NORTH MACEDONIA  |  PUBLIC FINANCE REVIEW, MARCH 2019 Annex 10. Guaranteed Minimum Income Comparisons Figure A10.1. GMI for a Couple with Two Children, 2014 Percent of minimum wage 140 120 100 80 60 40 20 0 ry nia ria al Spain lgium reece oatia venia rance 21 lta lic d m ia ia ds ia rg d lic nga oma Bulga ortug Cr F EU- Ma epub Polan ingdo Eston Latv erlan thuan bou Irelan epub Hu R P Be G Slo R K t h L i x e m R vak ited Ne Lu Cze ch Slo Un Source: OECD. 212 Annex 10. Guaranteed Minimum Income Comparisons NORTH MACEDONIA | PUBLIC FINANCE REVIEW