Assessing Intergovernmental Fiscal Relations Financing Higher Education Controlling Health Care Expenditures Managing Fiscal Risks in Public-Private Partnerships
VOLUME I: MAIN REPORT VOLUME II: COUNTRY CASE STUDIES (CD)
Contents FOREWORD ...... 7 Daniela Gressani CONTRIBUTORS ...... 9 CONFERENCE AGENDA ...... 10 OVERVIEW ...... 11 Thomas Laursen I. FISCAL CHALLENGES FOR THE EU8 COUNTRIES ...... 27 Paulina Bucon and Emilia Skrok 1. FISCAL DEVELOPMENTS 1995-2004 ...... 27 2. MEDIUM-LONG TERM FISCAL PROSPECTS ...... 43 3. QUALITY OF FISCAL POLICY ...... 46 II. INTERGOVERNMENTAL FISCAL RELATIONS ...... 58 William Dillinger 1. INTRODUCTION ...... 58 2. EVALUATION CRITERIA ...... 58 3. THE SYSTEM EX ANTE ...... 59 4. POLITICAL AND ORGANIZATIONAL REFORMS ...... 60 5. ASSIGNMENT OF FUNCTIONS ...... 62 6. REVENUES ...... 66 7. DEBT CONTROLS ...... 73 8. DIRECTIONS FOR FURTHER REFORM ...... 75 III. FINANCING HIGHER EDUCATION ...... 80 Mary Canning, Martin Godfrey, and Dorota Holzer-Zelazewska 1. INTRODUCTION ...... 80 2. HIGHER EDUCATION SYSTEMS IN A COMPARATIVE PERSPECTIVE ...... 81 3. DIRECTIONS FOR FURTHER REFORM ...... 93 4. CONCLUSIONS ...... 96 IV. CONTROLLING HEALTH EXPENDITURES ...... 99 Mukesh Chawla 1. INTRODUCTION ...... 99 2. HEALTH EXPENDITURES ...... 103 3. HEALTH SECTOR DEFICITS AND DEBTS ...... 107 4. KEY EXPENDITURE AREAS IN THE HEALTH SECTOR ...... 109 5. SPENDING ON AGEING POPULATIONS AND MEDICAL TECHNOLOGY ...... 114 6. DIRECTIONS FOR FURTHER REFORM ...... 117 7. CONCLUSIONS ...... 129 V. MANAGING FISCAL RISKS IN PUBLIC–PRIVATE PARTNERSHIPS ...... 135 Hana Polackova Brixi, Nina Budina, and Timothy Irwin 1. INTRODUCTION ...... 135 2. THE FISCAL EFFECTS OF PPPS ...... 137 3. HOW FISCAL INSTITUTIONS AFFECT THE FISCAL COST OF PPPS ...... 143 4. DIRECTIONS FOR FURTHER REFORM ...... 147 5. CONCLUSIONS ...... 154 SUMMARY OF CONFERENCE DISCUSSIONS ...... 157 Thomas Laursen Boxes Box 01. Objectives of Intergovernmental Finance Systems ...... 17 Box I.1. EU8 Tax Reforms: ...... 33 Figure A. EU8 Effective Corporate Income Tax Rates (macro backward-looking approach)...... 33 Table A. Top statutory tax rate on personal income ...... 34 Table B. Top statutory tax rate on corporate income ...... 34 Table C. VAT rates ...... 34 Box I.2. Public Spending Measurement Issues ...... 36 Box I.3. Factors Affecting Expenditure Developments ...... 36 Figure A. Public Expenditure and Output Gap (%) ...... 37 Figure B.1. Political Economy at Work in the Baltic Countries ...... 37 Figure B.2. Political Economy at Work in the Visegrad Countries ...... 37 Box I.4. Case Study: Ireland ...... 47 Box I.5. Case Study: Slovakia ...... 48 Figure A. GG expenditure and GDP growth (% of GDP and %, respectively) ...... 49 Figure B. GG Revenue, Expenditure and Balance (% of GDP) ...... 49 Figure C. Composition of GG Revenues (% of GDP) ...... 49 Figure D. Composition of GG Expenditures (% of GDP) ...... 49 Box I.6. EBA Methodology ...... 53 Box II.1. A Snapshot of the Economics of Expenditure Assignment ...... 59 Box II.2. Hungary’s Municipal Bankruptcy Procedure ...... 75 Box II.3. Calculating the Formula Grant in England and Wales ...... 76 Box IV.1. Expenditure on Drugs (OECD countries; 1998-2003) ...... 110 Box IV.2. Health Reforms in Slovakia ...... 118 Box IV.3. Managing Expenditure on Drugs – the Experience of Slovakia ...... 121 Box V.1. How Does the Use of PPPs Differ by Sector? ...... 137 Box V.2. PPPs in the EU8 ...... 139 Box V.3. Building Risk Management Capacity in Turkey ...... 146 Tables Table I.1. Effect of Pension Reforms in the EU8 (% of GDP) ...... 29 Table I.2. EU8 Tax Burden 1995-2004 (% of GDP) ...... 32 Table I.3. EU8 Economic Classification of Public Expenditure 2003 (% of GDP) ...... 39 Table I.4. EU8 Functional Classification of Public Expenditure 2003 (% of GDP) ...... 39 Table I.5. Snow Ball Effect on Debt (% GDP) ...... 41 Table I.6. EU8 Expenditure and Revenue at Central and Local Government Level 2003 ...... 42 Table I.7. EU8 Debt Stabilizing Primary Fiscal Balance 2004 (% of GDP) ...... 44 Table I.8. EU8 Medium-Long Term Spending Projections (% of GDP) ...... 44 Table I.9. EU8 Sustainability Gap Indicators (% of GDP) ...... 46 Table I.10. EU8 Patterns of Large Fiscal Contractions 1998-2004 ...... 47 Table I.11. Composition of Public Spending in 2002 – Ranking ...... 51 Table I.12. EU8 Panel Growth Regressions 1995-2004 (dependent variable: growth rate of per capita GDP) ...... 54 Table I.13. Robustness of Base Variables ...... 54 Table I.14. Tax Structure and Output Growth (shares of total taxes) ...... 55 Table I.15. Expenditure Structure and Output Growth ...... 56 Table II.1. Timetable of Post-Soviet Reforms ...... 60 Table II.2. Local Share of Total Public Education Spending ...... 63 Table II.3. Subnational Spending on Health and Social Assistance (% of total public sector spending on these functions) ...... 64 Table II.4. Personal Income Tax Shared with Subnational Governments (2002) ...... 67 Table II.5. Property Taxes (% of GDP) ...... 69 Table II.6. Summary of Municipal Debt Regulations ...... 74 Table III.1. Higher Education Indicators (EU15 and EU8 countries) ...... 83 Table III.2. Tertiary Graduates in Science and Technology Employed as Professionals or Technicians (2003) ...... 84 Table III.3. Higher Education and the Labor Market (EU8 and EU15 countries) ...... 84 Table III.4. Public Financing of Higher Education (EU8 and EU15 countries, 2001) ...... 85 Table III.5. Financial Aid to Students (EU8 and EU15 countries, 2001) ...... 86 Table III.6. Higher Education Expenses Borne by Parents and Students (first degree, public institutions: Poland, Hungary and Latvia; US$ PPP) ...... 88 Table III.7. Poland: Enrollment Rate by Age Group and Household Consumption Quintile (2003) .... 88 Table III.8. Extent of Autonomy Enjoyed by Universities (selected EU15 and EU8 countries) ...... 91 Table III.9. Appointment of Leaders of Universities (selected EU15 and EU8 countries) ...... 92 Table IV.1. EU8 – Standardized Death Rates (per 100,000); different causes (2003) ...... 101 Table IV.2. EU8 – Scope of Services Covered by Social Health Insurance ...... 104 Table IV.3. Health Sector Costs in Slovakia (1998-2003; percent of total) ...... 105 Table IV.4. Lithuania – Debt of Health Insurance Fund to Providers (billion Litas) ...... 109 Table IV.5. Beds per 100,000 Inhabitants (EU8; 1993-2002) ...... 112 Table IV.6. Physician Salaries (selected EU8 countries; average monthly in local currency) ...... 113 Table V.1. Selected Liabilities from PPPs and Other Privately Financed Infrastructure Projects ...... 140 Table V.2. Possible Real Fiscal Effects of PPPs on Net Worth ...... 141 Table V.3. Improving Fiscal Institutions for PPPs at a Glance ...... 154 Figures Figure O1. Employment and Relative Poverty Risk in the EU (2001 or latest data) ...... 11 Figure O2. Employment and Debt in the EU (2004) ...... 12 Figure O3. Size of Public Sector and Per Capita Income in the EU (2004) ...... 12 Figure O4. Tax wedge and Employment Rate in the EU ( 2004) ...... 13 Figure O5. Fiscal Decentralization and Real GDP Growth in the EU8 1995-2003 ...... 14 Figure I.1. EU8 Fiscal Developments 1998-2004 (% of GDP) ...... 28 Figure I.2. EU8 General Government Revenue and Expenditure (% of GDP) ...... 30 Figure I.3. Visegrad Countries - Divergence from EU Program Fiscal Targets (% of GDP) ...... 31 Figure I.4. EU8 Change in Tax Burden 1995-2004 Relative to Initial Level ...... 31 Figure I.5. EU8 Composition of Tax Burden 1995-2004 (% of GDP) ...... 33 Figure I.6. EU8 Tax Wedge on Labor Costs - Low Wage Earners ...... 35 Figure I.7. General government, total revenue and grants (% GDP) ...... 35 Figure I.8. EU8 Public Expenditure Composition (% of GDP) ...... 38 Figure I.9. General government, total expenditure (% GDP) ...... 40 Figure I.10. EU8 Hidden Fiscal Deficit 2000-03 (% of GDP) ...... 41 Figure I.11. EU8 Balances at Various Levels of Government 2001-2003 ...... 42 Figure I.12. EU8 Fiscal Plans 2005-07 (% or % of GDP) ...... 43 Figure I.13. EU8 Long-Term Public Debt Projections (% of GDP) ...... 45 Figure I.14. Tax Ratio and Output Growth (for all EU8, average 1995-2003) ...... 49 Figure I.15. Theoretical Efficiency of Various Categories of Government Expenditure ...... 50 Figure II.1. Structure of Local Government Expenditures as Shares of GDP ...... 62 Figure II.2. Structure of Local Government Revenues as % GDP ...... 66 Figure III.1. Gross Enrollment Rates in Tertiary Education (EU8 & selected EU15 countries, .1990/91 – 2002/2003) ...... 82 Figure III.2. Share of University Students in Public and Private Institutions (EU8 countries, 2002) .... 86 Figure III.3. Chances of Obtaining Higher Education by Family Background (indicated by father’s occupation: the Czech Republic, Poland, Hungary, Finland, and the USA, 1998; ratios for individuals aged 20-35) ...... 89 Figure IV.1. GDP and Health Expenditures, Selected EU, OECD and Other Countries (2002 or latest available year) ...... 99 Figure IV.2. Expected Years Spent in Poor Health (Selected EU, OECD and other Countries) ...... 100 Figure IV.3. Health Expenditures in EU8 Countries, Percent of GDP, 2003 ...... 103 Figure IV.4. Public and Private Expenditures on Health in Latvia ...... 103 Figure IV.5. Structure of Health Expenditures in Poland, 1999-2003 ...... 105 Figure IV.6. Structure of Public Expenditures on Health in Hungary ...... 106 Figure IV.7. Structure of Public Expenditures on Health in Latvia ...... 106 Figure IV.8. Structure of Health Expenditures, OECD average (2001) ...... 107 Figure IV.9. Fiscal Position of the Health System, Hungary ...... 107 Figure IV.10. Fiscal Position of the Slovak Health System (percent of GDP) ...... 108 Figure IV.11. Structure of Hospital Debts in Poland ...... 108 Figure IV.13. Total Pharmaceutical Expenditure in Latvia (2003 prices) ...... 110 Figure IV.14. EU8 Actual and Projected Total Fertility Rate (TFR) ...... 114 Figure IV.15. EU8 Actual and Projected Life Expectancy at Birth ...... 114 Figure IV.16. EU8 Countries: Population over 60, 1995-2025 ...... 115 Figure IV.17. Diffusion of MRI Units and CT Scanners, Selected Countries ...... 116 Figure IV.18. Price and Volume of Pharmaceuticals, Selected Countries, 2002 ...... 119 Figure IV.19. Expenditures on Drugs in Slovakia ...... 121 Figure V.1. Comparing Public Finance to a PPP with a Long-Term Purchase Contract ...... 142
Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 7 FOREWORD
The new Central European and Baltic members of the European Union – the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia, and Slovenia, henceforth the “EU8 countries” – all face the challenge of creating greater fiscal space for needed development spending, not least on infrastructure, to accelerate the process of real income convergence with the incumbent members of the EU.1 At the same time, they are under increasing pressure to improve the quality and efficiency of public service delivery, both to strengthen the formation of human capital needed to support high and sustainable growth and to ensure broad and equitable opportunities for all citizens.
Fiscal space is constrained first of all by the need to maintain sustainable fiscal positions in the context of the Economic and Monetary Union, but also by limited absorptive capacity for additional, productive public spending. The Stability and Growth Pact requires EU members to maintain at least balanced budgets over the economic cycle, and to contain fiscal deficits below 3% of GDP and debt levels below 60% of GDP, albeit with some flexibility. These thresholds are also key conditions for the adoption of the Euro by the new member states. Finally, the Visegrad countries (the Czech Republic, Hungary, Poland, and Slovakia) need to reduce excessive fiscal deficits to meet the requirements for joining the monetary union.
At the same time, fiscal space is constrained by existing pressures on public expenditures and fiscal revenues. Social spending is already compressed in some EU8 countries, and most of the EU countries face significant spending pressures over the medium-long term related to population aging (pensions, health, etc.). Public sectors also remain relatively large given income levels in some of these countries, and tax burdens may need to be reduced further to enhance efficiency, competitiveness, and employment in the EU.
Under these conditions, fiscal space can generally only be created through rationalizing existing spending programs. While some progress has been made in most EU8 countries, there is strong evidence of remaining inefficient or even wasteful expenditures, and – although politically difficult – curtailing such spending would clearly be justified on economic grounds. Reviewing persisting subsidies to certain sectors or enterprises is the most obvious place to start, as it would result in increased fiscal space as well as reduced distortions. But the biggest potential for savings in most countries lies in the social spending area where large inefficiencies are apparent. In most countries, the scope for creating fiscal space on the revenue side is limited but could be achieved through broadening tax bases and sources.
The aim of this study is to examine more specific ways of creating fiscal space in the EU8 countries. It considers ways to enhance the efficiency of key spending programs and bring in additional private financing, while supporting the broader growth process and social equity and governance concerns. The study is not meant to be a comprehensive review of fiscal policy in the EU8 countries, but rather to address a limited number of selected, priority issues. This study is also not meant to include in-depth analysis of country specific systems and specific country-oriented policy recommendations, but rather – given the commonality of issues across the EU8 countries as well as other countries in the region – to provide a cross-country comparative perspective, convey existing knowledge and current thinking on reform issues, and distill lessons from reforms both within and outside of the region as well as best practices where such can be identified. Finally, this study is part of a larger, multi-year effort that will continue to focus on priority issues for fiscal reform in the EU8 countries.
The study was prepared in close partnership with governments and non-government experts in the EU8 countries. The priority issues were selected in consultation with Finance Ministers from the region, and government officials from the various countries – as well as local experts and consultants – were closely involved in the work through the provision of information, background analysis, and discussion of work in progress. International experts were also involved either as direct contributors or peer reviewers. Early drafts of the study were discussed at the technical level in a series of regional thematic workshops, and the final overall study was discussed at a conference in Prague hosted jointly by the Ministry of Finance of the Czech Republic and the World Bank on November 15-16, 2005. We hope that government officials from the EU8 countries can benefit from the study through the associated dissemination of knowledge and analytical
1 “In its broadest sense, fiscal space can be defined as the availability of budgetary room that allows a government to provide resources for a desired purpose without any prejudice to the sustainability of a government’s financial position” (Heller, 2005: Understanding Fiscal Space, IMF Policy Discussion Paper, March). 8 capacity building, and also that policy-makers will find the study useful as they consider various reform alternatives. We also hope that other countries in the region as well as international partner organizations such as the European Commission will find the study relevant and useful.
This study includes a review of the current fiscal performance and prospects in the EU8 countries; intergovernmental fiscal relations; higher education financing; health care financing; and fiscal risks related to public-private partnerships. The main message of the study is that much progress has been made in placing fiscal policy on a more efficient and sustainable path in the process leading to EU accession, but that there are opportunities for a stronger orientation of fiscal policy toward growth with equity. In the social spending areas, the study argues for enhancing competition within adequate regulatory and quality control frameworks; increasing the role of private financing; improving incentives on both the supply and demand sides of the market; and providing more targeted support for low-income groups. Such reforms would not only enhance efficiency and fiscal sustainability, but also provide for more equitable access to key social services and strengthen governance in major institutions. In the institutional areas, the study argues for institution building to play a major role in managing fiscal risks from PPPs and in making intergovernmental fiscal relations more stable and predictable. For PPPs, this includes measures to enhance risk analysis, disclosure, accounting and management (including medium-long term fiscal projections). For intergovernmental fiscal relations, it includes measures to ensure adequate funding of central government mandated sector specific spending as well as greater security in the funding of discretionary expenditures.
The World Bank stands ready to further support the EU8 countries in these endeavors.
Daniela Gressani Country Director, Central Europe and the Baltic Countries Europe and Central Asia Department World Bank
Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 9 CONTRIBUTORS Authors: Hana Polackova Brixi, Senior Economist, World Bank (Beijing) Nina Budina, Senior Economist, World Bank (Washington, DC) Paulina Bucon, Research Assistant, World Bank (Warsaw) Mary Canning, Lead Education Specialist, World Bank (Warsaw) Mukesh Chawla, Lead Economist, World Bank (Washington, DC) William Dillinger, Lead Public Sector Management Specialist, World Bank (Washington, DC) Martin Godfrey, Consultant, World Bank (Washington, DC) Dorota Holzer-Zelazewska, Research Analyst, World Bank (Warsaw) Timothy Irwin, Senior Economist, World Bank, (Washington, DC) Thomas Laursen, Lead Economist, World Bank (Warsaw) Emilia Skrok, Economist, World Bank (Warsaw) Peer reviewers: Nicholas Barr, Professor of Public Economics, London School of Economics, UK Paolo Belli, Senior Economist (Health), World Bank (Washington, DC) Andrew Burns, Senior Economist, World Bank (Washington, DC) Kenneth Davey, International Development Department, University of Birmingham, Emeritus Professor of Development Administration Imre Hollo, Economist, European Investment Bank Andrzej Rys, Director, JAGIELLONIAN UNIVERSITY, Centre for Innovation, Technology Transfer, and University Development, Poland Jamil Salmi, Sector Manager, World Bank (Washington, DC) Allen Schick, Professor, Maryland School of Public Affairs, USA Sergei Shatalov, Lead Country Economist, World Bank (Washington, DC) Drahomira Vaskova, Deputy Director, Ministry of Finance, Czech Republic, Max Watson, Wolfson College, Oxford, Deborah Wetzel, Sector Manager, World Bank (Washington, DC) Conference discussants and session chairmen: Jaak Aaviksoo, Rector of the University of Tartu, Professor of Optics and Spectroscopy, Estonia Lucie Antosova, Institute of Economic Studies, Charles University, Prague Vladimir Bezdek, Advisor to the Bank Board, Czech National Bank Cheryl Gray, Sector Director, Poverty Reduction and Economic Management, Europe and Central Asia region, World Bank Roger Grawe, Senior Advisor, Roma Education Fund; former Country Director for Central Europe and the Baltic States, Europe and Central Asia region, World Bank Daniela Gressani, Country Director for Central Europe and the Baltic States, Europe and Central Asia region, World Bank Stepan Jurajda, Deputy Director for Research, CERGE-EI, Prague Filip Keereman, Head of Unit, Member States V: Czech Republic, Lithuania, Poland, Slovenia, Slovakia (ECFIN. B.5), European Commission, DG Economic and Financial Affairs Thomas Laursen, Lead Economist for Central Europe and the Baltics, Europe and Central Asia region, World Bank Danny Leipziger, Vice President and Head of Poverty Reduction and Economic Management Network, World Bank Tom McCarthy, Chief Executive, Irish Management Institute, Ireland Witold Orlowski, Professor, Warsaw School of Economics; former Advisor to the President of Poland Gabor Peteri, Director of Development, LGID, Hungary Tomas Prouza, Deputy Minister of Finance, Czech Republic Andrzej Rys (see above) Ondrej Schneider, Deputy Head of the Institute for Economic Studies, Head of the Department of European Economic Integration and Economic Policy, Faculty of Social Science, Charles University, Czech Republic Paul Bernd Spahn, Macro Fiscal Advisor, ITA Regional Center, Bosnia and Herzegovina Jan Svejnar, Professor of Economics, University of Michigan, USA Sweder van Wijnbergen, Professor of Economics, Universiteit van Amsterdam, the Netherlands Drahomira Vaskova, Deputy Director, Ministry of Finance, Czech Republic Andreas Woergoetter, Head of Country Studies V Division, Economics Department, OECD Miroslav Zamecnik, Consultant, Czech Republic 10 CONFERENCE AGENDA Prague, November 15-16, 2005
November 15:
20.00-22.00 Welcome Banquet Hrzansky Palace (hosted by Finance Minister Sobotka).
November 16:
8.30 Registration
9.00 Welcome Remarks (Daniela Gressani and Tomas Prouza)
9.15 Introductory remarks (Tomas Prouza and Danny Leipziger)
9.45 Fiscal Background (Emilia Skrok and Paulina Bucon) Discussant: Jan Svejnar Discussant: Filip Keereman Chair: Thomas Laursen 10.30 Coffee
11.00 Assessing Intergovernmental Fiscal Relations (William Dillinger). Discussant: Gabor Peteri Discussant: Paul Bernd Spahn Chair: Roger Grawe
12.00 Managing Fiscal Risks of PPPs (Nina Budina) Discussant: Sweder van Wijnbergen Discussant: Vladimir Bezdek Chair: Drahomira Vaskova 13.00 Lunch
14.00 Financing Higher Education (Mary Canning). Discussant: Jaak Aaviksoo Discussant: Tom McCarthy Chair: Stepan Jurajda 15.00 Coffee
15.30 Controlling Health Expenditures (Mukesh Chawla) Discussant: Andrzej Rys Discussant: Miroslav Zemecnik Chair: Lucie Antosova
16.30 Concluding panel discussion Andreas Woergoetter Ondrej Schneider Witold Orlowski Chair: Cheryl Gray
17.30 Closing remarks (Drahomira Vaskova and Thomas Laursen). Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 11
OVERVIEW2 i. EU8 countries continue – to varying extents – to be affected by their socialist legacies in terms of their economic and social systems. Generally, the social models of these countries are characterized by relatively low employment and efficiency, with employment rates ranging from a low of 52 percent in Poland to a high of 64 percent in the Czech Republic, still below the EU15 average.3 Equity, on the other hand, varies significantly among the EU8 countries, with relative “poverty rates” below the EU15 average in the Czech Republic, Hungary, and Slovenia, but significantly higher in Slovakia, Poland, and the Baltic countries.4 Following Sapir (2005), this places the Czech Republic, Hungary, and Slovenia among the European countries with low efficiency and high equity, similar to most continental economies (including France and Belgium), while the other EU8 countries are characterized by low efficiency and low equity, similar to the Mediterranean economies (including Spain, Italy, and Greece) (Figure O1). This is in sharp contrast to the high efficiency-high equity models of the Nordic countries.5
Figure O1. Employment and Relative Poverty Risk in the EU (2001 or latest data). 94
92 CZ SE 90 HU SI FI LU Nord ) 88 AT NL %
( Cont
e BE DE t
a FR r 86 y t r EU8 e
v EU15 o P
- 84 PL LV 1 LT 82 EE IT Med ES UK Anglo 80 GR PT SK IE 78 50 55 60 65 70 75 80 Employment rate (%) Notes: Poverty rate is the “risk” of relative poverty (measured as the share of the population with income, after social transfers, below 60% of median). Data for EU15, EU8, Anglo, Cont, Nord and Med are calculated as simple averages. LV (2002); SK (2003). Source: Eurostat. ii. While, this report relies mostly on the EU countries as main comparators, this may not always be the best benchmarks. The structural public finance and labor market problems in Europe are well known, and EU8 countries may also want to look beyond this region for guidance on key economic policy issues and best practices, including to other OECD countries and fast-growing emerging markets. Nevertheless, with recent EU accession, participation in the union’s broader economic and structural policy guidelines, and EU-wide data comparability, Europe is the natural first place for comparison. iii. The ultimate objective of economic reforms in general and social and labor market reforms in particular in the EU8 countries should be to move towards the NE corner of the efficiency-equity diagram. The experience and models of European countries suggests that there is no inherent trade-off between these two objectives, at least in the longer run (Schuknecht and Tanzi, 2000).6 Further, the EU8 countries are not likely to be close to the frontier where such a trade-off may exist. The aim of the World Bank’s programmatic fiscal reform analysis in the region is to discuss how public finance and social reforms can help make a push in the right direction while creating fiscal space for additional productive, growth enhancing public spending (and potentially tax
2 Prepared by Thomas Laursen. 3 Efficiency is characterized here is providing adequate incentives to work. Employment is measured as the share of employed 15-64 year olds to total population in that age group. 4 Poverty rates measured as the share of population with incomes below 60% of the median (as a proxy for the risk of falling into poverty). 5 Boeri (2003) refers to the “fatal attraction” of Western Europe in shaping social policies in Central and Eastern Europe. 6 This contrasts with earlier work that argued for a clear – and possibly increasing – trade-off between equity and efficiency (Okun, 1975; and de Mooj and Tang 2004). 12 reforms) within tight fiscal constraints and the need to ensure fiscal sustainability. As pointed out by Sapir (op cit.), efficiency is also a necessary condition for long-term sustainability. This claim is tentatively supported by the negative relationship between employment and debt in the EU, although the correlation is not strong, there is significant variation, and all the usual caveats to such bivariate correlation analysis apply (Figure O2).
Figure O2. Employment and Debt in the EU (2004).
115 IT GR 105
BE 95
85
DP 75 o G FI
f FR 65 DE s%
ba HU PT EU15 d 55 SE t NL
GG e ES 45 PL AT SK UK 35 CZ EU8 SI IE 25 LT 15 50 55 60 65 70 75 Employment rate (%) Source: Eurostat. iv. Fiscal space in the EU8 countries is needed to increase investment in physical – notably infrastructure – and human capital to raise the countries’ long-run growth potential and thus ensure relatively rapid convergence of real incomes and living standards with those prevailing in the rest of the EU. While education levels are generally high in the EU8, the structure of employment and production suggests that strengthening the quality of human capital remains an important challenge. Similarly, while much progress has been made, infrastructure indicators continue to compare unfavorably with those of Western Europe. Fiscal space is also required to cofinance large and increasing inflows of EU regional aid funds, the absorption of which also will be contingent on stronger technical and administrative capacity of all levels of government. Further, some EU8 countries (notably Hungary and Slovenia) retain relatively large public sectors given their income levels, and in most EU8 countries the tax burden on labor remains very high, discouraging employment (Figures O3 and O4). Thus, a further lowering of the tax burden – not least on labor – is also desirable to stimulate investment and employment growth.
Figure O3. Size of Public Sector and Per Capita Income in the EU (2004). 60 SE
55 AT
BE 50 FI FR
DP SI IT EU15 45 HU PT NL s GR DE LU a of G PL CZ 40 UK
e EU8 v EE ES eue% n LV SK
GG r 35 IE
LT 30
25 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 GDP per cap ita PPS Source: Eurostat. v. EU8 countries face tight fiscal constraints over the medium term and difficult challenges in ensuring fiscal sustainability over the longer term. As members of the EU, countries are subject to the Stability and Growth Pact (SGP), and the commitment to prepare for Euro adoption entails the need to prepare convergence programs consistent with the Maastricht criteria and – where relevant – submission to the Excessive Deficit Procedure. These Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 13
Figure O4. Tax wedge and Employment Rate in the EU ( 2004). 50
BE SE 45 DE IT PL HU LV CZ EU8 40 LT SI FI SK AT NL () EE o b% 35 GR EU15
ES n FR o ar
g 30 PT dl ee
a LU UK
Txw 25
20
IE 15 50 55 60 65 70 75 Employment rate (%)
Notes: LT and SI (2003); EE (2002). Source: Eurostat. institutional constraints boil down to commitments to gradually reduce fiscal deficits to below 3 percent of GDP – and in principle to continue fiscal consolidation and achieve broadly balances fiscal positions – and maintain public debt below 60 percent of GDP. In the revised SGP, new member states that have introduced private pension pillars are given some respite in that the fiscal cost of such reforms can be phased in gradually over the period 2005-10. While all EU8 countries are below the debt threshold (although Hungary is teetering on the edge of it), only the Baltic countries and Slovenia meet the deficit target. The Visegrad countries (the Czech Republic, Hungary, Poland, and Slovakia) plan to achieve this only during 2006-08, and uncertainty remains about the broad political commitment and ability to adhere to these timetables in some of these countries. Over the longer term, all countries face significant fiscal pressures from population aging that in most cases could render public finances unsustainable in the absence of further reform of social insurance systems.7 vi. Most countries in the region have significant potential for creating the required fiscal space. First, several countries – notably the ones that retain large public sectors – have room for reducing wasteful spending on direct or indirect subsidies, rationalizing social transfer programs, enhancing the efficiency of health care and education systems, and streamlining public administration. Second, most countries can further strengthen tax administration, broaden tax bases and sources, and reduce tax expenditures. Third, some countries can advance further on the privatization agenda, raising resources not only for financing pension and other reforms that reduce implicit government liabilities, but also reducing the need for ongoing fiscal support in various forms and paving the way for improved efficiency and profitability of the given enterprises and thus enhancing the future revenue potential. There is also scope for increased use of public- private partnerships where there are good reasons to believe that the private sector is more efficient than the public sector in creating and operating infrastructure and public services. On the other hand, shifting spending off the budget will not by itself create real fiscal space. Fourth, in several countries there is scope for strengthening public finance management (procurement, debt management, etc.). vii. The Lisbon strategy calls for the emphasis of public finances to be broadened from its focus on stability to include the contribution to sustainable growth, full employment, social cohesion, and competitiveness. The quality of fiscal policy remains an important concern in most EU8 countries, as indeed in many incumbent EU member countries. While the impact of fiscal policy on short-term growth has been extensively studied and discussed, surprisingly little remains understood about the role of public finances in promoting longer term growth.8 Modern endogenous growth theory clearly identifies a potentially important role through potential effects on factor accumulation (both human and physical capital), the provision of public goods, addressing externalities, etc. (see e.g. Barro, 1990). Most models distinguish between productive and non-productive
7 Countries with strong fiscal positions and low public debt levels as well as those that have implemented comprehensive pension reforms (including establishing a close link between contributions and benefits in their PAYG systems) are generally in a stronger position to deal with adverse demographics. 8 Recent literature has focused on conditions for expansionary fiscal contractions (see e.g. Giavazzi and Pagano, 1990; and Alesina and Perotti, 1995). 14 spending and distortionary and non-distortionary taxes, and several of these postulate a non-linear, hump- shaped relationship between aggregate spending or taxation and economic growth. Nevertheless, such classifications are ambiguous, and the empirical evidence is similarly non-conclusive (Volume II Annex 2). This being said, it is hardly controversial that public investment in key infrastructure and education – at least up to a certain point – are conducive or even critical to sustained growth, while high direct taxes that affect investment and employment have a negative impact. Flat tax systems have become popular in transition economies, but the empirical evidence on their role in promoting growth remains mixed and tentative, revenue effects are uncertain, and they may reduce the role of public finances in income redistribution. viii. The role of fiscal decentralization in promoting efficiency and growth is, on the other hand, controversial. While subnational fiscal autonomy may promote the effectiveness and efficiency of government interventions through better knowledge of local market conditions, there is also a risk of regulatory capture by local interests and expenditure indiscipline. Institutions undoubtedly matter a great deal – if these are not strong, privatization of key municipal enterprises and service providers may often be a better alternative. Some studies do find a positive association between decentralization and local government revenue autonomy on the one hand, and output growth on the other, but again the empirical evidence is far from conclusive.9 This is hardly surprising as for example the decentralization of education in Hungary has had questionable results in terms of education outcomes. However, it is interesting to note that fiscal decentralization in the EU8 appears to have been weakly associated with faster output growth, although there is obviously significant variation and growth clearly depends on many other factors (Figure O5).
Figure O5. Fiscal Decentralization and Real GDP Growth in the EU8 1995-2003.
3 10 0 2) 50
eal LV
9 8 o r
1- EE 9 ( ate f
tr 6 n% LT CU i 4 ggrowh ainL SK SI HU PL CZ
vera e 2
er capit EU15 nala DP p An u 0 G 15 2 0 2 5 3 0 3 5 Expenditure of local government level as a s hare of general government expenditure in 2003 Note: The share of local government in total spending was broadly unchanged during 1995-2003. Source: Eurostat; and staff calculations. ix. The World Bank’s analytical work on fiscal reform issues in the EU8 is of a multi-year, programmatic nature, with a focus on the following five main areas: (i) education; (ii) health; (iii) social insurance; (iv) public administration reform; and (v) fiscal risks. The work adopts a regional, cross-country approach reflecting the fact that many of the issues and reform considerations are similar across most countries in the region and that directions for reform may have broad applicability. At the same time, the progress in different areas of reform has been uneven across the region, and there may be valuable lessons to be learned from the reforms. The studies in this volume focus on two priority areas for public finance reform: health care financing and higher education financing. In both areas, there has been some, albeit uneven, progress on reform in recent years, but much remains to be done in most countries to promote service quality, efficiency, equitable access, financial sustainability, and governance. EU membership and enhanced mobility of people across borders in the union is already affecting both sectors, including through upward pressure on salaries, and more is likely to come. There is little doubt that enhanced competition, increased reliance on user payments, and more targeted subsidies is in the cards. In both health care and education, the system of intergovernmental fiscal relations plays an important role, and the volume sets the stage for the subsequent discussion by reviewing the current situation and key issues in this area. Along a somewhat different dimension, the study also includes a discussion of the fiscal risks of public-private partnerships (PPPs) and the conditions under
9 See e.g. Davoodi and Zou (1998). Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 15 which such arrangements may help to create real fiscal space.10 The regional focus of and priorities for the Bank’s analytical support have been widely endorsed by Finance Ministers in the region. Fiscal challenges for the EU8 countries (Chapter I) x. Fiscal trends in recent years have varied considerably among the EU8 countries. Two groups of countries emerge within the EU8 on the fiscal scene: those with relatively strong fiscal positions, modest debt, and small governments (the Baltic countries, and to some extent Slovakia); and those with relatively weak fiscal positions, sizeable debt, and large governments (Hungary, Poland, and the Czech Republic). Slovenia is a special case, with strong public finances but a large government. The composition of revenues and expenditures, meanwhile, remains an important issue in all EU8 countries, although the particular problems differ from country to country. Chapter I by Emilia Skrok and Paulina Bucon examines recent fiscal developments, assesses medium- long term fiscal prospects, and discusses the quality of fiscal policy in the EU8. xi. Following the temporary widening of fiscal deficits in the wake of the Asian and Russian crises in most countries, some countries pursued ambitious and sustained fiscal consolidation strategies while others allowed deficits to remain high or even widen further. In particular, the Baltic countries, Slovenia, and Slovakia all undertook a sustained adjustment effort, with fiscal deficits now around or below three percent of GDP (and even surplus in Estonia). Debt levels are low in the Baltic countries, moderate in Slovenia, and reaching comfortable levels in Slovakia.11 Fiscal policy has been more erratic in the other Visegrad countries.12 The Czech Republic managed to reverse a sharp widening of the fiscal deficit in 2001-02, but new pressures are building fast. Poland pursued an expansionary fiscal policy in the period 2001-04, especially in the last couple of years where output growth recovered strongly, but took steps to regain control in 2005. Hungary has to a large extent lost control over its fiscal policy, and debt levels are hovering around the critical 60 percent of GDP limit. xii. Fiscal consolidation efforts have been supported by strong output growth and in some countries expenditure reform and/or discipline while tax reforms have tended to lower revenues. Several countries have been pursuing tax reforms aimed at lowering the overall tax burden, and in most EU8 countries general government revenues as a share of GDP are now significantly lower than in the mid-1990s and than in the EU15. Slovakia has been a frontrunner in these efforts (starting from a relatively high level), reducing the tax burden by more than 10 percentage points of GDP. Cuts in corporate and personal income taxes have tended to lower the share of direct to total taxes, while the reliance on social security contributions and indirect taxes has increased.13 Overall, labor taxes remain relatively high in most EU8 countries. xiii. Expenditure developments have generally mirrored revenue developments, albeit with cyclical conditions and political cycles affecting short-term movements in some countries. Slovakia has been the only EU8 country to undertake a comprehensive restructuring of its social spending programs, with more piecemeal reforms in other countries that have tended to rely more on various administrative measures (notably Hungary). EU8 countries have generally also benefited from a decline in interest rates. On the whole, the structure of spending has not changed much over the past decade or so, although some countries have managed to reduce subsidies and public consumption to the benefit of higher investment. Increasing subsidies in the Czech Republic (and recently also Estonia and Lithuania), higher social spending and public consumption from already high levels in Hungary, and cut backs in social spending from low levels in Latvia and Lithuania are a source of some concern. Public investment is particularly low in Latvia. xiv. Intergovernmental fiscal relations have been influenced by an ongoing process of decentralization, with a tendency toward increasing budgetary autonomy of lower levels of government. The size of local governments relative to total general government in the EU8 is around 20-25 percent, broadly similar to EU15. Local governments in the EU8 have generally not been a source of fiscal deficits in recent years reflecting tight borrowing constraints in most countries, although they may have been behind some of the “hidden” fiscal deficits (which have amounted to an average of more than one percent of GDP annually in recent years).
10 While PPPs can play a role in the provision of healthcare and education services, this is not the focus in this report. 11 Debt developments have generally been influenced favorably by strong output growth, lower interest rates, and currency appreciation. 12 Fiscal outcomes in the Visegrad countries have generally fallen well short of targets agreed with the EU in the context of pre- accession economic programs and in some countries post-accession convergence programs. 13 More recently, changes in indirect taxes have been influenced by EU accession. 16 xv. Medium-term fiscal plans are laid out in countries’ Euro adoption convergence programs with the EU. The Baltic countries and Slovenia aim to maintain fiscal deficits below three percent of GDP and adopt the Euro from 2007-08.14 Slovakia is targeting to achieve the fiscal deficit threshold in 2006, and plans to adopt the Euro from 2009. Poland and Hungary were planning to achieve the fiscal target in 2007, but a new government taking office in Poland and recurring fiscal slippages in Hungary raise uncertainty about these commitments and Euro adoption may well be pushed back to beyond 2010 in both countries. Despite major outstanding reforms, the Czech Republic looks to be on target to reduce its fiscal deficit to below the critical level in 2008, with Euro adoption feasible in 2010-11. Existing fiscal plans would be consistent with maintaining public debt ratios below the required 60 percent of GDP. xvi. Looking further ahead, public finances – notably pension and health spending – in all EU8 countries will come under increasing pressure from population aging. While some countries (notably Poland and Estonia) have positioned themselves well through comprehensive pension reforms, others face rapidly rising and unsustainable debt levels in the absence of further reforms – not least in case of failure to comply with medium-term fiscal programs. xvii. The quality of fiscal policy remains an important concern in most EU8 countries. The Lisbon strategy calls for the emphasis of public finances to be broadened from its focus on stability to include the contribution to sustainable growth, full employment, social cohesion, and competitiveness. In this context, the quality of fiscal policy refers to the structure and efficiency of taxes and spending. The quality of fiscal adjustments also has important bearings on the success and sustainability of such adjustments. High quality fiscal adjustments – including expansionary episodes of fiscal adjustment – have been found to rely on expenditure cuts rather than tax increases, and on tackling those expenditures that are politically sensitive such as transfers, subsidies, and wages. The EU8 countries have undertaken several episodes of fiscal adjustment since the late 1990s, with about one-half of these episodes dominated by expenditure contraction.15 Fiscal contractions have relied on cuts in investment (Czech Republic, 1999 and 2004; Estonia, 2003; and Latvia, 2003), social spending (Lithuania, 2002; Latvia, 2001 and 2002; and Slovakia, 2003), and public consumption. xviii. The extent to which fiscal adjustments and public finances in the EU8 have been and are conducive to growth is difficult to assess. The literature remains somewhat inconclusive as far as the impact of size of government and structure of public finances on growth is concerned, although some consensus seems to be emerging that there may be a hump-shape on the former relationship – too little and too much government is bad, that certain spending areas tend to be more productive than others (notably investment in critical infrastructure and human capital is better than subsidies and excessive government employment), and that some taxes are more distortionary than others (indirect taxes less distortionary than direct taxes). Much has to do with the efficiency with which government collect and spend resources, both in a technical and economic sense. The tentative findings in Chapter I suggest that in the EU8, higher tax burdens do tend to be associated with weaker growth performance, and that the higher the share of direct taxes and social security contributions in total taxes, the more harmful is the tax system. Findings are less conclusive regarding expenditure composition, although investment and social benefits tended to be positively and negatively, respectively, related to growth. On the whole, EU8 public finance reforms in recent years appear to have been in the right direction, although much remains to be done. Intergovernmental fiscal relations (Chapter II) xix. Overall, the EU8 countries have made remarkable progress in reforming their systems of intergovernmental fiscal relations, and the eight new EU countries now have much more in common with the longer-standing members of the Union than they do with the Soviet-era systems they inherited. Subnational governments are an important part of the public sector in the EU8 countries, accounting for about one-quarter of total government spending. They provide basic public services in both the social sectors (education, health, and social assistance) and in infrastructure (water supply, sewerage and transport). How well they perform these functions has an important influence on the quality of both human and physical capital in these countries, and therefore on the prospects for economic growth. It also has equity implications, as the social services performed by municipal governments have important distributional effects. Further, subnational government operations may affect macroeconomic stability
14 Estonia, Lithuania, and Slovenia joined the ERM2 in July 2004, with Latvia following suit in April 2005. The first three countries target euro adoption for January 1, 2007, and Latvia one year later. Recently, these plans in Latvia and Estonia have been put into question due to increases in inflation. 15 However, since 2001, only Slovakia has implemented expenditure dominated fiscal adjustments. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 17
Box 01. Objectives of Intergovernmental Finance Systems What are the objectives of a system of intergovernmental finance? According to conventional economic theory, the primary aim should be to promote efficiency in the allocation of resources while the central government would be responsible for stabilization and income distribution. Theory argues that if the benefits of particular services are largely confined to local jurisdictions, welfare gains can be achieved by permitting the level and mix of such services to vary according to local preferences. Local consumers, confronted with the costs of alternative levels of service can express their preferences by voting for rival political candidates (i.e., voting with their hands) or moving to other jurisdictions (voting with their feet). In this respect, local politics can approximate the efficiencies of a market in the allocation of local public services by “pricing” municipal services and relying on the local political process and household mobility to clear the market. On this basis, a successful system of intergovernmental relations would be one in which each level of government (1) performs only functions whose benefits are confined to its boundaries and (2) derives its revenues from sources (such as local taxes and fees) that confront its residents with the costs of these functions. In practice, there are a number of limitations to this model. First, the benefits of public services can be difficult to define in geographical terms. While most of the benefits of spending on public health, for example, may be confined to a local jurisdiction, failure to control infectious diseases can have national or international implications. Second, spending on public services – particularly education and health – can be important instruments of distributional policy. Where these functions are assigned to local governments, their distributional implications cannot be ignored. Third, high levels of local taxation or widespread deficit spending could weaken the central government’s control over fiscal policy. In practice, the intergovernmental fiscal systems of the EU8 countries must be held to somewhat different standard: (i) Transparency and predictability: The principal challenge confronting the EU8 countries was to devise a system for financing local government that would be perceived as fair and stable. The emergence of multi-party democracy in the EU8 required transparency in order to allay charges of partisan favoritism and reduce the transaction costs that would result from continued reliance on bargaining. Predictability was required to provide local governments with a stable basis for budget planning; (ii) Equity: Because local governments assumed responsibility for social services, the new system had to ensure that such services would be available to people living in poor jurisdictions; (iii) Macroeconomic control: The system had to ensure that local government fiscal behavior would not threaten macroeconomic stability. Central governments needed sufficient control over aggregate local taxing to conduct fiscal policy and sufficient control over local borrowing to ensure that aggregate deficit targets were achievable; and (iv) Effectiveness: Finally, the systems needed to encourage effectiveness in the production of services under local control. The EU8 countries inherited over-scaled schools and hospitals and inefficiently run public utilities. Incentives for rationing services therefore had to be incorporated in the new system of intergovernmental finance. as independent fiscal actions by local governments could undermine central control over fiscal policy. Chapter II by William Dillinger takes stock of developments in intergovernmental fiscal relations and discusses current challenges in this area in the EU8. xx. Organizational reforms generally took the form of scaling back the responsibilities of deconcentrated units of central administration while creating autonomous units of self-government and, in the larger EU8 countries, new, consolidated regional governments. Privatization had two important implications for local governments. First, it eliminated important providers of public services. Second, it drastically altered the revenue base. Profits (where such were generated) from local-government-owned enterprises were no longer a source of revenue, and as central governments revised the tax structure to capture revenues from the growing private sector, the fundamental basis for tax sharing had to be altered. Functional decentralization required an increase in local government revenues. Privatization changed the nature of the tax base, and changed political conditions increased pressure for local fiscal autonomy while raising the risk that local governments would use that autonomy to thwart central government fiscal and equity objectives. xxi. The new local government legislations assigned an extensive list of functions to the newly independent municipalities. Education is the largest single item of local government expenditure, with local governments (including intermediate levels of elected government) accounting for about two-thirds of total public sector spending on education in most EU8 countries. The extent of municipal management control over education, however, is limited, owing to legal constraints which set out the framework for teachers’ salaries, teaching loads, and rules for recruitment and promotion. Local governments play a major role in managing healthcare in the EU8, but only a minor role in its financing. All these countries now finance the majority of healthcare costs through compulsory insurance schemes. In most of the countries, health insurance funds (HIFs) provide 18 funding directly to healthcare providers (primary physicians and secondary and tertiary healthcare facilities), and subnational governments thus account for only a small share of total public sector health spending in most countries. In all EU8 countries, central governments provide the vast majority of social protection, including old age and disability pensions, sickness benefits, and short term unemployment benefits.16 Local governments, nevertheless, are responsible for a variety of assistance to other vulnerable segments of the population, including the long-term unemployed and their dependents. They have varying degrees of discretion in determining who is eligible for such benefits and what form they should take. xxii. With the privatization of the housing stock, municipal responsibilities for housing management and maintenance have declined sharply in the EU8 countries. Municipalities have, on the other hand, assumed responsibility for water supply, sewage, district heating, solid waste management, and third tier roads, although these do not necessarily account for a large proportion of budgetary expenditure.17 Water, sewage, district heating and solid waste management are generally organized as separate, tariff-supported, enterprises. Poor standards of original design and construction, high energy requirements, changing patterns of demand, and long periods of deferred maintenance have resulted in deteriorating and obsolete infrastructure networks. xxiii. The revenues of local governments consist of discretionary funds, largely revenue sharing and funding for specific functions, principally education and health.18 Revenue sharing generally relates to the personal income tax (PIT) – in all EU8 countries, the shared taxes are administered by the central government at nationally uniform rates, and fixed shares of the revenues are then assigned to local governments. The distribution mechanisms for the PIT include a high degree of equalization. None of the EU8 countries assign a major unrestricted – in the sense that municipalities have the authority to set the rate and retain the revenue – tax base to local governments. The most widespread form of local taxation is the property tax, but local discretion over tax rates varies among countries and even in those that have moved to marked-based valuation, the property tax generates very little revenue. xxiv. All the EU8 countries have supplementary arrangements for financing education and health. The post-Soviet-era systems of education financing reflect an ongoing tension between ensuring a basic level of education financing in all jurisdictions, regardless of the strength of local tax bases, and encouraging efficiency in the use of education funding. All the EU8 countries finance the largest component of education – teachers’ salaries – through some form of intergovernmental transfer. A few countries rely solely on central government administrative procedures to ration the level of local spending on education, while most distribute funding for education on a per pupil (capitation) basis and allow local governments considerable discretion over how these funds are used. The latter approach ensures a minimum level of education financing in all jurisdictions while allowing the central government to ration the level of such spending through its control over the per-capita amount. In particular, it forces rationalization in jurisdictions with under-enrolled schools. However, the success of capitation based financing is challenged by the inability of local governments to dismiss staff as well as the unwillingness or inability of central governments to increase the level of capitation transfers to reflect centrally-mandated increases in costs, notably wages. xxv. The EU8 countries have also attempted to use the health financing system as a tool to achieve efficiency. All have created HIFs to finance the majority of primary, secondary, and tertiary healthcare costs. These funds are financed from payroll taxes, supplemented by central government payments on behalf of certain categories of non-contributors (pensioners, workers on maternity or sick leave, and unemployed). Health funds were generally created to protect healthcare spending from the vicissitudes of the annual government budget process (although funding remains subject to fluctuations in the payroll tax base and the rate of the payroll tax). In principle, they also function as a ceiling on healthcare spending: with few exceptions, spending on primary, secondary, and tertiary healthcare is not allowed to exceed the revenues of the HIF. However, in practice hard budget constraints are often breached through arrears and extraordinary budget support. Attempts to use fiscal instruments to ration the supply of health services on a more micro level have had mixed results. One
16 Pension systems have been partially privatized in several countries in the region. 17 They do, nevertheless, constitute an important source of contingent liabilities (as these enterprises often have high levels of arrears) and a claim on scarce investment resources. 18 In most of the EU8 countries, the specific purpose of this funding is to cover the full costs of the function to which it is assigned. Poland and Hungary are exceptions, with the level of funding for education determined independently of the costs of this function in the expectation that any shortfall will be funded from discretionary revenues. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 19 of the major sources of inefficiency in the Soviet model of healthcare was its excessive use of secondary and tertiary facilities. To varying degrees the EU8 countries have addressed this issue by introducing “family medici ne” models of primary care, but performance incentives have proven incompatible with cost containment due to the tendency of providers to respond to fee-for-service forms of compensation by increasing the volume of services they provide. Rationing secondary and tertiary care has been even more difficult. xxvi. EU8 countries have also strengthened controls on local government borrowing. Following liberal access of local governments to capital markets in the early 1990s and several instances of default, the majority of EU8 countries have introduced quantitative ceilings on municipal borrowing and debt service. Several of the EU8 countries also impose case by case administrative controls on local borrowing and restrict certain forms of borrowing. Further, several countries impose penalties on municipalities that fail to meet their debt obligations, and a few have introduced formal municipal bankruptcy laws (although the use of these measures has been limited so far). Central government bailouts remain a possibility in most countries. xxvii. Chapter II argues that the EU8 countries have made substantial progress in the reform of intergovernmental relations as measured against the criteria outlined above. Most countries have come a long way on transparency and predictability, although Hungary remains a source of concern. Countries in the region also manage to achieve a high degree of redistribution in the financing of social services, perhaps even too much: while there is a strong case for equalizing spending on social assistance, the case for equalizing spending on other functions financed from discretionary revenues – road maintenance, community amenities, or general administration – is less obvious. Local government finances have not been a source of fiscal and macroeconomic instability, reflecting limited fiscal autonomy and strict borrowing constraints in most countries. On the other hand, attempts to use the system of intergovernmental relations to encourage greater efficiency in the production of public services have been less successful. While the majority of the EU8 countries now finance education on a capitation basis, this has not been sufficient to prompt the closure of under-enrolled classrooms or schools. In the health sector, efforts to encourage primary providers to increase the volume of services they provide have been thwarted by over-billing, and efforts to ration secondary and tertiary care using variants of the German points system have run aground for similar reasons. xxviii. Over the longer term, Dillinger points out that sustainability of EU8 intergovernmental fiscal arrangements will depend on the ability of central governments to keep their agreements with local governments. In monopolizing the major sources of tax revenues, central governments have arrogated to themselves the power to decide how large a share of total public sector revenues local governments should receive, and how they should be spent. In the process, they have made themselves vulnerable to the claim that the level of funding which they provide to local governments is insufficient. In the case of sector specific financing, the implication is that central government mandated increases in costs of service provision must be accompanied by increases in the level of financing. In the case of funding discretionary expenditures, the main issue is risk related to the annual negotiations between the central and local governments. Experience from Western Europe suggests that countries might deal with these issues through either increasingly sophisticated estimates of the costs of providing discretionary – in effect treating these as if they were mandatory (England is a good example of this approach), or allow the local governments to decide for themselves on the appropriate level of spending on the basis of their constituents’ willingness to pay. The latter approach would require the assignment of a broad based tax such as the PIT to the local level. Some Scandinavian countries have gone this way (and is also cited in the European Charter of Local Self-Government and actively promoted by the Council of Europe), but it appears to have few supporters among the EU8. At present, central governments prefer to retain near-total control over tax policy, and local governments prefer to avoid the onus of taxation. Financing higher education (Chapter III) xxix. The EU8 countries have enthusiastically embraced mass higher education, and are rapidly closing the gap between themselves and the EU15 in enrollment rates and the incidence of higher education in the population of working age. This expansion is helping these countries to insert themselves into the global knowledge economy, as a high proportion of their graduates get jobs in the knowledge-intensive services sector. At the same time, resources per student are declining, raising concerns about the quality of higher education. Also, the emergence of private higher education institutions and introduction of fees for student who do not gain regular admission, have resulted in a dual-track system that is affecting the equity of access to higher education. Chapter III by Mary Canning, Martin Godfrey, and Dorota Holzer-Zelazewska reviews the experience with a variety of financing mechanisms in the EU8 countries and seeks to develop some 20 useful policy options to “level the playing field” for countries contemplating further reforms, which would include the introduction of variable fees, needs-based grants and loans to increase private financial flows into higher education while facilitating equal access across the board. xxx. Average enrollment in higher education in the EU8 countries is now equivalent to well over one-half of the relevant age group, and the annual number of tertiary graduates (per 1000 people) now exceeds that of the EU15 (although, on average, the proportion of the 25-64 population with higher education remains larger in the latter group of countries). With the coming of a market economy and substantially greater opportunities for university graduates, both in local private sectors and now potentially elsewhere in the EU, the private benefits from university education have increased considerably both in terms of higher income and lower rates of unemployment. The earnings premium for higher education in the labor market is particularly high in countries where higher education still has some scarcity value. Unemployment rates (similar on average in both the EU8 and EU15) vary between countries but are invariably lower for higher education graduates than for those with lower levels of education. Overall, the private rate of return to higher education is still high in the EU8 and, given the pattern of subsidization, likely to be higher than the social rate of return. xxxi. Most EU8 countries have managed to protect the percentage of GDP that is devoted to public expenditure on education, and the spending on higher education is comparable to the EU15 (around one percent of GDP). However, EU8 countries spend a much higher proportion of GDP per head per student in public tertiary institutions than EU15 countries, and a much higher multiple of expenditure per primary school student. The number of Euros spent per tertiary student, adjusted for differences in purchasing power, is less than one-half of the level in the EU15. Expenditure per head seems to vary inversely with enrollment rates, which suggests at least the possibility of a trade-off between the accessibility and quality of higher education from the student’s point of view. While such cross-country comparisons between countries with different salary levels, even when adjusted for purchasing-power differences, are fraught with problems, there is little doubt that an increase in unit expenditure will be needed.19 xxxii. The small role of fees in most EU8 countries does not mean that parents and students avoid paying for higher education. Many countries have instituted a dual system, maintaining fee-free higher education for regularly admitted state-supported students, while adding a special fee-paying track for those who fail to gain such admission. Institutions have also imposed and increased user charges for formerly heavily- subsidized board and residence facilities, which are borne by students or their parents, as are the costs of living for those who do not make use of such facilities. Also, although a significant proportion of educational budgets remains devoted to stipends, maintenance grants or other types of financial aid to students, they have been eroded by inflation to virtually nominal levels in many EU8 countries.The dual-track system means that there are two classes of students – those whose education expenses are nominal, and those who pay fees equivalent to several thousands of dollars. Those who obtain fee-free, state-subsidized places are disproportionately from privileged families (which have contributed to their academic success); poorer students, who are less successful in entrance examinations and cannot afford the alternative fee-paying track, are generally excluded from higher education. In order to facilitate entry to higher education for students who need financial support to pay fees and/or maintenance costs, several EU8 countries have instituted loan systems. xxxiii. The trend towards increased participation in higher education can probably not be combined with the improvement in quality and relevance that is needed in order to be competitive within the EU and to increase participation in the global knowledge economy, unless the amount of money available to institutions increases. At the same time, government budgets are severely constrained, and it is difficult to justify an increase in public allocations to higher education which yields high private rates of return to people who are disproportionately from privileged backgrounds. Also, the increasing internationalization of higher education and cross-border labor mobility from the new toward the old EU member countries reduces assurances that university graduates will eventually pay for their education through income taxes. Thus, it is almost inevitable that average fees in tertiary institutions would have to be increased to a level which covers a significant proportion of total costs. The level at which tuition fees are set is clearly not only
19 Tertiary institutions in the EU8 spend a lower percentage of their budgets on research and development than do their counterparts in the EU15, but even if this is taken into account, the contrast in unit educational spending remains. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 21 an economic but also a political issue. There is a strong case, on the grounds of revenue maximization, efficiency, autonomy, and equity, for making such fees variable, rather than fixed and uniform. xxxiv. The extension of fees to all students and an increase in their average level would imply a need for wide-ranging student loan systems. The authors argue in favor of income-contingent loans carrying “unsubsidized” interest rates.20 Repayments would be income contingent, calculated as a given percentage of the borrower’s earnings until the loan has been repaid. Those who take out loans effectively get their higher education free at the point of use, repaying only later if and when they can afford to do so. Thus, even the most risk-averse students (often assumed to come disproportionately from low- income backgrounds), may be willing to borrow in order to finance their studies. Several countries have introduced student loan schemes, but with varying success. The authors note that there are several pre- conditions for effective student loan schemes, including ways to keep track of people’s movements and systems of withholding at the point of wage and salary payment. Further, to the extent private banks are involved, there would need to be a system of clear and credible government guarantees and proper fiscal accounting for these. Student loans could be supplemented with scholarships and grants for those who truly need it.21 However, income contingent loans (and other subsidies) raise difficult issues of both feasibility and equity. Means-testing is complicated by the fact affluence may depend not only on current income, and further, there is no assurance that wealthy parents would pay for their children’s education. Means-tested loans and grants to higher education students are, at best, a blunt instrument and need to be backed up by other measures to promote equity in access. xxxv. Such reforms would allow the inequitable dual-track system to give way to a single-track system, under which all students pay tuition fees, and to a distribution of state subsidies based primarily on need, rather than on academic ability. Thus, a system of deferred tuition fees (i.e. combined with a well-designed system of student loans), supported by a system of scholarships or grants targeted to those who need them, is an essential part of any strategy for expanding quantity, improving quality and achieving equitable access. Students should be eligible for government subsidies whether they are attending a public or a private institution. At the same time, it is the role of the government to set and monitor quality standards for all tertiary institutions, regardless of ownership. The combination of variable fees, needs-based grants, and loans would also help to increase the current and future labor market relevance of the specializations chosen by students. xxxvi. The chapter also stresses that further reforms in the criteria for determining the amount of public money to be allocated to each tertiary institution would help increase transparency of funding and efficiency. Having started with fairly simple “money-follows-the-student” formulae, including competitive and experimental funding mechanisms to encourage innovation and research, would be useful in meeting the required policy goals. Further governance reforms, particularly in institutional management and procedures for appointing leaders of autonomous tertiary institutions, are likely to be needed to ensure that lump sums allocated through such mechanisms are used in a way that is sensitive to the public interest. While deregulation and increased autonomy of universities have made higher education more sensitive to student choice and labor market developments, closer links between the university and the private sector are desirable, also to attract private funds through endowments and other mechanisms. xxxvii. Decreased reliance on government funding and increased reliance on price incentives would not mean the elimination of a role for governments in relation to higher education. They would still be an important source of funds, organize and oversee student loan schemes, and be responsible for the promotion of equitable access. They would also have to ensure that quality assurance systems are in place, and would be able to design and use formula-funding schemes to achieve national objectives that go beyond those of the immediate market and to modify excessive competition between institutions. What is needed, rather than detailed interference in academic processes, is a combination of standard-setting and financing systems designed to ensure high-quality outcomes which would be communicated to stakeholders through accurate, impartial and easily available data. The rest could be left to higher education institutions, autonomous but accountable in their governance arrangements.
20 Interest rate subsidies add a lot to the costs of a loan scheme, divert funding from quality improvement to student support, and are deeply regressive (mainly benefiting successful professionals in mid-career whose loan repayments are switched off early because of the subsidy). 21 Institutions and governments could offer a few competitive scholarships to the strongest candidates to provide an incentive for competition in the qualifying process. 22 Controlling health care expenditures (Chapter IV) xxxviii. Following the transition from central planning toward market-based economies, the EU8 countries introduced a number of reforms in the finance, management and organization of the health sector. Reforms removed the state monopoly on healthcare by privatization and decentralization, and started the process of establishing an economic relationship between the healthcare provider and the patient through changes in healthcare financing. Almost all countries adopted the Bismarckian social health insurance model, and in all cases one or more Social Health Insurance Agencies started operating independently of the state budget. In most countries, subsequent reforms also addressed the pharmaceutical sector, and brought about changes in the health financing structure in an effort to address the manifold problems brought about by the insurance system. However, reforms have generally been less successful in securing sustainability of health care financing, improving efficiency, enhancing equity in healthcare financing and delivery, and managing the quality of health services. Chapter IV by Mukesh Chawla and Marzena Kulis takes stock of recent trends in health expenditure and discusses directions for reform consistent with the objectives of stabilizing the fiscal situation in these countries without adversely affecting the production, delivery and utilization of health services. xxxix. EU8 countries have generally been successful in safeguarding resources for the health sector and a reasonable health status of their people. Total spending on health in the EU8 countries is at levels roughly commensurate with per capita GDP in these countries, and is not out of line with other European countries and high-performing middle income countries. In some EU8 countries like Estonia and Latvia, however, health spending is on the low side relative to per capita GDP. The health status of the people in the EU8 countries is at levels commensurate with their levels of health spending and income, but is generally poor compared to the EU15. The difference between the EU15 and EU8 average life expectancy has decreased in recent years, but most EU8 countries have some five years shorter life expectancy than the EU15 average. xl. Health expenditures in the EU8 countries are mainly financed from public sources. On average, public financing accounts for about three-fourths of total health spending in EU8 countries, which is comparable to the EU15. Out-of-pocket payments in the health sector account for the bulk of the residual financing and have been increasing in recent years. Social health insurance is the dominant form of public financing of the health system in EU8 countries, although many also transfer funds from the state budget to the social insurance system. Most countries have a single agent who purchases health services on behalf of the insured. The public financing system in all EU8 countries provides generous coverage of health services, in most cases with little or no financial participation from the patient at the point of service. xli. Health expenditures in almost all EU8 countries are dominated by inpatient care and pharmaceuticals, which account for roughly 40 percent and 30 percent of total health expenditures, respectively. Primary and specialist outpatient care, long-term care, and administrative expenses comprise the balance. The general trend over time has been that of rising expenditures on pharmaceuticals, falling expenditures on inpatient care and increasing expenditures on outpatient care. This structure of spending is not very different from the OECD countries. The relatively fast growth in pharmaceutical costs in both groups of countries reflects higher volumes and the entry of new, more expensive drugs, rather than general price increases. xlii. In addition, excessive hospital infrastructure is a major drain on scarce health sector resources, and salaries and the cost of new technology are also rising fast. The EU8 countries inherited a disproportionately excessive number of hospitals and hospital beds, and while almost all countries have taken many bold steps to reduce the number of hospitals and acute-care hospital beds, but the number of beds per 100,000 inhabitants is still 30 percent higher than the average for EU15 countries. Salaries in the health sector – which account for more than 60% of health expenditures in these countries – are rising faster than average salaries. This reflects greater availability of outside opportunities following EU accession, and also the push to bring the ratio of salaries in the health sector to average salaries in the economy in line with EU15 proportions.22 Further, widespread and extensive use of new and expensive medical technology has made it a major cost- driver in the health systems in many EU15 countries, and this trend is likely to continue.23
22 The ratio of physician salaries to the national average is significantly lower than two (the average EU15 level) in many EU8 countries, but are likely to rise toward the EU15 average as medical professionals from the new member states are in great demand there. 23 The tax-funded health systems of Spain and the UK have the lowest density of MRI units and CT Scanners, and also report the lowest hospital capacities and inpatient expenditure shares, as well as the lowest total health expenditures. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 23 xliii. A defining feature of the health sector in almost all the EU8 countries is the widespread and growing indebtedness.24 Buoyed by generous benefit packages (offering ineffective and non-essential services), extensive infrastructures, and pressured in recent years by increasing pharmaceutical costs and higher salaries, expenditures on health in country after country have consistently surpassed the available resources, leaving huge unpaid bills for services already delivered. This situation is particularly severe in the Visegrad countries, although some of the Baltic countries are also beginning to feel the pressures. Regions with an excessive concentration of public hospitals and hospital beds expectedly also generate the largest debts in the health sector. Hospital contracts with health insurance companies do not cover the full costs of maintaining the large hospital infrastructure and associated fixed costs, and the resources allocated by the central and local governments (who own most public hospitals in most EU8 countries) are not sufficient to pay for all the expenses. In addition, public hospitals tend to provide services irrespective of the value of contracts concluded with the sickness funds, and bear costs that are refunded neither from health insurance nor from the state budget. xliv. Financing and delivering healthcare for the growing population of the elderly is likely going to be the largest future cost-driver in the health systems of the EU8. All countries in the region face the consequences of population ageing caused by reduced fertility and mortality rates on the one hand, and increasing life expectancies on the other. The share of people aged 60 years and older in the total population in EU8 countries, which was 16 percent in 1995, is projected to increase to 27 percent in 2025. Correspondingly, the old age dependency ratio is expected to rise sharply. Older persons are more likely to have greater health needs and are more likely to consume more expensive healthcare services, particularly during the last years of life. xlv. Thus, despite a decade of reforms, the underlying supply-side incentives in the health system continue to be weak and generally ineffective, almost matching the near-absence of demand-side incentives which promote cost-consciousness. While most countries have introduced risk-sharing at the primary care level by paying the providers on the basis of capitation, similar incentives are by and large absent at the secondary and tertiary levels. Most hospitals in the EU8 countries are owned by the local or central governments and most hospital managers have no claim to any residual balances that may result from good management of the facilities. Prices for health services are typically set at the level of the central government, usually at levels very close to the costs of production and delivery of services, and with few rewards and penalties for staying within/breaching the budget, health providers pay scant attention to the cost side of the equation. xlvi. The health insurance system in many countries continues to function like centralized collection and reimbursement agencies, and many of the tasks and responsibilities usually identified with an insurance system are lacking. On the delivery side, the shift from the more resource intensive inpatient care to less expensive outpatient care has not been accomplished, and the culture of over-hospitalization and seeking specialized care persists. Recent gains from the reduction in length of hospital stays per episode, and increases in bed occupancy rates in individual facilities have been negated by slow progress in addressing the over-supply of hospital infrastructure. On the financing side, not all healthcare systems have been able to find appropriate mechanisms of reimbursement so as to motivate providers to deliver better quality of services, and to produce them in an efficient and cost-effective manner. On the management side, even though most state-owned hospitals enjoy a fair degree of de jure autonomy, few efforts have been made to improve administrative capacity and management within hospitals. xlvii. The scope of services covered by the public system have generally not been clearly defined and articulated, and formal copayments remain limited in most countries. Further, the pervasiveness of informal payments in health in many EU8 countries has become a serious impediment to healthcare reform. Besides contributing to the general environment of corrupt practices and the growth of a parallel healthcare financing system, informal payments introduce perverse incentives in the health system, and compromise efforts to improve efficiency, accountability, and equity in the delivery of health services. The non-transparent and discretionary nature of informal payments adversely affect access to healthcare, particularly for the more vulnerable segments of the population who have to pay disproportionately large amounts for health services that are supposed to be available free of charge.
24 Debts refer to payments outstanding and past due, not liabilities which are a natural consequence of the financial management of health facilities. 24 xlviii. Much remains to be done to improve the quality of health care. Most EU8 countries do not have well-functioning quality controls that regularly incorporate evidence-based medicine in the production, delivery, and financing of healthcare services, and institutional mechanisms to review the quality of care at the health facility level are generally lacking. Clinical protocols for major diseases are outdated and non- compatible with evidence-based medicine and cost-effectiveness analysis. The existing quality improvement systems are fragmented and are not integrated across different elements of the healthcare system, including professional self-regulation, purchasing by health insurance funds, training of health professionals, and management of individual healthcare practices. xlix. The solutions to these three problems are obvious and include some combination of: (i) rationalization of the benefit package and introduction of supplementary health insurance; (ii) introduction of co-payments or other mechanisms to make people take greater responsibility of their health and healthcare-seeking behavior; (iii) creating competition among healthcare providers; (iv) consolidating and closing public hospitals and reducing hospital beds; (v) risk-pooling and rational purchasing of health services; and (vi) introducing provider payment mechanisms that promote incentives for cost management and quality enhancement. While rationalization of health spending may also pave the way for lowering health contributions and the tax burden on labor, the latter objective may also be achieved by moving toward greater reliance on broader based general taxes for financing health care. This is the direction in which several European countries have gone, and with good results – perhaps reflecting greater effectiveness of government interventions to control costs in the health sector in such systems. l. Creating a universally acceptable benefit package to be funded from public sources is an attractive, but difficult proposition. Health insurance funds (or governments) tend to exclude some treatments or entitlements, which do not solve the problem. For the fiscal sustainability of health systems, a flexible mechanism needs to be in place that allows for a benefit package that responds to population needs and can be serviced within the given budget constraints. This flexibility needs two types of decisions. First, setting priorities and matching them with the principle of covering catastrophic costs; and second, ensuring legislative support for the package, so that full political responsibility for such decisions is taken and adequate levels of resources are earmarked. li. To a great extent, demand for healthcare is controlled by the individual consumer, who has an incentive to over-consume if they are responsible for only a fraction of the costs. The key to containing excessive and unnecessary demand for health services therefore lies in making individuals and families sensitive to the costs of additional health spending, while still limiting each family’s maximum outlays to affordable levels. This was the logic behind the introduction of patient co-payments in the health sector in Slovenia and Slovakia, a move that in both countries resulted in significant savings. The adverse equity effects of co-payments can be neutralized to a large extent by exempting the poor and vulnerable from making official co-payments or through targeted grants, though it is not easy to set up a system by which the poor and the vulnerable are always appropriately identified. lii. Further reforms are also needed to rationalize hospital infrastructure and contain the rapid growth in pharmaceutical costs and new expensive technology. Estonia has been a frontrunner in terms of downsizing hospital infrastructure, and much can be learned from the reforms there. Lessons can also be learned from the ongoing restructuring of hospitals in Slovakia, focused on the largest cities where over-capacity is most severe and where access is less of a concern. Drug price regulation is important and necessary, but the experience of many countries shows that regulation of consumption of pharmaceuticals is also critical in order to contain expenditures. Most countries have adopted demand-side measures for controlling consumption, and cost- sharing has proven to be the most effective. Lowering drug costs requires a multi-faceted approach, necessarily involving the policy-makers working together with payers, doctors, pharmacists, and patients, and involving a package of measures aimed at pricing, prescription, and safety of drug use. Finally, EU8 countries may be well-advised to consider regulating the use of high-end technologies, similar to most EU15 countries.25 liii. Looking forward, the social health insurance system in the EU8 countries will come under even greater pressure unless urgent action is taken to address rising health expenditures. At the very least, the health reform package would have to consist of a combination of stricter supply-side measures, such as
25 Countries with tax-based systems such as the UK and Spain have national agencies for high technology and technology control measures in their national health policies. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 25 management of pharmaceutical expenditures and hospital restructuring, and demand-side measures, such as greater patient responsibility for their own health and greater patient contributions, including cost- sharing for pharmaceuticals. The objective of these measures in the short-run has to be the stabilization of the fiscal situation so that no new debts are created in the health system. Likewise, the emphasis in the medium term should be on improving efficiency and effectiveness, ensuring access to healthcare, and enhancing the quality of care. liv. Estonia, Slovenia and Latvia, all exposed to the same set of financial pressures as the other countries in the region, have effectively managed health care finances through good governance, strict adherence to the rules of financial discipline, and by simply not spending what they do not have. To be sure, all three countries also carried out a host of reforms to control pharmaceutical spending, reduce hospital infrastructure, improve hospital management and so on, but the underlying difference between these three countries and the others is that of sheer fiscal discipline. Also, important lessons may be learned from several other EU countries such as Austria that were in a similar state at the time of membership, but found ways of managing their budgets without bringing about huge destabilizing changes. Managing fiscal risks from PPPs (Chapter V) lv. Public–private partnerships (PPPs) have a long history around the world, and have become increasingly popular also in the EU8 countries. Poland, Hungary, and the Czech Republic, in particular, have all experimented with such arrangements, not least in the road sectors. Public-private partnerships may enhance efficiency in service delivery and help create fiscal space through real fiscal savings, but managing the associated fiscal risks and achieving these objectives requires a strong institutional framework. There is little doubt that in practice part of the attraction of PPPs is that they allow governments to secure much- needed investment in public services without immediately having to raise taxes or borrow – in the EU8 potentially getting around Maastricht criteria for debts and deficits. Chapter V by Nina Budina, Tim Irwin, and Hana Polackova-Brixi reviews the experience with the use of PPPs in the EU8, discusses the fiscal risks associated with such arrangements, and proposed institutional improvements to better manage such risks. lvi. PPPs involve fiscal risks. In some PPPs, governments commit to purchase the output of their private partners, whether or not they subsequently want the output. In these PPPs, governments take on debt-like obligations with a value roughly equal to the cost of the investment, even if they need not disclose any new liability. In other PPPs, final users pay, but governments guarantee certain outcomes. Governments might pay nothing now, but face the risk of paying later, perhaps when they can least afford it. The danger of PPPs is that governments’ desire to avoid reporting immediate liabilities may blind them to the future fiscal costs and risks. That not only increases fiscal vulnerability, but may also lead to poorly designed projects with a higher cost than necessary or the use of PPPs when either public finance or purely private finance would be better. lvii. The authors argue that the extent of fiscal risk depends intimately on the fiscal institutions that shape and constrain government decisions toward PPPs – that is, on factors such as fiscal targets (including the Maastricht criteria), budgeting procedures, accounting and auditing standards, and the assignment of responsibilities for fiscal decisions among different parts of government. Such fiscal institutions affect decision makers’ incentives. They may, for example, reward policy makers for minimizing short-term cash expenditures, or alternatively for reducing fiscal vulnerability. Institutions may also influence the information available to decision makers: they might generate information only on traditional liabilities created by publicly financed investment, for example, or also on liabilities incurred in PPPs. Finally, they affect decision makers’ capacity: for example, they may or may not allow governments to analyze the risks in PPP projects and decide which are best borne by the government. lviii. Better fiscal institutions can therefore increase the chance that PPPs will be well designed and appropriately used. First, governments can take steps to improve the awareness of risks among officials and politicians. Second, they can impose upon themselves and lower tiers of government stronger requirements to disclose information about PPP contracts and the fiscal obligations that they create. Third, governments can continue to improve their fiscal planning, budgeting, and accounting in ways that help them choose their expenditure and investment plans rationally. Fourth, they can improve their ability to manage risks by allocating responsibility for taking on risk, developing quantitative monitoring of exposure, and so forth. Under such circumstances, PPPs may be a valuable tool for promoting sound and fiscally responsible investment in infrastructure. 26 References Alesina, A., and R. Perotti (1995), “The Polital Economy of Budget Deficits,” IMF Staff Papers 42. Barro, R.J. (1990), “Government Spending in a Simple Model of Endogenous Growth,” Journal of Political Economy 98. Boeri, T. (2003), “Social Policy Models in Transition: Why Are They So Different from One Another?” In Holzman, R., M. Orenstein, and M. Rutkowski eds.: Pension Reform in Europe: Process and Progress. World Bank (Directions in Development). Davoodi, H., and H. Zou (1998), “Fiscal Decentralization and Economic Growth: A Cross-Country Study.” Journal of Urban Economics 43. De Mooij, R.A., and P.J.G. Tang (2004), “Reforming the Public Sector in Europe: Reconciling Equity and Efficiency.” CPB Netherlands Bureau of Economic Policy Analysis. European Commission (2004), “Public Finances in EMU 2004,” DG ECFIN. Giavazzi, F., and M. Pagano (1990), “Can Fiscal Contractions be Expansionary? Tales of Two Small European Countries.” In Blanchard, O., and S. Fischer (eds): NBER Macroeconomic Annual. MIT Press. Heller, P. (2005), “Understanding Fiscal Space,” IMF Policy Discussion Paper 05/4 (March). Okun, A.M. (1975), “Equality and Efficiency,” Brookings Institution, Washington. Sapir, A. (2005), “Globalization and the Reform of European Social Models,” Background document for the presentation at ECOFIN Informal Meeting in Manchester, September 9. Schuknecht, L., and V. Tanzi (2000), “Reforming Public Expenditure in Industrialized Countries: Are there Trade- offs?” Paper prepared for EC DG ECFIN workshop on Fiscasl Policy in the EMU – New Issues and Challenges (November) Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 27
I. FISCAL CHALLENGES FOR THE EU8 COUNTRIES26 1. During the past 10 years, there have been striking changes in fiscal policy in the EU8 countries. Most of the EU8 countries have engaged in fiscal consolidation aimed at reducing government deficits toward the Maastricht criterion for euro adoption of 3% of GDP. Fiscal consolidation has been complicated by the completion of the transition process which together with the implementation of the acquis communautaire has burdened the expenditure side of government budgets. At the same time, these have been pressured by growing tax competition. The structural nature of the fiscal imbalances in most of the countries certainly has not eased this task. The new member states have coped with these various pressures in strikingly different ways and with quite uneven results. Fiscal contractions based on expenditure restraint have been undertaken in the Baltic countries and Slovakia, while other EU8 governments have pursued erratic fiscal policies and postponed politically difficult reform measures. As a result, in 2004 budget deficits were comfortably below 3% of GDP only in the Baltic countries and Slovenia, while deficits in Poland and Hungary remained stubbornly high (close to 5% of GDP) despite an exceptionally supportive environment. The consolidation efforts have stabilized debt-to-GDP ratios and in some countries led to declining debt burdens, thus contributing to the sustainability of public finances.
2. However, looking further ahead, demographic and budgetary projections show that fiscal positions may easily turn unsustainable in the absence of further fiscal reforms. Although many EU8 countries implemented important reforms over the past years (pension systems reforms, fiscal decentralization, and education reforms, among others), many important challenges remain. Progress in health care system reforms have been slow in most countries, and in many social benefit systems remain insufficiently targeted. Strengthening both the sustainability and quality of public finances is of fundamental importance for delivering the Lisbon Strategy goals of building an inclusive labor market and stimulating the growth of employment, both essential for achieving greater social cohesion and combating poverty within the European Union.
3. It has become increasingly clear that, whatever steps are taken to improve fiscal sustainability and quality, it is equally important to ensure that domestic budgetary rules and institutions contribute towards sound public finances. In the EU8, fiscal discipline clearly seems to depend on the institutional setting: the experiences from recent years suggest that deficit bias, overoptimistic revenue projections, creative accounting and one-off measures may all be linked to underlying institutional weaknesses. 27
1. Fiscal Developments 1995-200428 General government balances
4. Fiscal developments in the EU8 have varied significantly in recent years. In all countries, the fiscal deficit widened in 1998-99 in the wake of Asian and Russian currency crises, but since then developments have differed substantially. The Baltic countries, Slovenia, the Czech Republic, and Slovakia have pursued ambitious fiscal consolidation, while Hungary and Poland have allowed deficits and public debt to widen to worrisome levels in recent years (Figure I.1).
26 Prepared by Paulina Bucon and Emilia Skrok. 27 The role of fiscal institutions in shaping fiscal policy outcomes in the new member states is discussed in Chapter IV of the EU 2005 Public Finance Report and in Ter-Minassian, T. (2005). The EC points out that a common source of fiscal slippage in the new member states was failure of spending ministers and local authorities to internalize the social costs of their demands (the so-called “common pool” problem) as well as fragmentation of the process of formation, monitoring and implementation of budgets. 28 This report is based on information available through end-August, 2005. 28
Figure I.1. EU8 Fiscal Developments 1998-2004 (% of GDP).
Czech Republic Estonia Latvia Lithuania GG debt (left scale) GG debt (left scale) GG debt (left scale) GG debt (left scale) GG balance (right scale) GG balance (right scale) GG balance (right scale) GG balance (right scale) 40 -2 8 3 20 0 25 -1 2 35 -3 -1 -2 1 30 -4 6 15 -2 20 0 -3 25 -5 -1 -3 -2 -4 20 -6 4 10 -4 15 -3 -5 15 -7 -5 -4 10 -8 2 -5 5 -6 10 -6 8 9 0 1 3 4 2 4 2 4 1 9 0 0 1 3 2 3 3 4 8 8 9 8 9 0 1 2 9 0 0 0 9 0 0 0 0 0 9 0 0 0 0 0 0 9 9 9 0 0 0 9 0 0 0 9 9 0 0 9 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 9 9 9 9 0 0 0 0 9 2 1 1 2 2 2 2 2 2 2 2 2 1 2 1 1 1 2 2 2 2 2 2 2 2 1 1 2 Hungary Poland Slovakia Slovenia GG debt (left scale) GG debt (left scale) GG debt (left scale) GG debt (left scale) GG balance (right scale) GG balance (right scale) GG balance (right scale) GG balance (right scale) 65 -2 50 -1 50 -2 30 -1 -3 -3 -2 -4 45 45 25 -4 -3 -5 -2 60 -6 -5 -4 -7 40 40 20 -6 -5 -8 55 -9 -3 -7 -6 35 35 -10 15 -11 -8 -7 -12 50 -9 30 -8 30 -13 10 -4 4 1 3 4 9 0 1 9 8 9 0 1 2 3 4 8 8 3 2 0 2 9 0 9 0 0 0 0 0 0 0 0 0 0 0 9 0 0 0 0 0 0 9 0 9 9 0 9 9 9 9 0 0 0 0 0 0 0 0 9 9 0 0 0 0 0 9 0 0 9 9 0 0 0 0 0 9 18 23 19 24 20 21 22 2 2 2 2 2 2 2 2 2 2 1 1 1 2 1 1 1 2 2 2 2
Notes: CZ - without one-off imputation of state guarantees in 2003; HU - excluding one-off expenditure in 2002; HU, PL - with the effect of pension reform. Source: Eurostat; staff calculations.
5. Among the Baltic countries, Estonia has been running a fiscal surplus in recent years, while both Latvia and Lithuania have reduced deficits to below three percent of GDP. In all three countries, public debt levels are low at less than 20 percent of GDP. Similarly, Slovenia has strengthened its fiscal credibility significantly, with deficits now contained to around two percent of GDP and the debt burden manageable. Slovakia has also taken great strides to bring its fiscal house in order, and is now close to complying with the Maastricht fiscal criteria. The same would appear to be true in the Czech Republic. In contrast, fiscal deficits in Hungary and Poland exceeded five percent of GDP in 2004, with debt levels – particularly in Hungary, close to the critical 60 percent of GDP limit. 29
29 Fiscal deficits include the effect of pension reform (see Table I.1). Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 29
6. The fiscal consolidation that has taken place in most countries in the region in recent years reflects mainly strong output growth and administrative measures of various kinds, with real fiscal reforms confined to Slovakia and to a lesser extent the Czech Republic. Slovakia has implemented important fiscal reforms on both the revenue and expenditure side since the late 1990s, and especially in the last couple of years (Box I.5). There have also been important reforms in the Czech Republic, including in the area of social spending as well as public finance management. The positive development in 2004 was further affected by a change in budgetary rules that allowed unspent funds to be carried over to the following year. Meanwhile, efforts to control public finances in Hungary have generally been of an administrative nature, including in 2004 a new methodology for accounting VAT refunds. The fiscal deficit in all countries, except perhaps Estonia, thus remains of a structural rather than cyclical nature.
Table I.1. Effect of Pension Reforms in the EU8 (% of GDP).
Notes: *) CZ - without one-off imputation of state guarantees in 2003 (Eurostat figure: 11.7%); HU - excluding one-off expenditure in 2002 (Eurostat figure 8.5%). Source: Eurostat; staff calculations.
7. Revenue and spending patterns have also varied among in the EU8 countries in recent years. Several countries have been pursuing tax reforms aimed at lowering the overall tax burden, and general government revenues have been on a declining trend in the Baltic countries and Slovakia reaching a level of around 35 percent of GDP (Figure I.2). Expenditures in these countries have been scaled down correspondingly. Meanwhile, revenues have remained broadly stable at around 45 percent of GDP in the other countries in the region, with expenditures trending upward (except in Slovenia). 30
8. Thus, on the fiscal scene, two groups of countries emerge: those with moderate size of government and low deficits (the Baltic countries and Slovakia), and those with relatively large governments and significant fiscal problems (Hungary, Poland, and to some extent the Czech Republic). Slovenia falls in between, with large government but small deficit.
Figure I.2. EU8 General Government Revenue and Expenditure (% of GDP).
Czech Republic Estonia Latvia Lithuania
60 Total expenditure/ GDP 50 Total expenditure/ GDP 45 Total expenditure/ GDP 45 Total expenditure/ GDP Total revenue/ GDP Total revenue/ GDP Total revenue/ GDP Total revenue/ GDP 55 45 50 40 40
45 40
40 35 35 35 35
30 30 30 30 2 7 8 9 0 1 3 4 5 7 0 3 4 9 7 1 2 3 5 6 2 6 8 9 1 8 0 4 5 6 9 9 0 0 0 0 0 0 0 9 9 9 0 0 0 0 9 9 9 9 9 0 9 9 0 9 9 0 0 9 0 9 9 0 0 0 9 9 9 0 9 0 9 0 0 0 9 9 9 0 0 9 9 9 9 0 0 0 0 0 9 0 9 9 9 9 0 0 0 9 0 0 9 9 9 0 0 0 9 9 2 2 2 1 2 2 2 2 1 1 1 1 1 1 1 1 1 2 2 2 1 1 1 1 2 2 2 2 2 1 23 17 20 22 24 15 16 18 19 21 Hungary Poland Slovenia Slovakia
55 Total expenditure/ GDP 55 Total expenditure/ GDP 55 Total expenditure/ GDP 75 Total expenditure/ GDP Total revenue/ GDP Total revenue/ GDP Total revenue/ GDP 70 Total revenue/ GDP 65 50 50 50 60 55 50 45 45 45 45 40 35 40 40 40 30 4 6 7 0 1 2 3 5 5 6 7 8 8 9 0 1 1 2 2 5 4 6 7 9 0 3 3 4 4 8 9 0 0 0 9 0 0 9 9 9 9 9 9 9 0 0 9 9 9 0 0 0 0 9 9 9 9 9 9 0 0 0 0 0 0 0 9 0 9 9 0 0 0 0 0 0 9 9 9 9 9 0 9 9 9 9 9 9 9 9 0 0 0 0 0 9 9 9 0 0 0 9 9 9 9 0 0 0 0 0 0 22 2 21 23 15 16 17 18 19 20 2 2 2 1 1 1 2 1 1 2 1 1 1 1 1 1 1 2 2 2 2 2 2 2 1 1 1 2 2 2
Notes: CZ - without one-off imputation of state guarantees in 2003; HU - excluding one-off expenditure in 2002; HU, PL - with the effect of pension reform. Source: AMECO (ESA 95); staff calculations.
9. Fiscal outcomes in the Visegrad countries have generally fallen well short of targets agreed with the EU, although membership of the EU in 2004 forced additional discipline.30 A comparison of the budget outcomes achieved in the period 2000-2004 with the targets established in the Pre-Accession Economic Programs (2001-2003) reveals significant slippages in terms of both deficit and debt levels, but outcomes in 2004 were broadly in line with post-accession Convergence Programs (Spring 2004 and Autumn 2004) (Figure I.3). Deficit overshoots were caused mainly by higher than expected expenditure whereas revenues in some cases even exceeded the projected targets. On the expenditure side, discretionary spending stimulated by election cycles appears to have played a significant role in some countries.
30 The Baltic countries, notably Estonia and to some degree Latvia, were more successful in achieving program targets, despite lower than expected revenues. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 31
Figure I.3. Visegrad Countries - Divergence from EU Program Fiscal Targets (% of GDP).
General Government Balance
Czech Republic Hungary Poland Slovakia actuals ac t uals actuals ac t uals PEP 2001 4 PEP 2001 4 PEP 2001 4 PEP 2 0 0 1 PEP2002 PEP2002 4 PEP2002 3 PEP2002 PEP2003 3 PEP2003 PEP2003 2 PEP2003 2 2 CP 2004 May CP 2004 May CP 2004 May 1 2 CP 2004 May CP 2004 Nov CP 2004 Nov CP 2004 Nov 0 CP 2004 Nov 0 0 1 -1 -2 0 -2 -2 -4 -3 -1 -4 -4 -6 -2 -5 -6 -8 -6 -3 -8 -10 -7 -4 -8 -12 -5 -10 -9 -14 -10 -6 -12 2000 2001 2002 2003 2004 2005 2006 2007 2000 2001 2002 2003 2004 2005 2006 2007 2008 2000 2001 2002 2003 2004 2005 2006 2007 2000 2001 2002 2003 2004 2005 2006 2007
General Government Gross Debt
Czech Republic Hungary Poland Slovakia actuals actuals ac t uals ac t uals PEP 2001 65 PEP 2001 PEP 2001 60 PEP 2001 55 60 PEP2002 PEP2002 PEP2002 PEP2002 PEP2003 50 PEP2003 PEP2003 55 PEP2003 CP 2004 May CP 20 0 4 May CP 2 0 0 4 Ma y CP 2004 May 55 45 CP 2004 Nov 60 CP 2004 Nov * CP 2004 Nov 50 CP 2004 Nov
40 45 50 35 55 40 30 45 35 25 50 30 20 40 15 25 * including pension reform eff ect 10 45 35 20 2000 2001 2002 2003 2004 2005 2006 2007 2000 2001 2002 2003 2004 2005 2006 2007 2000 2001 2002 2003 2004 2005 2006 2007 2008 2000 2001 2002 2003 2004 2005 2006 2007
Source: PEPs; CPs; Eurostat. Revenue developments 10. EU8 countries have generally reduced the tax burden (i.e. the total amount of taxes and social security contributions) significantly over the last decade, with the largest reductions occurring in countries whose initial tax-to-GDP was the highest (Figure I.4).
Figure I.4. EU8 Change in Tax Burden 1995-2004 Relative to Initial Level.
2
EU15 CZ 0 25LT 30 35 40 45 -2
HU -4
LV EE PL -6
-8 Cumulative change 1995-2004
-10
SK -12 Total taxes, including social contributions, % GDP - base year 1995
Note: EU15 is simple average. Source: Eurostat; staff calculations. 32
Table I.2. EU8 Tax Burden 1995-2004 (% of GDP).
CZ EE LV LT HU 1995 2003 2004 1995 2003 2004 1995 2003 2004 1995 2003 2004 1995 2003 2004 TOTAL TAXES 36.236.236.137.933.432.833.729.128.628.628.527.743.639.239.2 of which Indirect taxes 12.311.411.913.913.112.713.711.511.412.311.911.217.815.816.1 of which VAT 6.3 6.5 n/a 9.8 8.9 n/a 9.3 7.3 n/a 7.7 6.8 n/a 7.7 9.1 n/a Excise duties 2.3 2.5 n/a 2.5 3.3 n/a 2.2 3.3 n/a 1.5 2.9 n/a 3.6 3.3 n/a Direct taxes 9.69.89.410.98.78.67.88.58.48.88.18.010.79.79.6 of which PIT 4.8 4.9 n/a 8.4 7.0 n/a 5.4 5.9 n/a 7.5 6.6 n/a 6.7 7.1 n/a CIT 4.6 4.6 n/a 2.4 1.7 n/a 1.8 1.5 n/a 1.3 1.4 n/a 3.8 2.2 n/a Social security contributions 14.415.114.713.111.511.212.19.08.87.68.68.515.013.513.4
PL SI SK EU8 EU15 1995 2003 2004 1995 2003 2004 1995 2003 2004 1995 2003 2004 1995 2003 2004 TOTAL TAXES 39.435.835.441.140.339.840.630.730.537.634.233.842.041.741.9 of which Indirect taxes 15.815.315.116.116.816.715.611.512.014.713.413.413.613.913.7 of which VAT 7.1 8.2 n/a 0.0 8.9 n/a 9.5 6.8 n/a 7.2 7.8 n/a 6.8 6.9 n/a Excise duties 3.5 3.9 n/a 0.2 3.2 n/a 2.5 2.1 n/a 2.3 3.1 n/a 2.8 2.7 n/a Direct taxes 12.87.27.07.18.48.411.67.25.89.98.58.212.512.912.9 of which PIT 8.1 4.1 n/a 6.1 6.0 n/a 3.6 3.3 n/a 6.3 5.6 n/a n/a n/a n/a CIT 3.3 2.2 n/a 0.5 1.9 n/a 6.1 2.8 n/a 3.0 2.3 n/a n/a n/a n/a Social security contributions 11.714.113.117.814.914.714.412.312.613.312.412.115.613.314.2
Source: Eurostat; AMECO (ESA 95); staff calculations.
For the group of countries as whole, the tax burden declined from 37.6% of GDP in 1995 to 33.8% GDP in 2004, driven by a drastic decline in Slovakia of more than 10 percentage points but also substantial reductions in Estonia, Latvia, Poland and Hungary (only the Czech Republic and Slovenia maintained broadly unchanged tax burdens during this period) (Table I.2) This compares to almost 42 percent of GDP in the EU15. The tax burden remains close to the EU15 average in Slovenia and Hungary, whereas the Baltic countries are substantially below.
11. The structure of taxation in the EU8 has also undergone significant changes. In particular, the share of direct taxes in total tax revenues has declined while the share of social security contributions and indirect taxes has increased. This reflects ongoing cuts in corporate and personal income tax rates, while high social security contribution rates have been largely maintained in the face of increasing pressures on social spending and some indirect tax rates have been increased (or bases broadened) in relation to EU accession. Nevertheless, as a share of GDP, all major tax categories have declined in the region as a whole and in most EU8 countries. The notable exceptions are social security contributions in the Czech Republic, Lithuania, and Poland, and both direct and indirect taxes in Slovenia, all of which are now higher than in the mid-1990s.
12. Compared to the EU15, EU8 countries in 2004 collected a relatively larger share of tax revenues from indirect taxes (and to a lesser extent social security contributions) and lower share from direct taxes. These differences are also reflected to a certain extent in the differences in the statutory tax rates, where the EU8 countries have higher-than-average consumption tax rates, but lower-than-average corporate and personal tax rates. Since direct taxes are more effective instrument for income redistribution, the differences between the EU8 and the EU15 suggest a stronger focus on redistribution in the old member countries. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 33
Figure I.5. EU8 Composition of Tax Burden 1995-2004 (% of GDP).
Social security contributions 50 Direct taxes Indirect taxes 40
30
20
10
0 4 5 0 4 5 0 4 5 4 0 5 0 4 5 0 4 5 0 4 5 0 4 5 0 4 5 0 0 9 0 0 0 9 0 0 9 0 9 0 0 0 9 0 0 0 9 0 9 0 0 9 0 0 9 0 9 0 0 0 9 0 0 9 0 0 0 9 0 9 0 0 9 0 0 9 0 0 9 0 0 9 2 1 2 2 2 1 2 2 1 2 2 1 2 2 2 1 2 1 2 2 1 2 2 1 2 2 1 CZ EE LV LT HU PL SI SK EU15
Source: Eurostat; AMECO (ESA 95); staff calculations.
13. Since the mid-1990s, a number of EU8 countries have implemented important reforms to their tax systems (Box I.1). The reforms differed in coverage and depth, but were generally aimed at reducing the tax burden on labor, particularly at the low-middle end of the pay scale, and at reducing corporate income tax rates while broadening bases. At the same time, however, in some countries personal and corporate income tax bases were eroded through the introduction of various tax relief schemes. Slovakia implemented a comprehensive tax reform that unified tax rates and eliminated most exemptions.
Box I.1. EU8 Tax Reforms EU8 countries have generally been reducing corporate income tax (CIT) rates while curtailing favorable special tax regimes (see tables below). Top CIT rates now vary between 28% in the Czech Republic and 15% in Latvia and Lithuania (Estonia has no tax on re-invested earnings). Some countries (including Slovakia) have also removed the double taxation of dividends. At the same time, some countries (including the Czech Republic) have introduced accelerated depreciation and increased R&D allowances.The tendency to reduce corporate tax rates and change favorable special tax regimes in the EU8 states has been reflected in effective tax rates. In the second half of the 1990s, effective corporate tax rates were growing in the EU-15, but falling in the EU-8 countries. Since then, both trends appear to have reversed and some convergence taking place.
Figure A. EU8 Effective Corporate Income Tax Rates (macro backward-looking approach). 25 CZ EE LV LT HU SK 20 EU- 8 EU- 15
15
10
5
0 1995 1996 1997 1998 1999 2000 2001 2002
Note: EU8 is simple average. Source: World Bank EU-8 Quarterly Economic Report, October 2004. 34
Recent changes in personal income taxes in the EU8 countries included mainly adjustments in tax allowances and exemptions aimed at social assistance (the Czech Republic, Estonia, Lithuania), but in some countries also lowering of tax rates. Most countries retain different personal income tax rates depending on income level and source, with only Slovakia moving to a single rate (but retaining a certain tax-free income threshold). Top marginal tax rates now vary between 50% in Slovenia and 19% in Slovakia.EU8 countries have continued their efforts towards harmonization of their indirect taxes (VAT and excises) with the EU15. Standard rates VAT now vary between 18% in the Baltic countries and 25% in Hungary. All countries have retained reduced rates for selected goods and services, while mostly rationalizing the items included in this group. The majority of the EU8 countries requested transitional periods for applying the standard rate on certain items, especially constructing services and heating (as well as the turnover threshold for SMEs), and for harmonizing certain excise duty rates (notably, excise duty rates on mineral oils, tobacco and alcohol are significantly lower in the EU8). These excises are being gradually raised. Table A. Top statutory tax rate on personal income.
1995 1998 2003 2004 CZ 40 40 32 32 EE 26 26 26 26 LV 25 25 25 25 LT 33 33 33 33 HU 44 44 40 38 PL 45 40 40 40 SI 50 50 50 50 SK 42 42 38 19
Table B. Top statutory tax rate on corporate income.
1995 1998 2003 2004 CZ 41 35 31 28 EE 26 26 26 26 LV 25 25 19 15 LT 29 29 15 15 HU 19.6 19.6 19.6 17.7 PL 40 36 27 19 SI 25 25 25 25 SK 40 40 25 19
Table C. VAT rates.
Standard rate 2004 Domestic zero Reduced 1998 2000 2003 rate1 rate(s) Standard rate Czech Republic 22 22 22 no 5.0 19.0 Estonia 18 18 18 yes 5.0 18.0 Latvia 18 18 18 yes 5.0 18.0 Lithuania 18 18 18 no 9.0 (5.0) 18.0 Hungary 25 25 25 yes 15.0 (9.0) 25.0 Poland 22 22 22 yes 7.0 (3.0) 22.0 Slovak Republic 23 23 20 yes 15.0 (5.0) 19.0 Slovenia - 19 20 no 8.5 20.0
"Domestic zero rate" means tax is applied at a rate of zero to certain domestic sales. It does not include zero rated exports.
14. With only marginal changes in the taxation of labor, tax wedges remain high in the EU8 countries, especially for low wage earners (Figure I.6).31 Hungary was most successful in recent years in reducing its tax wedge, albeit from a high level.
31 Tax wedge on labor costs is measured as the hypothetical income tax on gross wage earnings plus employee and employer social security contributions, expressed as a percentage of the total labor costs (defined as gross earnings plus employer social security contributions plus payroll taxes, where applicable). Low wage earners defined as single persons without children earning 67% of the average of a full-time production worker (APW). Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 35
Figure I.6. EU8 Tax Wedge on Labor Costs - Low Wage Earners.3
SK
SL
PL
HU
LT
LV 2004 2002 EE 1998 CZ
EU15
0 1020304050 Note: Slovenia data are for 2003. Source: Eurostat.
15. The level of general government revenues and grants in the EU8 countries is relatively high compared to a broader group of countries but similar to the average in the Central Europe and Asia region (Figure I.7).
Figure I.7. General government, total revenue and grants (% GDP).
50
40
30
20
10
0 EU8 EU 15 Sub- North Africa Latin Saharan America Af ric a Pacific & North Carabean America & America South As ia Middle East Middle East Asia & Asia East Europe and Europe Central Asia Central
Note: 2000-2004 average Source: IMF World Economic Outlook Database; staff calculations.
Expenditure developments 16. Expenditure developments have largely mirrored revenue developments over the past decade, with even larger contractions in spending relative to revenue in countries that initially had large fiscal imbalances (notably the Czech and Slovak Republics).32 Thus, the EU8 average public expenditure-to-GDP ratio declined from 46.5% of GDP in 1995 to 42.7% of GDP in 2004. Incidentally, a similar pattern was followed in the EU15 countries, where, however, spending still amounted to nearly 48% of GDP in 2004.
32 There are several complications in making cross-country comparisons of public expenditure levels or composition (Box I.2). 36
Box I.2. Public Spending Measurement Issues Cross-country comparisons of levels or composition of public spending are complicated by a number of factors: • The extent to which countries rely on tax expenditures as opposed to direct expenditure. While there is no consistent information on the amount of tax expenditure across countries, largely reflecting the absence of an agreed operational framework for measuring tax expenditures, the impression is that these are substantial in some countries. For example, in Poland, personal income tax expenditure programs, introduced in 1992, were originally used to compensate lower-income taxpayers for government withdrawal of price subsidies. However, the number and cost of these programs rapidly increased in the following years. While the marginal rates for the three income tax brackets are 40%, 30%, and 19%, income tax exemptions and deductions allowed under tax expenditure programs lowered the effective rates applicable to these brackets to 28%, 18%, and 13%, respectively, in 2003. • The taxation of social benefits. In some OECD countries, transfers are subject to broadly the same tax treatment as wage income (e.g. in the Nordic countries), while in others they are predominantly untaxed (e.g. sickness benefits in Poland and Hungary). For a given amount of net social transfers paid by the public sector and received by households, the level of public expenditures will be higher in the first group of countries. • The reliance on (mandatory and/or voluntary) private insurance schemes for pensions, unemployment, and/or health care systems. • Financing projects through public-private partnerships (PPPs) as opposed to conventional public investment. Several countries are increasingly relying on PPPs to finance investment projects (e.g. Hungary, Poland, and Slovakia). While this would show up as lower public spending in the short term, fiscal costs may materialize later (see Chapter V for a further discussion of these issues). • The use of guarantees on loans as a substitute for direct capital transfers or subsidies. Guarantees issued to state-owned enterprises (e.g. Czech Republic and Poland) are generally not reflected in general government expenditure in the year they are extended although they may show up in the fiscal accounts later when the guarantees are called. Source: OECD ( 2004); Polackova et al. (2004).
17. In the Baltic countries, spending increased sharply towards the end of the 1990s in response to the Russia crisis, but subsequently reverted to its downward path. Spending levels of around 35% of GDP are now among the lowest in the EU. Developments in the Visegrad countries have varied significantly. The Czech Republic cut back spending dramatically in the mid-1990s, but since then it has remained broadly stable at around 45% of GDP. Following a surge in spending in the difficult years of 1996-97, Slovakia similarly undertook an ambitious rationalization of spending programs and downsizing of the public sector, and spending declined to around 40% of GDP by 2004. Poland made good progress in cutting back public spending in the second half of the 1990s, but this was largely reversed during the period 2001-04. Hungary also allowed a significant increase in spending during these years. Slovenia maintained a high and stable spending ratio.
18. While EU8 expenditure trends in a longer retrospective generally have been dominated by the need for downsizing bloated public sectors, shorter term developments in some countries have been influenced by external or domestic shocks as well as business and political cycles (Box I.3).
Box I.3. Factors Affecting Expenditure Developments The literature typically identifies three typical factors that may affect public expenditure developments: (i) over the medium-long term, increases in per-capita incomes raises demand for some public goods and services (e.g. healthcare and education), thus leading to higher government spending (Wagner’s law); (ii) failure to rein in spending after periods of higher spending related to shocks, business cycles, or political cycles (hysteresis effect); and (iii) cyclical developments related to the factors mentioned in (ii). It is clear from the discussion above, that there is little support for the “Wagner” effect in the EU8 countries. Expenditures in most countries have trended down as per capita incomes increased, mirroring tax reductions and lower revenues. This reflects the initially bloated public sectors in all EU8 countries. Similarly, there is no clear evidence of any hysteresis or ratchet like effects. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 37
Over the shorter term, there is some evidence of counter-cyclical expenditure developments in the Baltic countries, but not in the Visegrad countries or Slovenia (Figure A). In Slovakia, on the contrary, the widening of the output gap after 1998 was accompanied by a strong compression of public spending
Figure A. Public Expenditure and Output Gap (%).
Czech Republic Estonia Latvia Lithuania Total expenditure/GDP (left scale) Total expenditure/GDP (left scale) Total expenditure/GDP (left scale) Total expenditure/GDP (left scale) Output gap (right scale) Output gap (right scale) Output gap (right scale) Output gap (right scale) 60 1 50 1 45 2 45 2 0 1 55 0 1 -1 0 50 -1 45 -2 0 40 -3 -1 45 -2 -4 40 -1 -2 -5 -3 40 -3 40 -6 -2 35 -4 -7 35 -4 -3 -8 -5 30 -5 35 -9 35 -4 30 -6 5 7 9 6 8 0 3 4 2 1 5 7 8 8 6 6 7 0 5 5 6 9 1 0 7 4 9 1 2 3 8 9 1 4 3 2 4 0 2 3 9 9 9 9 0 9 0 0 0 0 9 9 9 9 9 9 9 9 9 9 9 0 9 9 0 0 9 0 9 0 0 0 0 0 0 0 0 0 0 0 9 9 9 9 9 0 0 0 0 0 9 9 9 9 9 9 9 9 9 9 9 0 9 9 0 0 9 0 9 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 2 2 2 2 2 1 1 1 1 1 1 1 1 2 1 1 1 2 2 2 1 1 1 2 1 2 2 2 2 2 2 2 2 2 2 Hungary Poland Slovakia Slovenia Total expenditure/GDP (left scale) Total expenditure/GDP (left scale) Total expenditure/GDP (left scale) Total expenditure/GDP (left scale) Output gap (right scale) Output gap (right scale) Output gap (right scale) Output gap (right scale) 55 2 55 1 70 1 55 2
0 65 1 0 1 60 50 -1 50 55 50 0 -2 -1 0 50 45 -3 45 45 -1 -2 -1 -4 40
45 -2 40 -5 35 -3 40 -2 5 5 5 7 7 6 6 6 8 9 7 9 8 0 8 1 9 1 0 2 0 1 2 4 4 2 4 3 3 3 9 9 9 9 9 0 0 0 0 0 9 9 9 9 9 9 9 9 9 9 9 9 0 9 0 9 0 9 0 0 0 0 0 0 0 0 0 0 0 0 9 9 9 9 0 0 9 0 0 0 9 9 9 9 9 9 9 9 9 9 9 9 0 9 9 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 15 16 18 17 20 21 19 22 24 23 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2
Notes: output gap measures as the gap between actual and potential gross domestic product at 1995 market prices (% of potential GDP at market prices). CZ -without one-off imputation of state guarantees in 2003;HU - excluding one-off expenditure in 2002; HU, PL - with the effect of pension reform Source: AMECO (ESA 95); staff calculations.
There is also some evidence of political cycle dominated spending patterns in Estonia, Latvia, and Hungary, where spending has seemed to increase in periods leading up to Parliament elections (Figure B). On the other hand, no such patterns can be identified in the other EU8 countries.
Figure B.1. Political Economy at Work . Figure B.2. Political Economy at Work in the Visegrad in the Baltic Countries. Countries. Total expenditure/GDP 50 80 CZ HU PL SK Parliamentary elections 45 CZ HU PL SK
40 65
35
30 Total expenditure/GDP 50 EE LV LT Parliamentary elections 25 EE LV LT
20 35 1992 1994 1996 1998 2000 2002 1992 1994 1996 1998 2000 2002
Source: AMECO (ESA 95); staff calculations. Notes: CZ -without one-off imputation of state guarantees in 2003; HU - excluding one-off expenditure in 2002; HU, PL - with the effect of pension reform. Source: AMECO (ESA 95); staff calculations. 38
19. The composition of government spending has not changed much over the last decade, although a few EU8 countries have managed to improve the structure of spending somewhat (Figure I.8). Generally, lower interest rates have reduced the debt service burden, thus creating some fiscal space for productive spending. Slovakia has gradually reduced subsidies and collective consumption to the benefit of higher public investment and social spending. Nevertheless, subsidies remain relatively high and capital spending relatively low. In contrast, the Czech Republic has been increasing subsidies and only in recent years reversed a squeezing of capital expenditures. Hungary has made some progress since 2000 in reducing subsidies and raising public investment. However, sizeable interest savings have also been used to raise social spending and government consumption. In Poland, the structure of public finances has also improved somewhat with a rationalization of subsidies, social transfers, and collective consumption along with more resources for investment, although much remains to be done (and the large increase in “other” spending warrants scrutiny). In the Baltic countries, the recent tendency to increase subsidies (Estonia and Lithuania) and cut back social spending (Latvia and Lithuania) is a source of concern.
Figure I.8. EU8 Public Expenditure Composition (% of GDP).
60
50
40
30
20
10
0 0 5 0 4 5 4 0 5 4 5 0 0 4 4 5 0 4 5 0 5 5 0 4 5 0 4 4 9 0 0 9 0 9 0 0 9 0 0 0 9 0 0 0 9 0 9 0 0 0 0 9 0 9 0 0 0 9 9 0 9 0 0 9 0 0 0 9 0 0 0 9 0 0 0 0 0 0 9 0 9 9 1 2 2 1 2 1 2 2 2 2 1 2 1 2 2 1 2 2 1 2 2 2 1 2 2 1 2 CZ EE LV LT HU PL SI SK EU15 Subsidies Interes t Social transfers in kind Social benefits other than in kind Collective consumption Gross fixed capital formation Other
Notes: ESA 95 methodology. Social transfers in kind consist mainly of education and health services, although other services such as housing and cultural and recreational services are also frequently included. Social benefits other than in kind include: social security benefits in cash provided by social security schemes; private funded social benefits; unfunded employee social benefits (incl. (a) the continued payment of normal, or reduced, wages during periods of absence from work as a result of ill health, accident, maternity, etc.; (b) the payment of family, education or other allowances in respect of dependants; (c) the payment of retirement or survivors’ pensions to ex- employees or their survivors, and the payment of severance allowances to workers or their survivors in the event of redundancy, incapacity, accidental death, etc. (if linked to collective agreements); (d) general medical services not related to the employee’s work; (e) convalescent and retirement homes.); and (f) social assistance benefits in cash (ex. transfers made in response to natural disasters, recorded under other current transfers or under other capital transfers). Source: AMECO (ESA 95); staff calculations.
20. Social protection/benefits remain the largest category of public spending in all EU8 countries, although below the average EU15 level (Table I.3 and Table I.4). Social benefits and transfers in 2003 amounted to around 20% of GDP in the Baltic countries, Slovakia, and Slovenia, and 24-28% of GDP in the Czech Republic, Poland, and Hungary (ESA95). The second largest spending category is collective consumption in the range of 8-12% of GDP, in all cases but Lithuania higher than the EU15 average. It is interesting to note that the government wage bill appeared particularly large in Hungary and Slovenia, and to a lesser extent in Poland. Gross fixed capital formation varied between only 1.5% of GDP in Latvia to more than 4% of GDP in the Czech Republic, with most countries around 3% of GDP – somewhat higher than the EU15 average. Subsidies remained sizeable in the Czech Republic and to a lesser extent in Slovakia, Hungary, and Slovenia, while other countries in the region were below Western European levels. Finally, interest payments absorbed a significant share of resources in Hungary and Poland in line with their relatively high debt levels. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 39
Table I.3. EU8 Economic Classification of Public Expenditure 2003 (% of GDP).
EU15* CZ EE HU LV LT PL SK SI GFS 2001 Compensation of employees 11.8 4.3 n/a 12.7 8.9 8.9 9.7 8.6 12.8 Use of goods and services 6.6 4.8 n/a 6.7 6.0 5.7 8.2 6.4 10.3 Interest 3.0 0.8 n/a 4.2 0.8 1.3 2.8 2.5 1.6 Subsidies 1.6 9.7 n/a 2.9 0.8 0.3 0.7 1.8 1.4 Grants 0.7 0.1 n/a 0.0 0.0 0.0 0.0 0.1 0.0 Social benefits 20.2 12.5 n/a 16.3 10.0 12.8 18.3 17.4 18.6 Other expense 2.6 9.8 n/a 3.7 6.6 1.4 2.0 2.6 1.7 Expense 48.241.9n/a46.533.231.043.539.346.3
ESA 95 Subsidies 1.2 2.7 0.9 1.5 0.8 0.8 0.4 1.7 1.5 Interest 3.4 1.3 0.3 4.2 0.8 1.3 3.1 2.5 2.1 Collective consumption 8.2 12.3 8.8 10.8 11.0 7.6 9.1 11.0 8.4 expenditure Social benefits other than social 16.3 12.3 9.6 14.2 9.6 9.2 17.6 11.4 17.2 transfers in kind Social transfers in kind 12.5 11.7 9.9 13.4 10.8 10.8 8.5 8.9 11.9 Other current expenditure 1.9 1.2 0.9 1.7 0.8 0.2 1.5 2.2 1.3 Total current expenditure 43.5 41.6 30.3 45.7 33.7 29.9 40.2 37.8 42.4 Gross fixed capital formation 2.3 4.2 3.4 3.4 1.5 3.0 3.4 2.6 2.8 Total expenditure 47.753.235.850.735.734.249.039.248.2
Notes: * 2002. GFS 2001 on accrual basis except for LV and SI (on cash basis). GFS 2001 for EU15: simple average excluding GR, IT, PT. Source: AMECO (ESA 95); IMF Government Finance Statistics 2001; staff calculations.
21. EU8 countries tend to allocate a relatively larger share of spending toward “efficiency-oriented” programs, about the same on “basic functions” of the state, and less on income redistribution.33 Efficiency- oriented outlays ranged from 13% of GDP in Latvia to 22% of GDP in Hungary, compared to 17% of GDP in the EU15). All EU8 countries (except for Poland) allocated a larger share of total spending to this group than the EU15 (more than 42% of total spending in the Czech Republic, Hungary and Lithuania). Outlays on economic affairs were particularly high in Hungary and the Czech Republic, outlays on health were relatively large in Slovakia, Slovenia, and the Czech Republic (and low in the Baltic countries), and outlays on education very high in Slovenia (and low in the Czech and Slovak Republics). Spending on basic government functions amounted to 8-9% of GDP (except for Hungary at more than 12% of GDP)–below the EU15 average of 10% of GDP. This group of outlays accounted for more than 20% (EU15 average) of total spending in Hungary, Latvia, Lithuania and Slovakia, and less than 20% in the remaining EU8 countries. Finally, spending aimed at income redistribution ranged from 10% of GDP in Lithuania to 21% of GDP in Poland, similar to the EU15 average and the only EU8 country to devote a larger share of total spending to this purpose.
Table I.4. EU8 Functional Classification of Public Expenditure 2003 (% of GDP).
EU15 CZ EE HU LV LT PL SK SI General public services 7.0 7.4 3.2 8.2 5.3 4.1 7.0 5.2 8.5 Defense 1.6 1.8 1.8 1.3 1.3 1.5 1.0 1.8 1.4 Public order and safety 1.6 2.2 2.7 2.1 2.4 2.0 1.2 2.0 2.0 Economic affairs 4.6 11.7 3.8 5.7 3.7 4.2 3.2 5.1 3.5 Environment protection 0.7 2.0 0.6 0.8 0.4 0.5 0.6 0.7 0.5 Housing and community amenities 0.9 0.9 0.6 1.1 0.9 0.6 1.5 1.1 0.4 Health 6.2 6.5 4.1 5.7 3.3 4.3 3.0 2.3 6.8 Recreation, culture and religion 1.0 1.3 2.2 2.2 1.4 0.8 0.8 1.0 0.9 Education 5.7 4.9 6.4 6.1 6.2 6.0 6.3 4.3 5.9 Social Protection 19.1 14.6 10.4 17.0 10.8 10.1 19.9 15.7 18.3 Total Outlays 48.353.235.850.235.734.144.539.248.2
Source: Eurostat.
33 The “efficiency-oriented” expenditures comprise categories of economic affairs (including support programs and subsidies to mining, manufacturing, agriculture, energy and service industries), environment protection, health and education. “Basic functions” of the state and pure public goods include outlays on general public services, defense, public order and safety. These two broad categories essentially pursue “economic” objectives as opposed to programs aimed at income redistribution: social protection and housing and community amenities. 40
22. In a broader perspective, the level of general government expenditure in the EU8 countries is slightly above the average of the Europe and Central Asia region while below the levels observed in some of the most developed European countries (Figure I.9).
Figure I.9. General government, total expenditure (% GDP).
50
40
30
20
10
0 EU 8 EU15 Sub- North Africa Lat in America Saharan Africa Pacific & North Carabean America & America South Asia South Middle East Middle East Asia & Asia East Europe and Europe Central Asia Central Note: 2000-2004 average
Source: IMF World Economic Outlook Database; staff calculations.
Debt developments 23. Public debt levels in the EU8 have generally stabilized in recent years, although at high levels in some of the Visegrad countries (Figure I.1). In the Baltic countries, debt burdens have eased further to low levels (especially in Estonia) on account of buoyant output growth and strong fiscal positions. In Slovakia, debt levels have also come down from relative high levels a few years ago, and seem to have stabilized at just over 40% of GDP. In the Czech Republic, debt has shot up since the late 1990s on the back of sizeable fiscal deficits and assumption of contingent liabilities, but appears to have leveled off at around 35% of GDP. In Poland and Slovenia, debt has stabilized at just below 50% and 30% of GDP, respectively. Following a strong consolidation phase in 1998-2001, trends have reversed and the general government debt-to-GDP ratio has climbed back toward the critical 60% of GDP ratio.
24. In addition to generally improved fiscal balances, the broadly favorable debt developments in most EU8 countries have been aided by strong output growth, lower interest rates, and exchange rate appreciation. Decomposing debt dynamics into factors within and outside the government’s control, the latter “snowball effect” has generally been negative (reducing debt ratios) in recent years in the EU8 countries, with Poland during 2001-03 the main exception (Table I.5 and Appendix 1). Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 41
Table I.5. Snow Ball Effect on Debt (% GDP).
1999 2000 2001 2002 2003 2004 EUR-15 0.6 -0.4 1.5 1.2 2.3 0.2 Czech Republic 0.4 0.1 -0.2 0.4 -0.3 -1.5 Estonia 0.2 -0.4 -0.3 -0.2 -0.1 -0.2 Latvia 0.0 -0.3 -0.2 -0.5 -0.6 -1.1 Lithuania 1.9 0.7 0.2 -0.1 -0.5 -0.9 Hungary n/a -2.6 -1.5 -1.9 -1.4 -0.4 Poland n/a -0.8 1.4 1.9 1.2 -0.9 Slovenia n/a 0.2 -0.5 -0.6 -0.1 -0.3 Slovakia n/a -0.5 0.3 -0.3 -1.2 -1.8
Source: AMECO (ESA 95), staff calculations.
25. On the other hand, “hidden” fiscal deficits have generally added to the debt burdens of EU8 countries. Hidden fiscal deficits in the EU8 amounted to 1.3% of GDP annually (unweighted average) in the period 2000-2003 (Figure I.10). Hidden deficits were particularly large in Slovakia (5% of GDP in 2002-2003), and the Czech Republic and Estonia (roughly 2% of GDP in 2000-03). However, in Estonia the high hidden deficit comes from large accumulation of financial assets (securities other than shares). This is quite typical for governments in surplus with relatively small debts. In the Czech Republic, the hidden deficit reflects in part the reclassification as government of the Czech Banking Consolidation Agency. Finally, the finding for Slovakia is influenced by large privatization inflows, which have been adjusted for in the measurement of the hidden deficit but which were not in fact used to reduce debt (but rather accumulate reserves to finance pension reform).
Figure I.10. EU8 Hidden Fiscal Deficit 2000-03 (% of GDP).
5.0 2000-2001
4.0 2002-2003
3.0
2.0
1.0
0.0
-1.0 CZ EE LV HU PL SK LT
Notes: simple average; in some countries the revaluation effect is calculated using central government foreign debt currency structure. Source: AMECO (ESA 95); EBRD; MOFs; CBs; Eurostat; and staff calculations.
Intergovernmental fiscal relations 26. In recent years, the management of public finances has been influenced by a process of decentralization whereby the budgetary autonomy of lower levels of government has been increasing.34 In 2003, both expenditure and revenue shares at the local levels of government were around 20% of general government totals, comparable to EU15 average levels (Table I.6). Local governments were relatively larger
34 Intergovernmental fiscal relations are discussed in more detail in Chapter II. 42 in the Czech Republic, Hungary, Latvia, and Poland. At the same time, the share of central government in general government revenue and expenditure was generally significantly higher in the EU8 countries than in the EU15, reflecting the lesser reliance on other elements of the general government (notably social security funds) in the former group of countries.
Table I.6. EU8 Expenditure and Revenue at Central and Local Government Level 2003.
Total expenditure Total revenue % of total % of total EU 15 Central government 44.4 42.4 Local government 19.7 20.5 CZ Central government 68.6 61.9 Local government 21.6 26.2 EE Central government 67.1 68.9 Local government 22.2 19.5 LV Central government 53.2 50.1 Local government 24.3 24.8 LT Central government 52.0 48.3 Local government 18.6 19.6 HU Central government 52.2 47.6 Local government 21.9 24.4 PL Central government 44.9 38.9 Local government 24.2 25.6 SI Central government 51.9 50.3 Local government 15.8 16.5 SK Central government 52.9 47.9 Local government 15.2 16.5
Notes: ESA 95. The general government sector is divided into four subsectors: (a) central government (S.1311); (b) state government (S.1312); (c) local government (S.1313); and (d) social security funds (S.1314). The subsector central government (S.1311) includes all administrative departments of the State and other central agencies whose competence extends normally over the whole economic territory, except for the administration of social security funds. Included in subsector S.1311 are those non-profit institutions which are controlled and mainly financed by central government and whose competence extends over the whole economic territory. The subsector local government (S.1313) includes those types of public administration whose competence extends to only a local part of the economic territory, apart from local agencies of social security funds. Included in subsector S.1313 are those non-profit institutions which are controlled and mainly financed by local governments and whose competence is restricted to the economic territories of the local governments. Source: Eurostat; and staff calculations.
27. Local governments in the EU8 have generally not been important sources of fiscal deficits in recent years (Figure I.11). In 2003, local government budgets in most EU8 countries – even those with the largest local governments – were broadly balanced. The situation was not significantly different in 2001-02, although local government deficits in these years tended to be slightly higher. This likely reflects to a large extent binding borrowing constraints on local governments. Nevertheless, the figures do not reveal any potential hidden deficits and accumulation of arrears or contingent liabilities.
Figure I.11. EU8 Balances at Various Levels of Government 2001-2003.
5
0
-5
-10 Central government Local government Social security funds -15 2003 2002 2001 2003 2002 2001 2003 2002 2001 2003 2002 2001 2003 2002 2001 2003 2002 2001 2003 2002 2001 2003 2002 2001 2003 2002 2001 EU 15 CZ EE LV LT HU PL SI SK
Source: Eurostat. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 43 2. Medium-Long Term Fiscal Prospects Fiscal objectives 2005-08
28. The Baltic countries and Slovenia already comply with the Maastrict fiscal criteria for Euro adoption, and the Visegrad countries plan to follow suit in 2006-08. According to the most recent convergence programs, Slovakia would reduce its general government deficit to below 3% of GDP by 2006, Poland and Hungary by 2007, and the Czech Republic by 2008 (Figure I.12). All countries are planning an improvement in their respective primary, cyclically adjusted balances by 2007: in the period 2005-2007, the Czech Republic would run the highest primary, cyclically adjusted deficit, while Hungary would have reached a primary surplus already in 2005. The medium fiscal plans for the Baltic countries and Slovenia envisage sustaining the fiscal deficit below 3% of GDP in the period 2004-07 and fulfilling other pre-conditions (inflation, interest rates) for joining the EMU in 2007. While all countries (except Hungary) project a worsening of their debt levels over the period, they would remain safely below the 60% of GDP threshold. In the Visegrad countries, these prospects, however, rely on significant fiscal adjustment in the period under review.
29. The envisaged fiscal consolidation would rely mainly on reduced expenditures. Revenue rations would edge up in Lithuania and Slovakia, while easing in the Czech Republic, Estonia, and Hungary. Expenditures would be cut back by about 2.5 percentage points of GDP in the Visegrad countries (notably the Czech Republic and Hungary) and by 0.6 percentage points of GDP in the Baltic countries (driven by Estonia).
Figure I.12. EU8 Fiscal Plans 2005-07 (% or % of GDP). Czech Republic Estonia Latvia Lithuania Total expenditure/ GDP Total expenditure/ GDP Total expenditure/ GDP Total expenditure/ GDP Total revenue/ GDP Total revenue/ GDP Total revenue/ GDP Total revenue/ GDP 50 GG balance (right scale) -2 50 GG balance (right scale) 2 45 GG balance (right scale) 0 40 GG balance (right scale) -1
45 45 40 35 40 -3 40 1 35 35 35 -1 30 -2 30 30 -4 30 0 25 25 25 25
20 -5 20 -1 20 -2 20 -3 2004 2005F 2006F 2007F 2004 2005F 2006F 2007F 2004 2005F 2006F 2007F 2004 2005F 2006F 2007F
Hungary Poland Slovenia Slovakia Total expenditure/ GDP Total expenditure/ GDP Total expenditure/ GDP Total expenditure/ GDP Total revenue/ GDP Total revenue/ GDP Total revenue/ GDP Total revenue/ GDP 55 GG balance (right scale) -1 55 GG balance (right scale) -1 55 GG balance (right scale) 0 45 GG balance (right scale) -1
50 50 -2 50 -2 40 45 45 -3 45 -1 -2 40 -3 40 -4 40 35
35 -4 35 -5 35 30 -2 -3 30 30 -6 30 -5 25 25 25 -7 25
20 -6 20 -8 20 -3 20 -4 2004 2005F 2006F 2007F 2004 2005F 2006F 2007F 2004 2005F 2006F 2007F 2004 2005F 2006F 2007F Note: HU, PL – plan including 100% cost of pension reform. Source: Eurostat (Convergence Programs November 2004). There are several risks underlying these medium-term fiscal plans: macro-risks, including weaker output growth (e.g. as a result of higher oil prices and failure to engineer a recovery in Western Europe); political cycle risk (pre-election spending); failure to make adequate fiscal room for co-financing of EU structural funds or the provision of generous topping up of direct payments to agriculture; realization of contingent liabilities, including related to PPPs; and in a somewhat longer perspective, population ageing and pressures on pension and health care expenditures as well as fiscal costs related to compliance with the EU social inclusion policy and Lisbon strategy. 44 Medium-long-term fiscal sustainability 30. Current fiscal positions in some of the EU8 countries are unlikely to be sustainable over the longer term. The simplest way of looking at fiscal sustainability is to compare current primary fiscal balances with those that would be needed to maintain current debt levels.35 These basic calculations show that current primary fiscal balances are lower than what would be required to stabilize current debt ratios in the Czech Republics, Lithuania and Poland. Only Estonia and Latvia have some room for lasting fiscal expansions if they wish to maintain their current debt burdens.36
Table I.7. EU8 Debt Stabilizing Primary Fiscal Balance 2004 (% of GDP).
Actual primary Primary balance Real intererst rate 4% 6% balance 2004 2005-2007 * CZ -1.1 -0.4 -1.8 -2.4 EE -0.2 -0.1 2.0 0.2 HU -2.2 -1.2 -1.1 0.4 LV -0.7 -0.5 0.0 -0.8 LT -0.9 -0.5 -1.5 -0.9 PL -1.1 -0.3 -4.2 -0.5 SK -1.2 -0.4 -1.1 -0.4 SI -0.8 -0.3 0.0 0.1 Notes: *Projections from Convergence Programs. The primary fiscal balances required to stabilize current debt levels were calculated according to the equation: ⎛ 1+ n ⎞ x = ⎜ −1⎟ *b ⎝ (1+ g)(1+ P ) ⎠
where n – real interest rate, g – real GDP growth, – inflation rate, b – current government debt as a percentage of GDP.Real growth rates assumed in calculations: 4% for CZ, HU, PL, SK, and SI; and 6% for the Baltic countries (based on the average real growth rate projected for 2005-2007 in Convergence Programs); Inflation rates used are average projections for 2005-07 from Convergence Programs. Source: Burnside (2004); Eurostat; and staff calculations.
31. Clearly the assessment of long-term fiscal sustainability is a much more complex issue than suggested by the simple exercise presented above. More sophisticated analysis provided by the European Commission (2005) that takes into account underlying demographic changes and expenditure dynamics confirms that the Czech Republic faces severe challenges, and that fiscal pressures will also mount in several other EU8 countries over the longer term on current policies. Age-related expenditures (pensions and health) are projected to rise rapidly in the Czech Republic and Slovenia, reaching 15% and nearly 20% of GDP in 2009, respectively, and over 25% of GDP in both countries by 2050 (Table I.8). Meanwhile, some of the countries that have undertaken comprehensive pension reforms can look forward to an easing burden of age-related expenditures (notably Estonia and Poland).
Table I.8. EU8 Medium-Long Term Spending Projections (% of GDP).
Age related expenditure Total non-age related Total revenues Pen si on s H eal th car e Edu cati on Others expenditure 2009 2050 2009 2050 2009 2050 2009 2050 2009 (co nst) 2009 2050 CZ 8.6 15.2 6.5 9.3 3.8 3.6 30.2 47.5 47.5 EE 6.4 3.7 4.6 4.6 26.2 37.5 36.9 HU 7.4 7.6 34.6 42.4 42.4 LV 5.0 5.2 4.1 4.5 5.9 5.8 20.4 34.8 33.3 LT 5.3 7.0 4.6 4.6 25.1 34.8 33.3 PL 7.1 4.5 4.4 3.5 31.8 42.5 42.5 SK 6.9 7.4 5.0 6.6 3.4 3.4 0.8 0.6 20.3 36.9 36.9 SI 12.8 18.2 6.8 9.6 26.1 46.3 46.3
Source: EC (2005).
35 Of course in some EU8 countries debt levels should be brought down to reduce vulnerability and the risk of short-term financing difficulties, while in others (notably the Baltic countries) there is room to expand public debt from a pure fiscal sustainability point of view. 36 The assessment of fiscal sustainability is a much more complex issue that goes beyond the simple analysis presented in this chapter. A thorough overview of theoretical and practical issues regarding sustainability of public finances can be found in Burnside (2005). More recent analysis of fiscal sustainability has focused on the issue of uncertainty: Barnhill and Kopits (2003) adopt a Value-at-Risk methodology to address the effects of macroeconomic volatility and contingent liabilities; IMF (2003) models stochastic processes for key variables and uses Monte Carlo simulation to compute the probability distribution of debt; Xu and Ghezzi (2003) model the probability of default using simulated debt dynamics; and Mendoza and Oviedo (2004) model the effects of stochastic revenue flows on natural borrowing limits faced by a government. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 45
32. Even assuming that EU8 countries achieve their medium-term budgetary targets (baseline scenario) and incorporating the full impact of legislated structural measures, long-run projections reveal a significant risk of unsustainable public finances (measured against the 60% of GDP debt reference value) in almost all of the EU8 countries (except for Estonia, Poland and Slovakia) (Figure I.13). The Czech Republic would breach the debt limit even before 2030.
33. The risk of unsustainable public finances clearly increases considerably if EU8 countries do not achieve the targeted medium-term fiscal consolidation (2004 budget scenario). In this case, only Estonia (and with a stretch, Poland) manage to stay within the critical debt limit. This is a real source of concern given the tendency for fiscal slippages in the past (discussed above).
Figure I.13. EU8 Long-Term Public Debt Projections (% of GDP). Czech Republic Estonia
500 100 Programme scenario ("baseline") 50 400 Re f e r e n c e v alu e 2004 budget scenario 0 300 -50 200 -100 Programme scenario ("baseline") 100 -150 Reference value 2004 budget scenario 0 -200 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050
Latvia Lithuania
150 150 Programme scenario ("baseline") Programme scenario ("baseline") Reference value Reference value 2004 budget scenario 2004 budget scenario 100 100
50 50
0 0 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050
Hungary Poland
200 100 Programme scenario ("baseline") Reference value 150 2004 budget scenario 50
100 0
50 -50 Programme scenario ("baseline") Re f e r e n c e v alu e 2004 budget scenario 0 -100 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050
Slovenia Slovakia
250 200 Programme scenario ("baseline") Programme scenario ("baseline") 200 Reference value Reference value 150 2004 budget scenario 2004 budget scenario 150 100 100
50 50
0 0 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050
Source: EC (2005). 46
34. Accordingly, most countries will need to undertake significant fiscal adjustment over the medium-long term in order to preserve sustainability of their public finances. “Sustainability gap” calculations suggest an adjustment need of up to 8 percentage points of GDP in the Czech Republic, even in the baseline program scenario illustrated above (Table I.9). The size of the required adjustment in this case reflects age-related expenditure projected to exceed 25% of GDP in 2050, including expenditure on pensions that are projected to reach 15% of GDP. In contrast, in Hungary, where expenditure on pensions is projected to be at comparable levels in 2009 and 2050, the implied adjustment need amounts to only 0.6 percentage points of GDP––much lower than in the Czech Republic despite the fact that Hungary currently has a much larger deficit and higher public debt. Hungary and Poland provide good examples of successful pension reforms that have gone a long way toward securing sustainability of their pension systems, although it would in part be achieved through substantially lower replacement rates from the PAYG systems that may not be wholly offset by pension entitlements from the private pension funds (World Bank, 2005). Significant adjustment would also be needed in Latvia and Lithuania, as well as in other EU8 countries (except Estonia) to the extent they fail to comply with their medium-term fiscal targets.
Table I.9. EU8 Sustainability Gap Indicators (% of GDP).
Programme scenario ("baseline") 2004 budget scenario S1 S2 S1 S2 CZ 4.3 8.0 7.0 10.7 EE -2.5 -1.8 -3.7 -2.9 HU -0.2 0.6 1.2 1.9 LV 0.8 2.1 1.0 2.3 LT 0.4 2.6 1.4 3.6 PL -2.9 -1.8 0.2 1.3 SK -0.1 1.2 1.8 3.2 Notes: S1 measures the difference between the current tax ratio and the tax ratio that – if implemented now and sustained – would ensure a debt level within 60% of GDP in 2050 (a positive sustainability gap indicates an adjustment need). S2 indicates the change needed in tax revenue as a share of GDP that guarantees the respect of the inter-temporal budget constraint of the government, i.e., that equates the flow of revenues and expenses over an infinitive horizon. Source: EC (2005).
35. It should be recalled that the purpose of such long-term debt projections is to signal possible imbalances on the basis of current policies and demographic trends. To the extent policies or demographics change, so will long-term projections. The assessments presented in this chapter are therefore of a highly tentative nature– –providing an overview of possible future developments they do not focus on country-specific issues and limitations. More in-depth country-specific analysis is required to draw firm policy conclusions.
3. Quality of Fiscal Policy 36. The Lisbon European Council of March 2000 called for the emphasis of public finances to be broadened from its focus on stability to include the contribution to sustainable growth, full employment, social cohesion and competitiveness. The quality of public finances in this context refers to the structure of taxation and public spending as well as mechanisms to ensure a high level of efficiency of public spending. This section explores the quality of fiscal adjustments in the EU8 and the impact of tax and spending structures on growth.37 Quality of fiscal adjustments 37. There is a growing consensus among economists and policy makers that the success of efforts to consolidate public finances depends crucially on the “quality” of fiscal adjustments. “High quality” fiscal adjustments are characterized by adjustments that emphasize expenditure cuts rather than increased revenues and on tackling those expenditures that are politically most sensitive such as transfers, subsidies, and wage expenditures (Buti and Sapir, 1998; Alesina and Perotti, 1995 and 1997; Perotti, 1998; and von Hagen, Hughes Hallett and Strauch, 2001). Success is defined in terms of the persistence of the consolidation effort.38
38. Most EU8 countries carried out fiscal adjustments at some point during the last 10 years, with nearly all episodes of fiscal contractions that have been undertaken since 1998 based on expenditure contractions
37 Annex 1 in Volume II summarizes the recent literature on fiscal policy and growth. 38 Annex 2 in Volume II provides an overview of the adjustment experience of EU member states during the Maastricht convergence process and after the start of the third stage of the EMU. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 47
(Table I.10).39 Moreover, nearly one-half (seven out of 15 episodes) of these contractions were based on expenditure restraint where the contribution of spending cuts was at least 70% (expenditure-dominated contractions), although these were hardly present in last four years.
Table I.10. EU8 Patterns of Large Fiscal Contractions 1998-2004.
1998 1999 2000 2001 2002 2003 2004 Czech Republic Estonia Latvia Lithuania Hungary*) Poland Slovenia Slovakia Legend: - fiscal contractions episodes - "expenditure-dominated" contractions episodes (contribution of spending >= 70%)
Notes: *for Hungary in 1999-2000 the effect is estimated due to lack of data. Calculation based on methodology proposed in Von Hagen (2004); fiscal contraction based on expenditure restraint is defined as a change of the expenditure/GDP ratio as a percent of the change in the deficit/GDP ratio exceeding 70%. Source: Eurostat; and staff calculations. 39. Fiscal contractions in the EU8 countries in the period under review relied mainly on cuts in gross fixed capital formation or other capital transfers (Czech Republic, 1999 and 2004; Estonia, 2003; and Latvia, 2003), social benefits other social transfers in kind (Lithuania, 2002; Latvia, 2001 and 2002; and Slovakia, 2003) and collective consumption. The fiscal consolidation in the Baltic countries in 2000 is particularly noteworthy as it was characterized by an average deficit reduction of about 4.4% of GDP, lasted two or more consecutive years, and tackled forcefully government consumption as well as social benefits and social transfers in kind. In contrast, the contraction episode in Hungary in 2004 relied mainly on revenue adjustment, with a contribution of tax increases of 60%.
40. Ireland and Slovakia provide good examples of high quality, sustainable fiscal adjustment (Box I.4 and Box I.5). In both countries, fiscal adjustment was pursued over an extended period of time and relied on scaling back unproductive but politically sensitive expenditures.
Box I.4. Case Study: Ireland In the first half of the 1980s Ireland suffered from very low economic growth accompanied by large fiscal deficits and mounting public debt reaching 112 percent of GDP in 1987.The government elected in February 1987 focused its attention mainly on the expenditures side of the budget: general government expenditures was reduced from 52 percent o GDP in 1987 to 43 percent of GDP in 1990, while at the same time revenues decreased by 3.5 percentage points of GDP. The cyclically adjusted primary balance improved by about 8 percent of GDP during 1985-6 to 1990, and debt was reduced to about 90 percent of GDP by the turn of the decade.Spending cuts were directed mainly to transfers (down by 2.6 percent of GDP), the government wage bill (down by 1.5 percent of GDP) and public investment (down by 1.5 percent of GDP). Public employment was reduced by 10 percent during the period 1987-1989 (mainly in the health sector) through a hiring freeze, early retirement schemes and voluntary redundancy schemes. Also, in 1987 a social partnership was concluded, limiting wage increase to about 1 percent below inflation (ex-post).Tax reforms concentrated on personal income tax cuts (the upper rate was cut by 9 percentage points to 56 percent, while the standard rate was reduced by three percentage points to 32 percent) and lower corporate taxes (CIT rate reduced from 47 to 43 percent).The fiscal adjustment was part of a broader policy package and took place on the backdrop of a favorable external environment. The fiscal adjustment was preceded by a devaluation of Irish currency and pegging to the German mark along with liberalization of capital flows. Enhanced credibility was associated with lower interest rates (real rates dropped by 3-4 percentage points) and a recovery in investment. Also, lower interest rates generated a positive wealth effect, which – along with lower taxes – boosted private consumption. Meanwhile, the external environment was characterized by declining international interest rates (following the 1987 stock market crash) and recovery in the world economy benefiting the export-oriented Irish economy.On this background, Ireland experienced a significant improvement in its economic situation. Already during the years of fiscal contraction (1987-89), growth gained strong momentum (reaching over 6 percent in 1989) and outpaced that of G7 countries by a large margin, especially during 1990-91.
39 An episode of fiscal adjustment is defined as one where the general government deficit (ESA95) decreased by at least by 1% of GDP (this differs from the literature on fiscal adjustment in OECD countries, which looks at primary structural or cyclically adjusted deficits and assumes slightly higher and more persistent contractions; these more rigorous definitions are mainly used for checking “expansionary” episodes, which is not our aim). 48
Box I.5. Case Study: Slovakia* In 1996, a large current account deficit emerged as domestic demand surged, including because of a very expansionary fiscal policy. These imbalances persisted into 1997. In an effort to reverse the deterioration in the external accounts, monetary policy was tightened, but its effectiveness in slowing domestic demand was undermined by a further loosening of fiscal policy. The associated high real interest rate environment created strains in the banking sector and led to increasing financing difficulties for the government. The new government in 1998 responded with a series of measures focused on the expenditure side of the budget (reductions in public investments, social welfare and subsidies). An ambitious fiscal plan introduced in 1999 was based on further expenditure cuts, including the postponement of some major investment projects and a freeze in civil service wages as well as on tax increases (VAT, excises, improving tax collection and administration, and introduction of temporary import surcharge) and significant adjustments in administered prices to bring them more in line with costs. The government also strengthened control over public enterprises and accelerated their privatization. Better control was exercised over activities of extra- budgetary funds and rising arrears in health and transport sectors. However, the new government also inherited a large amount of bad loans and government guarantees for domestic and foreign borrowing by state-owned enterprises, and revealed contingent liabilities drove the fiscal deficit in 2000 above 12 percent of GDP. Nevertheless, by 2002, the government has shrunk by about 10% of GDP compared to 1997. The 2002 government adopted a more ambitious and comprehensive reform program, aimed at both tax and expenditure reforms and improving the overall business climate. On the expenditure side, the key reforms were in the areas of welfare (measures to reduce abuse of social benefits; and a mixture of benefit cuts and increased linkages to job search efforts to improve incentives to work), health care (introduction of co-payments, limiting the basic benefits package, and drug policy), elementary and secondary education (incentives for self-governments to rationalize the number of schools and replace historical-base financing with per-student payments), pensions (increase of retirement ages to 62 years, tightened link between future benefits and contributions, and introduction of a second funded pillar from 2005), and subsidies (deregulating administrated prices thus removing energy subsidies and cutting back support to agriculture, mining and construction). At the same time, fiscal management was strengthened (including establishment of a Debt and Liquidity Management Agency and a Treasury), and many public services (like schools, social care, and roads) were decentralized.40 On the tax side, the reform of 2004 featured a single rate of 19 percent for VAT, personal and corporate income tax, along with a widening of tax bases by eliminating most exemptions. Some taxes, like on dividends, were abolished to prevent double taxation. As a result of these fiscal reforms and a strong recovery in growth from 2002, both fiscal expenditures and revenues were reduced by about 15% of GDP and 13% of GDP, respectively, from 1997 to 2004.41 Social transfers dropped by almost 6% of GDP, and subsidies were lower by 3% of GDP. From 2002, there were also significant reductions in interest payments (1.5% of GDP) thanks to lower interest rates and better debt and liquidity management. The reduction in revenues reflected lower direct taxes (4% of GDP), indirect taxes (3% of GDP), and social security contributions (1% of GDP). Such significant fiscal changes would not have been possible without strong political backing. Although the strategies of the two reformist governments (1998 and 2002) in this respect were slightly different, both profited from the prospects of EU membership, the challenge that united coalition and opposition parties on all major issues. The 1998 government of a wide political spectrum was able to build on the financial crisis in 1998-1999.42 The 2002 centre-right government was more coherent on economic issues, and came to office well prepared with all major reforms drafted and a main message of “do it fast, while support is maintained.” The reforms went much beyond the EU agenda, although the government argued that this was the only way to ensure a secure landing in euro zone.
*) Valuable comments were received from the Institute of Financial Policy of the Slovak Ministry of Finance. 40 Further, in 2003 the Ministry of Finance cut employment by 30%, with savings used to increase remuneration of the remaining employees. 41 Three (inconsistent) time series of general government statistics in ESA95 are available: data from the Statistical Office contains non-consolidated figures for 1993-2002 and 2004, and consolidates figures for 2003; data published by the EC (Spring 2005) are consolidated for 2003 and 2004; and finally the Ministry of Finance, for its own analytical purposes, uses consolidated figures for 2002-2004 (similar to those of the EC). When comparing 1997 and 2004 (or 2002), one has to estimate the size of transfers within the general government in 1997 and adjust the series accordingly. 42 The left was represented by the Party of the Democratic Left, the centre by the Party of Civic Understanding and the right by the Slovak Democratic Coalition and Hungarian Coalition. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 49
Figure A. GG expenditure and GDP growth Figure B. GG Revenue, Expenditure and Balance (% of GDP and %, respectively). (% of GDP). 15 Total revenue 75 GDP GG expenditures Total expenditure 10 65 Deficit 55 5 45 35 0 25
-5 15 5 -10 -5
-15 -15 Y95 Y96 Y97 Y98 Y99 Y00 Y01 Y02 Y03 Y04 Y93Y94Y95Y96Y97Y98Y99Y00Y01Y02Y03Y04
Figure C. Composition of GG Revenues (% of GDP). Figure D. Composition of GG Expenditures (% of GDP). Social contributions 40 40 Interest VAT Subsidies 35 Corporate income tax 35 Social welfare Personal income tax Public sector wages 30 30
25 25
20 20
15 15
10 10
5 5
0 0 Y93 Y94 Y95 Y96 Y97 Y98 Y99 Y00 Y01 Y02 Y03 Y04 Y93 Y94 Y95 Y96 Y97 Y98 Y99 Y00 Y01 Y02 Y03 Y04 Note: in Figure B, the deficit in 1993 reached 31% of GDP. Source: 1993-2001 unconsolidated data of Statistical Office of Slovakia; 2002-2004 consolidated data from Ministry of Finance.
Fiscal policy and growth 41. The literature on the relationship between fiscal policy and growth is inconclusive. Various studies have looked at both the tax and spending side of public finances, but in neither case have there been consistent findings as regards the role of the overall level or composition of tax and spending on growth. There is more consensus that efficiency is key, both in tax policy and collection and in spending across different areas of the budget.43 In the EU8 countries, a simple scatter chart suggests that there might be a negative (albeit weak) relation between the overall tax burden and growth, but this may of course well be spurious correlation (Figure I.14).44 Figure I.14. Tax Ratio and Output Growth (for all EU8, average 1995-2003).
EU8 : Growth versus tax burden 1995-2003 5 4 0 4 5 3 a Tx/GDP 0 3 5 2 -5 0 5 10 15 GDP per capita growth
43 Annex 3 in Volume II provides an overview of expenditure efficiency considerations. 44 Tax shares may not be the best measures of tax distortion, but high tax burdens tend to be correlated with high marginal tax rates (Widmalm 2001). 50
42. Results of an econometric panel analysis covering the period 1996-2004 show that both the tax burden and the structure of taxes also seems to matter for growth (Appendix 2). The total tax burden is negatively related to growth in the sample, although the finding is not robust (Table A.1 and A.2). Further, there is a robust negative relation between the share of direct taxes plus social security contributions (presumably more distortionary taxes) and economic growth, with an increase in the share of these taxes by one percentage point associated with 0.3 percentage point lower growth (Table ). Personal income taxes are found to have a positive, albeit fragile, relation to growth, but this may reflect that the effects on work and investment incentives are not properly captured by the ratio of personal income taxes to total tax revenue, or that higher labor taxes have not been shifted onto higher real wages. On the other hand, both corporate income taxes and social security contributions have a negative and robust impact on growth. Finally, indirect (presumably less distortionary) taxes appear to have a positive and robust correlation with economic growth.
43. At the same time, none of the variables reflecting expenditure structure were robustly correlated with growth (Table). Nevertheless, two variables – gross fixed capital formation and social benefits other than social transfers in kind – were robust in some combination of conditional variables. In these cases, gross fixed capital formation had a positive impact on growth while social benefits other than social transfers in kind were associated with lower growth.
44. These preliminary results warrant cautious interpretation because of the limitation of the methodology used and quality/availability of data. There is as yet no comprehensive analysis of the relationship between tax/ expenditure structure and growth in the NMS and our analysis is only a first, small step in that direction. Synthetic indicator of the composition of public spending 45. The European Commission has developed a synthetic indicator to evaluate the composition of public spending across member countries (EC, 2003). The methodology developed by the EC aims to categorize expenditure items in terms of their efficiency effects and impact on fostering long-term growth (Figure I.15). Each spending component has a range of 2 points between the best and the worst performer. All items enter the indicator with the same weight. The higher the sum of scores in all categories, the better is the composition of public spending relative to other Member States.
Figure I.15. Theoretical Efficiency of Various Categories of Government Expenditure.
Efficiency
Category 2 Category 3 Category 4
Expenditure/ GDP
Category 1
Source: EC (2003).
Source: Benedek, D., O. Lelkes, A. Scharle, and M. Szabó (2004). Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 51
46. The EC has only calculated the indicator for EU15 countries, but we have supplemented this with our own calculations for the EU8 (Table I.11). Slovenia, Poland and – to a lesser extent – Slovakia and Hungary compare favorably to other EU member countries, reflecting good scores in education, health, and social protection (Slovenia), compensation of employees, social protection, and housing and community amenities (Poland), health, compensation of employees, and housing and community amenities (Slovakia), and public investment and social protection (Hungary). On the other hand, Latvia and the Czech Republic come out near the bottom of the table, with low scores in health, social protection and public investment (Latvia) and education, compensation of employees, and social protection (Czech Republic). These rankings do not appear to be correlated with the level of public spending, supporting the lack of evidence between spending and growth found above.
Table I.11. Composition of Public Spending in 2002 – Ranking.
Ranking Total expenditure (% of GDP) 1LU43.7 2SI48.1 3UK41.6 4PT*46.0 5FR*52.7 6PL*48.9 7FI49.8 8DE47.8 9 SK* 39.2 10 HU 52.6 11 LT 34.5 12 IT* 48.4 13 DK 56.2 14 AT 50.7 15 NL 47.8 16 BE 50.2 17 CZ 46.9 18 LV 35.8 19 GR* 50.0
Notes: The ranking does not include Cyprus, Estonia, Ireland, Malta, Spain and Sweden due to incomplete data. Calculations are based on GFS 2001 methodology (GG outlays by function and expenditure by type: education, health, social protection, housing and community amenities, compensation of employees, and interest) and ESA 95 (gross fixed capital formation and collective consumption). *FR - health and social protection figures for 2001; GR - education, health, social protection and housing and community amenities for 1999; compensation of employees and interest for 2000; IT - GFS data for 2000; PT - GFS data for 2001; SK - GFS data for 2003. Source: IMF GFS 2001-03; Eurostat; Polish Ministry of Finance; and staff calculations.
47. Clearly, the synthetic indicator has been designed only to benchmark good practices in public spending composition in member countries. Thus, it can only provide a highly suggestive signal regarding the quality of public spending in one member country relative to its peers. The indicator only provides ranking in relative terms and does not imply that higher-ranked countries are actually setting standards of good practice. Moreover, measuring only the level of spending on certain expenditure categories leaves out the issue of the output/outcome of these outlays and the indicator thus does not reflect the efficiency with which resources are used (for more detailed discussion on the issue of efficiency see Annex I.3 in Volume II). Data comparability and uncertain weighting of different expenditure component ratings are obviously also major shortcomings in this approach. Further, the indicator assesses public spending through the lens of its contribution to growth and as such does not capture other fiscal policy objectives such as stabilization and redistribution. The structure of public spending may result from social choices and more in-depth analysis is thus required to draw any policy conclusions. 52 APPENDIX 1 Decomposition of Changes in the Debt/GDP ratio 48. The “snowball” effect. Debt dynamics can be expressed as the change in the gross debt ratio resulting from the primary deficit, the “snowball” effect (contribution of interest and nominal growth) and the stock- flow adjustment:
D t - D t 1 PD t D t 1 i t y t SF t Y t Y t 1 = Y t + { Y t 1 • 1 + y t } + Y t Snowball effect where: Dt –general government gross debt, Yt –GDP at current market prices, yt –nominal GDP growth rate, SFt –stock-flow adjustment PDt – primary deficit, it- implicit interest rate, calculated as interest paid as % of gross debt at the end of period t-1.
49. The “hidden” fiscal deficit. Measuring “hidden” deficits entail first decomposing debt developments into the effects of initial debt, output growth, fiscal balance, and a residual (stock-flow adjustment - SF) which would include results from financial operations not related to deficit financing, privatization receipts, impact of exchange rate changes on foreign denominate debt, etc.
The change in debt-to-GDP ratio can be written as: g d d = b d + sf t + 1 t ( 1 + )( 1 + g ) t , where
αε(1+ i) sf = d + p + hd (1+ π )(1+ g) t
d, b sf, hd, p – general government gross debt, GG deficit, stock-flow adjustments, hidden deficit, privatization receipts as % of GDP;
π1t –GDP deflator; gt –real GDP growth; i – nominal interest rate on debt; a - share of foreign currency debt in total debt; ε - depreciation of the domestic currency (indicated by increase in exchange rate as units of domestic currency per unit of foreign currency); and t- time dimension (variables without time dimension are t+1).
50. In general, these would be expected to more or less cancel out over time. However, large and persistent stock-flow adjustments (especially if they have a negative impact on debt developments) should give cause for concern, as they “may be the result of inappropriate recording of budgetary operations and can lead to large ex-post upward revisions of deficit levels” (EC, 2003). This may reflect “quasi-fiscal” activities or “bad subsidies that serve special interest groups,” financed through accumulation of building contingent liabilities (Kharas and Mishra, 1999). Accordingly, the SF can be split into three components: privatization flows, exchange rate effects and remaining statistical adjustments (so called “hidden deficits”). However, interpreting “hidden deficit” involves a careful scrutiny of its size and the components. Remaining statistical adjustments usually include45: differences between the accrual and cash bases of recording transactions (example PL - 0.5% of GDP over the period 2000 to 2004); differences in the gross and net recording of transactions with financial assets; (i.e. accumulation of currency and deposits with banks, securities other than shares bonds issued by non-government units), loans and shares; and other adjustments reclassified of some units from non-government sectors to government and vice- versa, or early debt reimbursements.
45 Based on EC (2005). Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 53 APPENDIX 2
51. The analysis departs from the recent empirical growth literature (e.g. Kormendi and Meguire 1985; and Barro 1991). Previous studies have revealed that many of the structural and policy indicators found to affect growth are fragile and sensitive to small alterations in the conditioning information set (Levine and Renelt 1992; Folster and Henrekson 2001). We thus use a variant of the extreme-bounds analysis (Box I.6.).
Box I.6. EBA Methodology46 Extreme-bounds analysis involves the following steps (see e.g. Leamer 1983 and 1985): Imagine that there is a pool of variables (I and Z) that previously have been identified to be related to growth and one is interested to know whether the inclusion of a particular variable M is robust. Thus, one would estimate an equation of the form: Y = α + βmM+ βiI + βzZ + βdD + ε, where: Y is per capita GDP growth; I is a set of variables always included in the regression (e.g. the initial level of income, the investment rate, the secondary school enrolment rate, and the rate of population growth – following Levine and Renelt 1992); Z is subset of variables identified by past studies as potentially important explanatory variables of growth; usually up to three variables are taken from a pool of N variables available; and M is the variable of interest. Extreme-bounds testing involves varying the subset of Z-variables included in the regression to find the highest and lowest values for the coefficient on the variable of interest, (βm), that standard hypothesis test do not reject (at 0.05 significance level). Thus, the extreme upper bound is defined by the group of Z-variables that produce the maximum value of βm plus two standard deviations (βm + 2σm). A result is “robust” if βm remains significant and of the same sign at the extreme bounds. In contrast, if one finds a single regression for which the sign of the coefficient changes or becomes insignificant, the result is “fragile”. Thus, alteration in the conditioning information set may change the statistical inferences regarding the relationship between Y and M.
52. We considered the following basic I-variables in line with recent empirical studies: (i) the initial level of real GDP per capita; (ii) the investment-GDP ratio; (iii), the rate of population growth; (iv) the level of human capital (proxied by secondary school enrollment in line with common practice); and (v) the tax-to-GDP ratio (or in the expenditure analysis, the expenditure-to-GDP ratio). The latter was included in the base set to control for the overall tax (or spending) burden across countries and time while examining the role of the tax (or expenditure) structure (Widmalm, 2001). The Z variables included were the inflation rate (Stockman, 1981), the export-GDP ratio (measure of country openness), and the share of government consumption in GDP as indicator of the expenditure mix (or in the expenditure analysis, the tax-to-GDP ratio). The M variables of interest were in the case of the tax analysis: M – direct taxes and social security contributions; M1 – indirect taxes; M3 – social security contributions; M4 – personal income taxes; and M5 – corporate income taxes (all as a share of total taxes). In the expenditure analysis, the M variables were: M1 – subsidies; M2 – interest; M3 – social benefits other than social transfers in kind; M4 – collective consumption; M5 – social transfers in kind; and M6 – gross fixed capital formation (all as a share of total spending).
53. The empirical results are broadly in line with expectations (Table I.12.). In the first step, growth was regressed on the initial level of GDP, the investment share and the tax-to-GDP ratio (Model I).47 All coefficients have the expected signs and are significant. In the second step, population growth and secondary school enrollment were added as explanatory variables (Models II and III). The inclusion of these variables did not change the signs of the first three variables. Population growth has the “wrong” sign, but is not significant. School enrollment has the expected, positive and significant coefficient. However, the inclusion of this variable reduces the significance of the tax-to-GDP ratio. In the final step, the model is extended to include the Z variables used for the EBA analysis (Model IV). These variables are all of the expected sign and significant.
46 The EBA method has been criticized for “reverse data-mining” (Sala-i-Martin, 1997) and multi-collinearity (Levine and Renelt, 1992). 47 Results were similar for the expenditure analysis and are not shown here. The spending-growth elasticity was estimated at -0.36. 54
Table I.12. EU8 Panel Growth Regressions 1995-2004 (dependent variable: growth rate of per capita GDP).
Model IIIIIIIV Initial GDP per capita (Y0) -0.00*** -0.00*** -0.00*** -0.00*** (-2.9) (-2.5) (-3.8) (-5.9) Investment share of GDP (INV) 0.20*** 0.18*** 0.22*** 0.17** ( 3.0) (2.8) ( 2.9) (2.2) Tax-to-GDP ratio (TAX) -0.15** -0.13* -0.12* -0.15** (-1.8) (-1.6) (-1.4) (-1.7) Population growth (POP) -0.62 -- (-0.63) Secondary school enrolment ratio (HUM) 0.13*** 0.19*** (2.8) (4.0) Government consumption share of GDP -0.46*** (GOV) (-2.8) Openness (OPEN) 0.05** (1.8) Average inflation rate (INFL) -0.03*** (-2.1) Numbers of observations 80 80 61 59 R-sq (overall): 0.27 0.29 0.36 0.49
Note: Random effects estimation. ***(**)[*] stands for 1% (5%)[10%] significant
54. The initial level of GDP, the investment ratio, and the human capital variable are all robustly correlated with output growth (Table I.13.). Meanwhile, population growth and tax ratios are fragile, although the latter retains a negative sign.
Table I.13. Robustness of Base Variables.
M-variable Bound Bm t-ratio N R2 Z Robust Investment share min 0.17 2.18 59 0.48 OPEN,GOV,INF base 0.22 2.87 61 0.46 - yes max 0.28 3.39 59 0.39 OPEN, GOV Initial GDP min -0.00 -4.72 59 0.39 OPEN,GOV base -0.00 -3.78 61 0.36 - yes max -0.00 -3.51 61 0.38 INF Population min -1.83 -1.80 59 0.52 OPEN,GOV,INF growth base 0.50 0.42 61 0.36 - no max 1.49 0.92 59 0.33 OPEN Tax-to-GDP min -0.19 -1.58 59 0.36 OPEN,INF base -0.12 -1.35 61 0.36 - no max -0.07 -0.67 61 0.39 GOV Human capital min 0.13 2.80 61 0.40 GOV base 0.13 2.84 61 0.36 - yes max 0.19 3.95 59 0.48 OPEN,GOV,INF
Note: “base” corresponds to Model III above
55. The structure of taxes also seems to matter for growth. We find a robust negative relation between the share of direct taxes plus social security contributions (presumably more distortionary taxes) and economic growth, with an increase in the share of these taxes by one percentage point associated with 0.3 percentage point lower growth (Table I.14.).48 Personal income taxes are found to have a positive, albeit fragile, relation to growth, but this may reflect that the effects on work and investment incentives are not properly captured by the ratio of personal income taxes to total tax revenue, or that higher labor taxes have not been shifted onto higher real wages. On the other hand, both corporate income taxes and social security contributions have a negative and robust impact on growth. Finally, indirect (presumably less distortionary) taxes appear to have a positive and robust correlation with economic growth.
48 These results are based on Model I, but do not change fundamentally in the alternative specifications. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 55
Table I.14. Tax Structure and Output Growth (shares of total taxes).*
M-variable Bound Bm t-ratio N R2 Z Robust Direct taxes and social min -0.31 -3.84 78 0.40 OPEN security contributions base -0.30 -3.38 80 0.40 - yes (M) max -0.26 -2.59 80 0.42 GOV Indirect taxes (M1) Min 0.26 2.60 80 0.41 GOV base 0.30 3.39 80 0.40 - yes max 0.31 3.85 78 0.40 OPEN Direct taxes (M2) min 0.08 0.80 78 0.30 OPEN, INF base 0.09 0.99 80 0.26 - no max 0.16 1.79 78 0.27 OPEN, GOV Social security min -0.36 -4.26 78 0.46 OPEN, GOV contributions (M3) base -0.30 -3.61 80 0.40 - yes max -0.28 -3.76 80 0.40 INF
Personal income taxes min 0.08 0.94 78 0.26 OPEN, INF (M4) base 0.11 1.45 80 0.31 - no max 0.15 2.49 80 0.38 GOV, INF Corporate income min -0.22 -2.44 78 0.33 OPEN taxes (M5) base -0.22 -2.58 80 0.34 - yes max -0.19 -2.13 80 0.36 GOV, INF
*based on Model I 56. None of the variables reflecting expenditure structure were robustly correlated with growth (Table I.15.). Nevertheless, two variables – gross fixed capital formation (M6) and social benefits other than social transfers in kind (M3) – were robust in some combination of conditional variables. In these cases, gross fixed capital formation had a positive impact on growth while social benefits other than social transfers in kind were associated with lower growth.
57. Our findings should be interpreted with cautious given a number of problems that are commonly encountered in this type of cross-section regressions. The most important of these may be a potentially severe simultaneity problem, including potentially arising from business cycle effects and Wagner’s law (the tendency for government expenditure to be higher at higher levels of per capita GDP). As a result, the independent variable (like taxes or government spending as a share of GDP) is correlated with the error term in the growth regression, and this bias could impact regression results. While the possible simultaneity effect is an argument in favor of panel regressions with shorter time spans (as used by us), it at the same time limits the room for including potentially important lags of the explanatory variables. Further, there may be concerns about data quality, and measurement or misspecification errors may give rise to heteroskedasticity and misleading results.49 However, heteroscedasticity most often appears in a form where the error term is correlated with one of the independent variables or with the dependent variable, and this does not seem to be a major problem in our data.50
49 The linear ordinary least squares regression (OLS) makes the assumption that the variance of the error term in regression is constant (homoscedasticity). 50 Visual inspection of residuals plotted against fitted values show the distribution of the residuals in our pooled regression did not seem overly heteroscedastic. The more formal tests for heteroscedasticity (Breusch-Pagan / Cook-Weisberg test and White’s general test) also confirm that heteroskedasticity was not a problem in our pool regression. Not surprisingly, therefore, common corrections for heteroscedasticity such as the robust/White standard errors hardly change the results. 56
Table I.15. Expenditure Structure and Output Growth.
M-variable Bound βm t-ratio N R2 Z Robust Subsidies (M1) min -0.11 -0.25 52 0.39 INF base -0.01 -0.21 52 0.38 - no max 0.66 1.67 52 0.65 TAX, OPEN Interest (M2) min -0.36 -1.65 52 0.67 TAX,OPEN,INF base 0.01 -0.06 52 0.38 - no max 0.02 -0.06 52 0.39 INF Social benefits other min -0.54 -2.56 52 0.48 INF than transfers in kind base -0.48 -2.60 52 0.47 - no (M3) max -0.27 -1.45 52 0.67 TAX, OPEN INF Collective min -0.15 -1.24 52 0.61 OPEN, INF consumption (M4) base -0.09 -0.63 52 0.39 - no max 0.12 0.77 52 0.50 TAX, INF Social transfers in kind min 0.14 0.38 52 0.65 TAX, OPEN, INF (M5) base 0.29 0.62 52 0.39 - no max 0.29 0.62 52 0.39 - Gross fixed capital min 0.36 1.26 52 0.41 INF formation (M6) base 0.38 1.41 52 0.41 - no max 0.64 2.96 52 0.65 OPEN Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 57 References Barnhill, T.M., Kopits Jr., G. (2003). ”Assessing Fiscal Sustainability under Uncertainty.” Manuscript, George Washington University. Barro, R.J. (1991). “Economic Growth in a Cross-Section of Countries.” Quarterly Journal of Economics, 106(2): 407-443. Benedek, D., Lelkes, O., Scharle, A., Szabó, M. (2004). “The Structure of General Government Expenditure and Revenues in Hungary, 1991-2002.” Ministry of Finance Working Paper No.9. Burnside, C. (2004). “Some Tools for Fiscal Sustainability Analysis.” Duke University, November 2004 Burnside, C. [ed.] (2005). “Fiscal Sustainability in Theory and Practice.” The World Bank, August. Buti, M., Sapir, A. (1998). “Economic Policy in EMU: a Study by the European Commission Services.” Clarendon Press, Oxford, Oxford University Press, New York. European Commission (2003). Public finances in EMU - 2003. European Economy, 3/2003. European Commission (2005). Public finances in EMU - 2005. European Economy, 3/2005. Fölster, S., Henrekson, M. (2001). “Growth Effects of Government Expenditure and Taxation in Rich Countries.” European Economic Review, 45(8): 1501-1520. IMF (2003), “Sustainability Assessments-Review of Application and Methodological Refinements.” Manuscript, Policy Development and Review Department, International Monetary Fund, June. Kharas, H., Mishra, D. (1999), “Hidden Deficits and Currency Crises.” Mimeo. World Bank. Kormendi, R.C, Meguire, P.G., (1985). “Macroeconomic Determinants of Growth: Cross-country Evidence.” Journal of Monetary Economics, 16: 141-164. Leamer, E.E. (1983). “Let’s Take the Con out of Econometrics.” American Economic Review, 73(1), 31-43. Leamer, E.E. (1985). “Sensitivity Analyses Would Help.” American Economic Review, 75(3), 308-13. Levine, R., Renelt, D. (1992). “A Sensitivity Analysis of Cross-country Growth Regressions.” The American Economic Review, 82(4): 942-963. Mendoza, E.G., Oviedo, P.M. (2004). “Fiscal Solvency and Macroeconomic Uncertainty in Emerging Markets: The Tale of the Tormented Insurer.” Manuscript, University of Maryland, April. OECD (2004). “Enhancing the effectiveness of public spending: Experience in OECD countries”, WP No 380 Polackova Brixi H., Valenduc Ch., Zhicheng Li Swift (2004). “Tax Expenditures – Shedding Light on Government Spending through the Tax System. Lessons from Developed and Transition Economies”. World Bank. Sala-i-Martin, X. (1997). “I Just Ran Four Million Regressions.” NBER Working Paper No. 6252, National Bureau of Economic Research.
Ter-Minassian, T. (2005). “Fiscal Issues in Accesion Countries.” Mimeo. International Monetary Fund, Washington, D.C. Von Hagen, J., Hughes Hallet, A., Strauch, R (2001). „Budgetary Consolidation in EMU.” European Commission, Economic Papers No. 148. World Bank (2004). “EU-8 Quarterly Economic Report. Special Topic: Corporate Income Taxation and FDI in the EU-8.” The World Bank, Warsaw, October. World Bank (2005). “EU-8 Quarterly Economic Report. Special Topic: Sustainability of Pension Systems in the EU- 8.” The World Bank, Warsaw, October. Widmalm, F. (2001). “Tax Structure and Growth: Are Some Taxes Better Than Others?” Public Choice, 107(3-4): 199-219. Xu, D., Ghezzi, P. (2003). “From Fundamentals to Spreads: A Fair Spread Model for High Yield Emerging Markets Sovereigns.” Manuscript, Deutsche Bank, July. WB (forthcoming) Bulgaria Public Finance Policy Review, 58 II. INTERGOVERNMENTAL FISCAL RELATIONS51
1. Introduction 58. Subnational governments are an important part of the public sector in the EU8 countries. They provide basic public services in both the social sectors (education, health, and social assistance) and in infrastructure (water supply, sewerage and transport). They account for about one-quarter of government spending. How well they perform these functions has an important influence on the quality of both human and physical capital in these countries, and therefore on the prospects for economic growth. It also has equity implications, as the social services performed by municipal governments have important distributional effects. Further, subnational government operations have implications for macroeconomic stability, as independent fiscal actions by local governments could undermine central control over fiscal policy.
59. The period since the fall of the Berlin Wall and the breakup of the Soviet Union has seen dramatic changes in the political and economic institutions of the EU8 countries. All abolished the Communist party’s monopoly on political power, abandoned central planning as a device for allocating resources and setting targets for deliverables, and were prompted, in reaction to the high degree of centralization under the former regimes, to grant substantial power to subnational units of administration. The EU8 countries have spent the last 15 years devising a new system of intergovernmental arrangements that will respond to these changes. The aim of this chapter is to see how far along they have come and how well they now perform.
2. Evaluation Criteria 60. According to conventional economic theory, the primary aim of a system of local government finance is to promote efficiency in the allocation of resources (Box II.1). Theory argues that if the benefits of particular services are largely confined to local jurisdictions, welfare gains can be achieved by permitting the level and mix of such services to vary according to local preferences. Local consumers, confronted with the costs of alternative levels of service can express their preferences by voting for rival political candidates (i.e., voting with their hands) or moving to other jurisdictions (voting with their feet). In this respect, local governments can approximate the efficiencies of a market by “pricing” municipal services and relying on the local political process and household mobility to clear the market. On this basis, a successful system of intergovernmental relations would be one in which each level of government (1) performs only functions whose benefits are confined to its boundaries and (2) derives its revenues from sources (such as local taxes and fees) that confront its residents with the costs of these functions.
61. In practice, there are a number of limitations to this model. First, the benefits of public services can be difficult to define in geographical terms. While most of the benefits of spending on public health, for example, may be confined to a local jurisdiction, failures to control infectious diseases can have national or international implications.
62. Other aspects of the theory confront practical obstacles as well. As described in Box II.1, theory assigns the role of income redistribution to the central government. It is presumed that central governments will use the instruments under their control (e.g., progressive taxation and direct payments to low income households) to achieve a desired level of income equality, leaving local governments to focus only on the level and mix of services they provide. But spending on public services – particularly education and health – can be important instruments of distributional policy. Where these functions are assigned to local governments, their distributional implications cannot be ignored. Theory also assumes that macroeconomic stability is not a local government concern. But high levels of local taxation or widespread deficit spending could weaken the central government’s control over fiscal policy.
63. As a practical matter, the intergovernmental fiscal systems of the EU8 countries must be held to somewhat different standard: Transparency and predictability: The principal challenge confronting the EU8 countries was to devise a system for financing local government that would be perceived as fair and stable. The emergence of multi-party democracy in the EU8 required transparency in order to allay charges of partisan favoritism
51 Prepared by William Dillinger. Signe Zeikate provided research assistance and prepared the section on debt controls. Ken Davey and Deborah Wetzel were peer reviewers. An earlier version of this chapter was discussed in a regional workshop in Budapest in June 2005. Current Issues in Fiscal Reform in Central Europe and the Baltic States 2005 59
and reduce the transaction costs that would result from continued reliance on bargaining. Predictability was required to provide local governments with a stable basis for budget planning. Equity: Because local governments assumed responsibility for social services, the new system had ensure that such services would be universally available, regardless of the strength of the local tax base. Macroeconomic control: The system had to ensure that local government fiscal behavior would not threaten macroeconomic stability. Central governments needed sufficient control over aggregate local taxing to conduct fiscal policy and sufficient control over local borrowing to ensure that aggregate deficit targets were achievable.
Box II.1. A Snapshot of the Economics of Expenditure Assignment Which functions should be assigned to which level of government? The literature on fiscal federalism provides a useful framework for thinking about this question. It begins by assigning three roles to the public sector as a whole: macroeconomic stabilization, income redistribution, and resource allocation (in the case of market failure). The model assigns the first two roles to the central government. The central government is assigned responsibility for stabilization on the grounds that local economies have no access to an independent monetary policy and are too open for effective countercyclical measures to be effective. The income redistribution function is also assigned to the central government, on the grounds that local attempts to address income disparities are likely to provoke inefficient migration: higher-income groups will move to low-tax areas and low-income groups to high-benefit areas (even with their populations in situ, local governments in poor regions would have little income to redistribute anyway). Subnational governments enter the picture only with respect to the third function: resource allocation. If the benefits of particular services are largely confined to local jurisdictions, residents can choose the level and mix of services that best matches their preferences (and pocketbooks). In this way, a system of intergovernmental finance can approximate the allocative efficiencies of a market in the allocation of local public goods.