Comparing Fiscal Federalism [1 ed.] 9789004340954, 9789004340930

Citation preview

Comparing Fiscal Federalism

Studies in Territorial and Cultural Diversity Governance Series Editors Francesco Palermo Joseph Marko Editorial Board Cheryl Saunders (University of Melbourne) G. Alan Tarr (Rutgers University, Camden, nj) Anna Gamper (University of Innsbruck) Nico Steytler (University of Western Cape) Petra Roter (University of Ljubljana) Joshua Casellino (Middlesex University) Stefan Oeter (University of Hamburg) Ilze Brands-Kehris (Director hcnm Office, The Hague)

volume 10

The titles published in this series are listed at brill.com/tcdg

Comparing Fiscal Federalism Edited by

Alice Valdesalici Francesco Palermo Associate Editor

Annika Kress

leiden | boston

Library of Congress Cataloging-in-Publication Data Names: Valdesalici, Alice, editor. | Palermo, Francesco, editor. Title: Comparing fiscal federalism / edited by Alice Valdesalici, Francesco Palermo ; associate editor, Annika Kress. Description: Leiden ; Boston : Brill, [2018] | Series: Studies in territorial and cultural diversity governance, issn 2213-2570 ; Volume 10 | Includes bibliographical references and index. Identifiers: lccn 2017057291 (print) | lccn 2017059298 (ebook) | isbn 9789004340954 (e-book) | isbn 9789004340930 (hardback : alk. paper) Subjects: lcsh: Intergovernmental fiscal relations. Classification: lcc hj197 (ebook) | lcc hj197 .c656 2018 (print) | ddc 336--dc23 lc record available at https://lccn.loc.gov/2017057291

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface. issn 2213-2570 isbn 978-90-04-34093-0 (hardback) isbn 978-90-04-34095-4 (e-book) Copyright 2018 by Koninklijke Brill nv, Leiden, The Netherlands. Koninklijke Brill nv incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi, Brill Sense and Hotei Publishing. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, ma 01923, usa. Fees are subject to change. This book is printed on acid-free paper and produced in a sustainable manner.

Contents Preface ix List of Acronyms xi Notes on Contributors xiv Introduction: Methodological Approach and Structure of This Book 1 Alice Valdesalici and Francesco Palermo

Part 1 Framework and Principles 1 Defining Fiscal Federalism 11 Alice Valdesalici 2 From a Formal to a Substantial Approach: Sources of Law and Fiscal Federalism 22 Sara Parolari 3 Financial Autonomy vs. Solidarity: A Dialogue between Two Complementary Opposites 40 Cheryl Saunders 4 The Practicalities of Economic Federalism: A Critical Review of How to Apply the Lessons of Fiscal Autonomy in Practice 60 Andrew Hughes Hallett 5 The Principles of Separation and Correspondence, the Comparative Method, and the Problem of Semantic Change 83 Matteo Nicolini

Part 2 Foundations Section 1 The Distribution of Powers

vi

Contents

6

Accountability and Revenue Assignment across Levels of Government: Rules, Practices, and Challenges 119 Maria Flavia Ambrosanio, Paolo Balduzzi, and Claudia Peiti

7

Taxing Powers of Subnational Entities: Between Domestic and Supranational Constraints 149 Gisela Färber

8

Can Lessons from Equalisation Transfers in Industrial Countries be Applied to Reforms in Emerging-Market Countries? 169 Ehtisham Ahmad and Giorgio Brosio

9

Fiscal Decentralisation and Decentralising Tax Administration: Different Questions, Different Answers 190 Richard M. Bird

Section 2 Intergovernmental Financial Relations 10

Intergovernmental Financial Relations: Institutions, Rules, and Praxis 223 Elisabeth Alber

11

Accommodating Diversity While Guaranteeing Stability: The Role of Financial Arrangements 274 Annika Kress

Part 3 New Perspectives on Fiscal Federalism 12

Local Governments in African Federal and Devolved Systems of Government: The Struggle for a Balance between Financial and Fiscal Autonomy and Discipline 299 Nico Steytler and Zemelak Ayitenew Ayele

13

Fiscal Sovereignty in a Globalised World: The Pressure of European Economic Governance on Domestic Public Finance 328 Jan Schnellenbach

C ontents

14

Fiscal Federalism in Times of Crisis: An Iron Law of Centralisation? 347 Karl Kössler and Martina Trettel

15

Comparative Research and Fiscal Federalism 376 Ronald L. Watts

16

A Post Scriptum to Ron Watts: The Trajectory of Fiscal Federalism 382 Francesco Palermo Index 387

vii

Preface The subject of Fiscal Federalism is undergoing an intense, worldwide discussion. Despite its strategic relevance in the political debate, there is, on the whole, insufficient knowledge of, and agreement with regard to, the very concept, its principles, and its consequences. This sometimes leads to an ­overemphasis of political considerations instead of taking into account the socio-­economic consequences of choices. This risk affects central institutions, as well as both the subnational and local levels. This book on comparative fiscal federalism is linked to the research area of the Institute for Comparative Federalism of Eurac Research on financial relations and fiscal federalism. The Institute’s studies, consultancies, and publications focus on the distribution of powers and intergovernmental relations in financial and fiscal matters in federal systems, adopting a comparative approach and predominantly following a legal-constitutional perspective. Against this background, the major challenge of this volume was to bring together scholars from different disciplines and to promote, through their interaction, interdisciplinary insight in the field. In fact, one of the limits to a shared perspective on financial relations in federal systems is represented by the deeply divided disciplinary approaches that still exist on the same matter. The aim of this book is to overcome some of these differences by looking at problems and solutions, thereby transcending disciplinary differences through an issue-oriented approach. It is up to the community of scholars and practitioners to determine whether and to what extent this effort has been successful. It is with the utmost sadness that we note the passing of the author of one of the chapters, Ronald L. Watts, on 9 October 2015. Ronald L. Watts was Professor Emeritus at Queen’s University, Fellow and Director of the Institute of Intergovernmental Relations at Queen’s University, and a Board Member and Chairman of the Research Committee of the Institute for Research on Public Policy. For years, he was—to quote the former President, Nico Steytler—“the heart and soul of the International Association of Centres for Federal Studies”. He was one of the founding members of the Association in 1977 and its president from 1992 to 1997. On several occasions, he was a consultant to the Government of Canada during constitutional deliberations, most notably in 1980–81 and 1991–92, and he was also a constitutional advisor to governments in several other countries. Besides being one of the most prominent scholars in the field of comparative federalism, he was also a wonderful and inspiring person, and was always an example to all of us. We will never forget how warmly he encouraged new generations of federalism scholars and how seriously he

x

Preface

took the work of others. We took the liberty of expanding on his concluding remarks, as they are still perfectly suited to the current and future challenges of comparative fiscal federal studies. We are grateful to all the authors for their commitment and cooperation in bringing this book together. It is our hope that this volume can provide a useful contribution to the debate on one of the most important and more tangentially studied aspects of (comparative) federalism and that it will inspire further studies in this field. Alice Valdesalici and Francesco Palermo Bolzano/Bozen, August 2017

List of Acronyms ac Autonomous Community (Spain) acir Advisory Commission on Intergovernmental Relations (usa) aeat Agencia Estatal de Administración Tributaria (State Tax Agency; Spain) AIReF  Autoridad Independiente de Responsabilidad Fiscal (Independent Authority for Fiscal Responsibility; Spain) anc African National Congress arra American Recovery and Reinvestment Act (usa) Base Erosion and Profit Shifting beps BVerfG Bundesverfassungsgericht (Federal Constitutional Court; Germany) BVerfGE Decisions of the BVerfG (Germany) b-vg Bundes-Verfassungsgesetz (Federal Constitutional Law; Austria) ca Constitution Act (Canada) cbo Congressional Budget Office (usa) cbsa Canada Border Security Agency ccf Canadian Council of Federation cet  Contribution Economique Territoriale (Territorial Economic Contribution; France) cfe Cotisation Foncière des Entreprises (Local Business Property Tax; France) cgc Commonwealth Grants Commission (Australia) cit Corporate Income Tax coag Council of Australian Governments CoB Controller of the Budget (Kenya) cpff  Consejo de Política Fiscal y Financiera (Council for Fiscal and Financial Policy; Spain) cra Commission on Revenue Allocation (Kenya) cra Canada Revenue Agency css Centrally Sponsored Schemes (India) cvae  Cotisation sur la Valeur Ajoutée des Entreprises (Local Value Added Tax for Businesses; France) ec European Commission ecb European Central Bank efsf European Financial Stability Facility efsm European Financial Stabilisation Mechanism emu European Monetary Union eprdf Ethiopian People’s Revolutionary Democratic Front esm European Stability Mechanism eu ets European Union Emissions Trading System

xii

List of Acronyms

fag  Finanzausgleichsgesetz (Financial Equalisation Law; Austria & Germany) fc Finance Commission (India) fdre Federal Democratic Republic of Ethiopia ffc South African Finance and Fiscal Commission fmm First Ministers’ Meeting (Canada) f-vg Finanz-Verfassungsgesetz (Financial Constitutional Law; Austria) gdp Gross Domestic Product gfs Government Finance Statistics gg Grundgesetz (Fundamental Law; Germany) gra General Revenue Assistance (Australia) gst Goods and Services Tax hst Harmonized State Tax iaffr  Intergovernmental Agreement on Federal Financial Relations (Australia) IGfR Intergovernmental Financial Relations ifi Independent Fiscal Institution imf International Monetary Fund imu Imposta Municipale Unica (Property Tax; Italy) irap  Imposta regionale sulle attività produttive (Regional Tax on Production; Italy) iuc Imposta Unica Comunale (Unique Municipal Tax; Italy) jcpaa Joint Committee for Public Accounts and Audit (Australia) jcpc Judicial Committee of the Privy Council kanu Kenya African National Union lofca  Ley Orgánica de Financiación de la Comunidades Autónomas (Organic Law on the Financing of the Autonomous Communities; Spain) ltb Local Tax Bureau mec Members of the Executive Council (South Africa) mtbr Marginal tax back rate mtef Medium-Term Expenditure Framework (South Africa) MTO Medium Term Objective npp National Partnership Payment nspp National Specific Purpose Payments (Australia) obr Office for Budget Responsibility (uk) oecd Organisation for Economic Co-operation and Development omt Outright Monetary Transaction (eu) osag Office of the State Auditor-General (Nigeria) pbo Parliamentary Budget Officer pc Planning Commission (India)

List of Acronyms

xiii

pit Personal Income Tax rmafc Revenue Mobilisation Allocation and Fiscal Commission (Nigeria) rts Representative Tax System sat Servicio de Administración Tributaria sc Stability Council (Stabilitätsrat; Germany) sepdm Southern Ethiopia Democratic Movement sfsf State Fiscal Stabilization Fund (usa) sgp Stability and Growth Pact (eu) sjlga State Joint Local Government Account (Nigeria) smdup Shako-Mejenger Democratic Unity Party (Ethiopia) sng Subnational Government snnp Southern Nations Nationalities and Peoples’ State (Ethiopia) spg Special Purpose Grant stb State Tax Bureau sufa Social Union Framework Agreement (Canada) tari Tassa sui rifuti (Garbage Tax; Italy) tasi Tributo per i servizi indivisibili (Duty for Indivisible Services; Italy) teu Treaty on European Union ts Tigray State (Ethiopia) tsc Treasury Selection Committee (uk) tscg Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (eu) usd United States Dollar vat Value-added Tax

Notes on Contributors Ehtisham Ahmad is a Visiting Senior Fellow of the lse Asia Research Centre. He is also a Senior Fellow at the Center for Development Research (zef), University of Bonn. Dr. Ahmad has held senior positions over the past two decades at the International Monetary Fund (Senior Advisor, Office of Executive Directors; Advisor and Division Chief, Fiscal Affairs Department); and was also a member of the core team for the World Bank’s 1990 World Development Report on Poverty. Elisabeth Alber is Senior Researcher, Project Leader (Federal Scholar in Residence Program) and Group Leader for Training and Outreach Activities at the Institute for Comparative Federalism of Eurac Research, Bolzano/Bozen, has been trained in International Sciences and Diplomacy at the Universities of Turin (Italy) and Turku (Finland). After working experiences at the European Center for Minority Issues (Germany) and the European Union (Belgium) she joined Eurac Research in 2006. Her publications, research and teaching focus on minority rights, territorial autonomies, comparative federalism and regionalism as well as participatory democracy in compound States. Since 1999, she is also workshop facilitator and consultant of intercultural activities as well as democracy and federalism programs, both in Europe and overseas. Maria Flavia Ambrosanio is Professor of Public Economics at the Catholic University of the Sacred Heart in Milan. Her research interests are mainly concentrated on Fiscal federalism, Local finance, International taxation and Fiscal competition. In addition to many publications in these fields, she authored several chapters on international handbooks. She is member of Cifrel (Inter-University Center for the study of local public finance) and worked as a consultant for several regional and national institutions. Zemelak Ayitenew Ayele is Assistant Professor at the Centre for Federal Studies of Addis Ababa University. He was formerly a National Research Foundation (nrf) Postdoctoral researcher under the South African Research Chair in (SARCHi) Multilevel Government, Law and Policy, at the Dullah Omar Institute for Constitutional Law, Governance and Human Rights, University of the Western

Notes on Contributors

xv

Cape (uwc). He has widely published in the area of local government and federalism in Ethiopia and Africa. Paolo Balduzzi is Assistant Professor of Public Economics at the Catholic University of the Sacred Heart in Milan. His main research interests include Political economy, Fiscal federalism, Local finance, Pensions, Intergenerational inequality, and Public services. He published on peer reviewed national and international academic journals (e.g: Economics & Politics, Journal of Economics, The Italian Journal of Public Economics) and authored several chapters on international text and handbooks. He is member of Cifrel (Inter-University Center for the study of local public finance) and secretary of Italents, a think-tank devoted to the monitoring of new migrations. Richard M. Bird is Professor Emeritus of Economics, Rotman School of Management, and Senior Fellow of the Institute for Municipal Governance and Finance, Munk School of Global Affairs, University of Toronto. He has been a visiting professor in the United States, the Netherlands, Australia, Japan, and India, a consultant to many national and international organizations, and has published extensively on taxation and fiscal issues, particularly in developing countries. His most recent books are Taxation and Development: The Weakest Link? (Edward Elgar, 2014) and Is Your City Healthy? Measuring Urban Fiscal Health (Institute of Public Administration of Canada, Toronto, 2015). Giorgio Brosio is Professor Emeritus at the University of Torino (Italy) and former President of Italian Society of Public Economics. He works as a consultant for several international organization such as the International Monetary Fund, the World Bank, the European Union, the Inter-American Development Bank, the Asian Development Bank, or the Economic Commission for Latin America and the Caribbean. Gisela Färber is Professor of Public Finance at the German University of Administrative Science Speyer, Germany and head of the research section “The State and Administration in Multilevel Politics” at the German Institute for Public Administration (föv). She is and has been member of various expert and parliamentary commissions, as well as advisory councils of federal and states’ governments in Germany and internationally.

xvi

Notes on Contributors

Andrew Hughes Hallett is University Professor of Economics and Public Policy at George Mason University; Professor of Economics, Copenhagen Business School; and fellow of the Royal Society of Edinburgh. He has been Professor of Economics at Princeton University; Bundesbank Professor at the University of Berlin; visiting professor at Harvard, Frankfurt, Rome and Paris Universities. He specialises in the theory of economic policy; macroeconomics; policy coordination; fiscal and monetary policies; federalism; and institutional design. He was a member of the Scottish Council of Economic Advisors from 2007; Scottish Fiscal Commission; Scottish Growth Commission; and consultant to the European Central Bank; European Commission; World Bank; imf; European Parliament. Karl Kössler is Senior Researcher at the Institute for Comparative Federalism of Eurac Research, Bolzano/Bozen (Italy) with a background in both comparative constitutional law and political science. His main fields of interest and expertise are comparative federalism and autonomy studies and, more broadly, constitutional design in diverse societies. He has provided consultancy to the Council of Europe and to various national and subnational governments. Dr. Kössler has also lectured on the above-mentioned subjects in Europe and beyond both at universities and in master programmes targeted at post-docs and civil servants. He is co-author of the book Comparative Federalism: Constitutional Arrangements and Case Law (Oxford: Hart Publishing, 2017). Annika Kress is Researcher at the Institute for Comparative Federalism of Eurac Research, Bolzano/Bozen with a background in Sociology of Law. Her main research interests are institutional innovation and participatory democracy, comparative federalism and comparative constitutionalism. Matteo Nicolini is Assistant Professor of Public Comparative Law in the Department of Law at the University of Verona. His fields of research include public comparative law, Italian and European constitutional law, federalism and regionalism, judicial review of legislation, law and literature, African law, legal geography. He is the author of numerous publications in Italian, Spanish, and English. His monographs include Partecipazione regionale e «norme di procedura». Profili di diritto costituzionale italiano ed europeo (Naples: Edizioni Scientifiche Italiane, 2009); (with Francesco Palermo) Il bicameralismo. Pluralismo e limiti della rappresentanza in prospettiva comparata (Naples: Edizioni Scientifiche

Notes on Contributors

xvii

Italiane, 2013); L’altra Law of the Land. La famiglia giuridica “mista” dell’Africa australe (Bononia: bup, 2016). He is also the editor of several collections, including (with Francesco Palermo and Enrico Milano) Law, Territory and Conflict Resolution: Law as a Problem and Law as a Solution (The Hague: BrillMartinus Nijhoff, 2016). Francesco Palermo is Professor for Comparative Constitutional Law in the Faculty of Law, University of Verona and Director of the Institute for Comparative Federalism at Eurac Research. He was the President of the Council of Europe’s Advisory Committee on the Framework Convention for the Protection of National Minorities, is Full Member of the Group of Independent Experts on the Council of Europe’s Charter for Local Self-Government and since 2016 President of the International Association of Centers for Federal Studies (iacfs). He has been teaching in several universities in Europe and overseas and is the author of more than 200 academic publications in several languages and editor of three book series. His main fields of expertise are comparative federalism, minority rights, comparative constitutional law, European integration, judicial review, constitutional transitions and legal terminology. Sara Parolari Ph.D. (2007), University of Trento, is Senior Researcher at the Institute for Comparative Federalism of Eurac Research. Her research and publications mainly focus on comparative, Italian and European constitutional law, federal and regional studies, devolution issues, fiscal federalism. She authored several articles on these topics in academic journals as well as chapters on international handbooks. She has provided consultancy to the Council of Europe and to regional and local governments in Italy. Claudia Peiti is Senior Economist at Ref Ricerche. She graduated in Economics and International Finance at the Catholic University of the Sacred Heart in Milan. She focused her further research endeavors in Public Finance, initially working for the Department of Economics and Finance of the Catholic University and then joining for two years the Public Finance Unit at the Budget Office of the Italian Senate. In July 2010, she obtained her Master’s degree in International Trade, Finance and Development at the University Pompeu Fabra of Barcelona. Since September 2010, she works in Ref Ricerche, an independent research centre in Economics. Her main research interest include Local Public Finance, with special attention on Fiscal Federalism related topics.

xviii

Notes on Contributors

Cheryl Saunders is Laureate Professor Emeritus at Melbourne Law School. She is a former President of both the International Association of Constitutional Law and the International Association of Centers for Federal Studies. She has specialist research interests in comparative constitutional law, method and theory, including comparative federalism. She is a senior technical advisor to the constitution building program of International idea. Jan Schnellenbach is full professor for Microeconomics at the Brandenburg University of Technology at Cottbus. Before moving to Cottbus, he has worked in research and teaching positions at the University of Marburg and the University of Heidelberg, and has graduated with a PhD in economics from the University of St. Gallen. His research interests include political economics, the economic analysis of federal systems, and applied behavioral economics. His research is published in renowned journals such as the European Economic Review, the Journal of Economic Behavior and Organization and the European Journal of Political Economy. Nico Steytler is the South African Research Chair in Multilevel Government, Law and Policy the Dullah Omar Institute of Constitutional Law, Governance and Human Rights of the University of the Western Cape, Cape Town, South Africa. He was a technical advisor to the Constitutional Assembly drafting the 1996 South African Constitution as well as a technical advisor to the Western Cape Provincial Legislature on the drafting of a provincial constitution (1996–1997). He was a member of the Municipal Demarcation Board (2004–2014) and is a commissioner of the Financial and Fiscal Commission (2013–2018). Internationally, he has advised on multilevel government in Kenya, the Democratic Republic of Congo, Ethiopia, Zimbabwe, Libya, Nepal, and Sri Lanka, and was a un expert adviser to the Yemeni Constitutional Drafting Committee (2014). He was the president of the International Association of Centre for Federal Studies (2010–2016). Martina Trettel is Senior Researcher at the Institute for Comparative Federalism of Eurac Research. She got her PhD in Comparative Public Law at the University of Verona in May 2017. Her main research interests are Institutional Innovation

Notes on Contributors

xix

and participatory democracy, Comparative fiscal federalism and Comparative Constitutional Justice. Alice Valdesalici is Senior Researcher at the Institute for Institute for Comparative Federalism of Eurac Research and adjunct Lecturer in Comparative Public Law at the University of Verona. She got her PhD in Italian and European Constitutional Law, together with the title of Doctor Europaeus, at the Graduate School of Law of the University of Verona in May 2016. She has been visiting scholar at the Instituts d’Estudis Autonòmics in Barcelona (May–June 2013; May–June 2014) and at the Deutsche Forschungsinstitut für öffentliche Verwaltung Speyer (March 2013). Her main research interests are Comparative Fiscal Federalism, Federal and Regional Studies, Italian and European Constitutional Law, Institutional Innovation. She authored several articles on these topics in academic journals as well as edited books. She has provided consultancies to the Council of Europe and to regional and local governments in Italy. Ronald L. Watts was Professor Emeritus at Queen’s University, Fellow and Director of the Institute of Intergovernmental Relations at Queen’s University, and a Board Member and Chairman of the Research Committee of the Institute for Research on Public Policy. He was one of the founding members of the International Association of Centres for Federal Studies (iacfs) in 1977 and its president from 1992 to 1997. On several occasions, he was a consultant to the Government of Canada during constitutional deliberations, most notably in 1980–81 and 1991–92, and he was also a constitutional advisor to governments in several other countries.

Introduction: Methodological Approach and Structure of this Book This book takes a broad and interdisciplinary approach to the notion of fiscal federalism, investigating the current de jure and de facto allocations of financial and fiscal powers, as well as the understanding of the functioning of financial relations among different levels of government. With this purpose in mind, the analysis presented in each individual chapter combines a theoretical and a case-study approach and involves scholars from different disciplines in order to provide a comprehensive analysis of developments and trends in fiscal federalism. The research method is based on comparative investigation. In particular, the authors of the various chapters deal with two or more case studies selected from different geographical areas.1 As the book adopts an extensive definition of fiscal federalism, the choice was not to request a set number of case studies.2 The basic idea was to include two or more federal systems3 from two or more 1 The geographical areas are drawn from the classification provided by J. Loughlin et al., Routledge Handbook of Regionalism and Federalism (Abingdon: Routledge, 2013). For the case studies, we were inspired by the list included in G. Anderson, Federalism: An Introduction (Toronto: Oxford University Press, 2008), adding Italy, Nepal, and the United Kingdom were added. 2 However, the following list has been offered as an indication of possible case studies divided by geographical areas: (a) Africa: Comoros, Democratic Republic of Congo, Ethiopia, Nigeria, South Africa, Sudan; (b) Asia: India, Indonesia, Nepal, Malaysia; (c) Asia-Pacific: Australia, Belau, Federated States of Micronesia; (d) Central and Latin America: Argentina, Brazil, Mexico, St. Kitts and Nevis, Venezuela; (e) Europe: Austria, Belgium, Bosnia and Herzegovina, Germany, Italy, Spain, Switzerland, the United Kingdom; (f) Middle East: Iraq, Pakistan, United Arab Emirates; (g) North America: Canada, United States of America; (h) the Russian Federation. 3 The notion of a “federal system” is used according to the definition provided in F. Palermo and K. Kössler, Comparative Federalism: Constitutional Arrangements and Case Law (Oxford: Hart Publishing, 2017), 8–9, to include “all those systems in which at least two political tiers of government exist, thereby combining self-rule and shared rule and thus making use (to a greater or lesser extent) of the federal toolkit”. It thus applies “not only to denominate fully fledged federal states (synonymously called federations). It also includes what are often referred to as regional or devolved states”, as well as compound or federal-type states. “The latter are federal systems in legal terms, as they are—like consolidated federal countries—constitutionally unitary states, subordinated to the national constitution and with constitutionally entrenched political autonomy of the subnational units. Notwithstanding the use of federal system and federal country as umbrella terms, the differentiated notions federal or regional © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_002

2

Introduction

geographical areas in every single chapter in order to demonstrate, as much as possible, ongoing trends on a worldwide scale for each topic. Furthermore, the comparison can include both mature and emergent federations, federal or federal-type states, according to the choice not to adopt a formal understanding of fiscal federalism in particular or of federalism in general. As a consequence, all types of federal systems are potential case studies regardless of their formal denomination as hybrids, quasi-federations, regional or devolved states, or federacies. This openness is also a consequence of the interdisciplinary approach, as different disciplines could entail diverse classifications of the same cases. Along the path paved by these preliminary methodological guidelines, the book is then structured in three parts: “Framework and Principles”, “Foundations”, and “New Perspectives on Fiscal Federalism”. Part 1, “Framework and Principles”, outlines the overall legal frameworks in the countries studied and examines their basic principles. At the outset, in Chapter 1, “Defining Fiscal Federalism”, Alice Valdesalici explores the evolution of the concept of fiscal federalism and provides a working definition for the purposes of this book. The legal frameworks that regulate financial regimes and financial relations from a formal point of view are explored in Chapter 2, “From a Formal to a Substantial Approach: Sources of Law and Fiscal Federalism”. This theme epitomises a crucial element when it comes to financial and fiscal matters, although it is often neglected in the literature. The sources of law in this field present many peculiarities in comparison with other areas of regulation. An investigation by Sara Parolari looks at different systems belonging to different legal families (common law and civil law) and speculates about how a legal system’s form can influence its substance. Once financial constitutions are portrayed in their formalistic sense, the focus moves on to the principles at stake. Although there is no one-size-fits-all model of fiscal federalism, every system can be considered a balance of a common pool of fundamental rules. Emblematic in this respect is the co-existence or devolved states are used when specific countries are meant or when a distinction between these notions needs to be made. For example, even if both Spain and Germany are, using our terminology, federal systems, the former is a regional state and the latter a federal state”. Furthermore, in general, this book uses the term state to mean a country as a whole (frequently associated with the adjectives unitary and federal); however, it can refer to a subnational entity when writing about a country that uses this specific terminology (e.g. states in the United States, India, Nigeria, etc.).

Introduction

3

of autonomy and solidarity in financial relations. This is the issue addressed in Chapter 3, “Financial Autonomy vs. Solidarity: A Dialogue between Two Complementary Opposites”. With the aim of comparing different approaches to the design and functioning of financial relations in federal-type systems by reference to the balance between the two above-mentioned principles, the author, Cheryl Saunders, integrates sound theoretical argumentation with an effective investigation of case studies, including two examples from different geographical regions, different systems of law, and different constitutional traditions. The case for fiscal decentralisation is presented in Chapter 4, “The Practicalities of Economic Federalism: A Critical Review of How to Apply the L­ essons of Fiscal Autonomy in Practice”, by applying the distinction between instrument independence and target independence to fiscal policies in federal systems. Providing an overall picture of economic theories on fiscal federalism, the author, Andrew Hughes Hallett, shows that there are distinct advantages and shortcomings in vesting subnational entities with revenue responsibilities. In making the case for tax autonomy as an enhancer of both efficiency and political accountability, the author highlights the importance of intergovernmental coordination and debt control mechanisms to realise the gains of fiscal autonomy. Another recurrent trait d’union is the principle of correspondence, which is the subject of Chapter 5, “The Principles of Separation and Correspondence, the Comparative Method, and the Problem of Semantic Change”. The author, Matteo Nicolini, explores the various linguistic variations and legal enactments of the principle. In referring to a number of case studies, Nicolini, a legal scholar, outlines the profound dissimilarities in terms of both legal provisions and operational structures, ascertaining the shift of the classical paradigm that “revenue follows functions”. Part 2, “Foundations”, is then devoted to the design of the fundamental principles of fiscal federalism through a combination of the current constitutional-­ legislative settings (as they result from constitutional adjudication) and an ­exploration of the functioning of financial systems. According to the proposed working definition,4 these include two sub-themes: a typical topic of federal studies, i.e. the distribution of powers (Section 1), and intergovernmental financial relations (Section 2). Regarding Section 1, “The Distribution of Powers”, the topics assigned to the individual chapters were selected from a commonly applied classification in fiscal federalism studies, with each chapter taking into consideration both the revenue structure and a functional perspective. As to the former, this 4 See A. Valdesalici’s chapter, “Defining Fiscal Federalism”, in this volume.

4

Introduction

typically comprises the distribution of revenue powers, which combines the power to tax (e.g. tax-base sharing) and the power over revenue (i.e. taxrevenue sharing), as well as the intergovernmental transfers system. As to the latter, the functional perspective contemplates the allocation of different public functions to the different levels of government, paying particular attention to the legislative and the administrative power related to tax and financial issues.5 In particular, the book focuses on the so-called revenue assignment problem. The general issue is addressed in Chapter 6, “Accountability and Revenue Assignment across Levels of Government: Rules, Practices, and Challenges”. The authors, Maria Flavia Ambrosanio, Paolo Balduzzi, and Claudia Peiti, explore different practices of revenue assignment in a set of oecd countries against a review of the literature dealing with the link between subnational governments’ accountability and sources of subnational financing. Adopting this standpoint, instruments of, and procedures for, revenue distribution across and within the different layers of government are of vital importance. While ensuring that subnational entities cover their spending needs, they determine the extent to which subnational entities have been made responsible for their financing. Chapter 7, “Taxing Powers of Subnational Entities: Between Domestic and Supranational Constraints”, is then devoted to the tax autonomy of subnational jurisdictions. The author, Gisela Färber, provides a thorough examination of fiscal decentralisation in federal systems, combining a theoretical analysis with a case-study approach. After emphasising the rationale and the main aims that lead to the distribution of taxation powers, the author explores the relationship between tax competition in a globalised world and decentralised taxation competencies, highlighting, in particular, the resulting problems for subnational governments. Furthermore, the chapter’s spectrum of analysis ­includes the impact of supranational institutions and the regulations they impose, focusing, in particular, on the constraints on decentralised taxation powers that eu treaties can lead to. 5 In this regard, German scholars of constitutional law use the expression Finanzhoheit (financial power or sovereignty) to refer to state functions in this field, and they include not only legislative, administrative, and jurisdictional functions, but also power over revenue and spending power as typical additional functions concerning financial relations in federal-type states. See K. Stern, Das Staatsrecht der Bundesrepublik Deutschland (München: C.H. Beck, 1980) 1089. With specific reference to spending responsibilities, see C. Gröpl, “Finanzautonomie und Finanzverflechtung in gestuften Rechtsordnungen”, DVBl—Deutsches Verwaltungsblatt (2006) 1080.

Introduction

5

The other side of the coin is represented by intergovernmental transfers, including equalisation transfers. Horizontal imbalances are a recurrent feature of all state systems, and this holds true particularly when it comes to federaltype systems. A certain degree of differentiation is inherent to the concept of federalism; however, it has to be kept within certain limits for the sake of cohesion and unity. Chapter 8, “Can Lessons from Equalisation Transfers in Industrial Countries Be Applied to Reforms in Emerging-Market Countries?” is devoted to this complex issue. With the aim of drawing lessons from existing practices in industrial countries for emerging-market countries, the authors, Ehtisham Ahmad and Giorgio Brosio, focus on the rationale, the origin, and the evolution of national equalisation systems in a group of oecd countries, both unitary and federal, and then move on to equalisation systems in a selection of emerging-market economies, providing some policy reform suggestions as a conclusion. While one part of taxing authority concerns the legislative power to tax, another part consists of tax administration. In this respect, attention is paid to the extent to which the administration of tax collection should be decentralised. In general, one can assume that collection powers follow the authority to regulate a tax itself. At the same time, however, there are highly fiscally decentralised states in which there is one single central tax administration authority, and others that are fiscally centralised in which each government has its own authority. Chapter 9, “Fiscal Decentralisation and Decentralising Tax Administration: Different Questions, Different Answers”, is dedicated to this matter. After dealing with the question of how one may characterise a tax as being decentralised, the author, Richard M. Bird, examines a variety of case studies and provides key arguments for and against decentralising the administration thereof. Regarding Section 2, “Intergovernmental Financial Relations”, two central issues were selected that are gaining more and more momentum in federalism studies. The first is dealt with in Chapter 10, “Intergovernmental Financial Relations: Institutions, Rules, and Praxis”. The author, Elisabeth Alber, scrutinises the constitutional set-up of different institutions of intergovernmental cooperation such as finance commissions or fiscal watchdogs. The investigation of the institutional dimension is a crucial issue, in particular due to the predominance of cooperative paradigms in financial relations worldwide. Looking at their role, composition, and functioning when it comes to financial decision-making processes is crucial for understanding the scope of subnational autonomy. The second topic has always been central to federalism studies, but is becoming more and more important in times of fiscal distress. Chapter 11,

6

Introduction

“Accommodating Diversity While Guaranteeing Stability: The Role of Financial Arrangements”, investigates asymmetrical fiscal arrangements and relations in a number of test cases. The author, Annika Kress, takes the consolidated distinction between de facto and de jure asymmetries as a point of departure and provides an analysis of existing cases, giving evidence of the major elements of differentiation from a comparative perspective in order to uncover trends of how financial systems are adapted in order to accommodate subnational diversity. Finally, Part 3 of the book, “New Perspectives on Fiscal Federalism”,6 focuses on a selection of emerging issues in financial and fiscal relations, making reference to both the national and the supranational arena. As to the domestic dimension, it is worth noting that fiscal federalism studies do not typically address the local level of government, as they consider it to be a subordinated tier.7 In this respect, for instance, legal scholars emphasise the lack of legislative powers, something that they consider to be an essential feature of genuine tax autonomy. Notwithstanding these political and legal circumstances, the decision was made to include a chapter on local government, as the local tier is becoming a more and more important player in financial relations. Without disregarding or denying the fundamental differentiation between local entities and constituent units, there is a clear need to take this additional tier of government into consideration. Although the role of local governments can vary significantly depending on the system of reference, the municipal level is generally assigned responsibilities with regard to important public functions and services that have a major impact on citizens’ living conditions. As such, the local factor is pivotal in understanding how federal systems function. Based on this assumption, Chapter 12, “Local Governments in African Federal and Devolved Systems of Government: The Struggle for a Balance between Financial and Fiscal Autonomy and Discipline”, scrutinises the fiscal and financial autonomy of local governments in Africa’s four most important federal-type countries: Ethiopia, Kenya, Nigeria, and South Africa. With this in mind, the authors, Nico Steytler and Zemelak Ayele, discuss the existence of a link between the extent of constitutional recognition and local governments’ 6 Quoting V. Tanzi, “The Future of Fiscal Federalism”, European Journal of Political Economy, 24 (2008) 705–712, at 711. 7 See N. Steytler, “Comparative Conclusion”, in N. Steytler (ed.), Local Government and Metropolitan Regions in Federal Systems (Montreal: McGill–Queen’s University Press, 2009) 393– 436, at 435.

Introduction

7

degree of political autonomy and emphasise the key role of resources and spending discretion in this respect. In doing so, the chapter examines whether any patterns emerge from the four above-mentioned African countries. A broad understanding of fiscal federalism calls, at the same time, for an openness to the supranational dimension and for an exploration of global problems (e.g. economic crises) and actors as key determinants of financial relations at not only the international but also the national and subnational level. Needless to say, the European system of economic governance represents an emblematic example in this respect, taking into consideration the impact it has on both the national and subnational layer in terms of financial and political autonomy. The effects of such a system on national public finance are thoroughly explored in Chapter 13, “Fiscal Sovereignty in a Globalised World: The Pressure of European Economic Governance on Domestic Public Finance”. The author, Jan Schnellenbach, paints a picture of such a system, reviewing its rationale, illustrating the adjustments adopted, and discussing its functioning, as well as possible alternative solutions, with an eye on the impact of eu constraints on the fiscal sovereignty of nation states. The topic is then examined from the standpoint of subnational autonomy in Chapter 14, “Fiscal Federalism in Times of Crisis: An Iron Law of Centralisation?” In this respect, the authors, Karl Kössler and Martina Trettel, offer an interesting comparison that refers to different historical horizons (the current crisis vs. the Great Depression), as well as to diverse models of federations (aggregative vs. devolutionary). In this light, the aim of the chapter is to assess whether and to what extent the economic and financial crisis prompts the increasing centralisation of financial systems at the expense of subnational autonomy on both the revenue and the spending side. The added value and possible shortcomings of adopting a comparative approach are broadly emphasised in the conclusions. Chapter 15, “Comparative Research and Fiscal Federalism”, is the republication of a paper by the late distinguished scholar Ronald L. Watts, published more than 10 years ago, whose arguments and foundations are still valid today. In this respect, Watts illustrates how comparative analysis can be considered the foremost methodological approach to gaining knowledge about the phenomenon in all its facets. The chapter is followed by Chapter 16, “A Post Scriptum to Ron Watts: The Trajectory of Fiscal Federalism”. The author, Francesco Palermo, tries to provide an answer to Watts’ question concerning the pattern of decentralisation. In particular, Palermo debates the future trajectories of the federal and regional ideas and warns about the practice of fragmented reforms that limit their focus to fiscal

8

Introduction

federalism, leaving unaddressed the institutional dimension. Even though this approach raises sound criticisms in terms of the balanced regulation of intergovernmental relations, financial relations have been gaining momentum as a consequence of this trend. Alice Valdesalici and Francesco Palermo Bolzano/Bozen, August 2017

Part 1 Framework and Principles

∵

chapter 1

Defining Fiscal Federalism Alice Valdesalici 1

Ongoing Trends in Fiscal Federalism

The subject of fiscal federalism is a topical and central issue on political agendas and in academic debate due to its importance for the very existence of any federal system.1 The allocation of financial resources to each tier of government is typically a dynamic process that frequently undergoes formal and informal changes. Changes to this process can shift powers and responsibilities in favour of either the central or the decentralised tiers of government. Indeed, both centralisation (or re-centralisation) and decentralisation can occur simultaneously in this area. This holds particularly true against the backdrop of the ongoing economic and financial crisis. Countries’ constitutional design and institutional settings in fiscal and financial matters continue being revised in order to better adapt to current challenges that result from both the supranational and domestic spheres of economic governance. All in all, financial stability has to be achieved while boosting economic performance. Federal systems achieve this either by adjusting existing legal frameworks or by attempting to overhaul the entire system starting from its principles and foundations.2 In any case, the rationale behind these processes lies in the struggle between global competitiveness and autonomy claims. Multilevel systems have to come to terms with the challenges that arise from the interdependence of governments and governance actors. Who is in charge of adopting decisions in this field? Who is responsible for their implementation? Which tier of government is better at attributing legislative and executive powers in financial and fiscal 1 The notion of “federal system” is used according to the definition provided in “Introduction: Methodological Approach and Structure of This Book” in this volume, at fn no. 3. 2 On ongoing reform processes and debates, see H. Blöchliger and C. Vammalle, Reforming Fiscal Federalism and Local Government: Beyond the Zero-Sum Game (Paris: oecd Fiscal Federalism Studies, 2012); E. Alber and A. Valdesalici, “Reforming Fiscal Federalism in Europe: Where Does the Pendulum Swing?” L’Europe en Formation, 1 (2012) 325–366; and also, S. Ortino et al. (eds.), The Changing Faces of Federalism: Institutional Reconfiguration in Europe from East to West (Manchester: Manchester University Press, 2005).

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_003

12

Valdesalici

matters? How can a state best achieve stability and meet redistributive objectives, while remaining a competitive player in the international arena? How can it reduce public expenditure and public debt, while still allowing for an allocation of responsibilities that matches citizens’ needs and preferences best? How can it respond effectively to the increasing democratic deficit, which has a particular impact on decision-making processes in financial matters at all levels and in all contexts? The prevailing opinion holds that decentralisation of public services enhances efficiency.3 At the same time, however, the public sector is specialising and becoming more complex, while the functions performed by the various tiers of government are increasingly overlapping. Many federal systems have failed to perform these functions, as is shown by the existence of rampant public debt and the weakening of the welfare state. As a result, the need to rethink existing structures is stronger than ever. Growing pressures for change are emerging that require a flexible approach whereby territorial entities make claims for more autonomy on both the expenditure and the revenue side (at least at the margin),4 while equality and solidarity remain for the sake of territorial cohesion and unity. As this scenario suggests, there is permanent friction between supranational integration, national (fiscal) sovereignty and subnational autonomy that challenges the classical paradigms of fiscal federalism. 2

Going Back to the Origins of the Concept

In the beginning, this topic was within the purview of economic studies and of scholars of public finance, in particular, whose investigations referred to the phenomenon in terms of federal finance.5 Only later did the expression fiscal

3 At the same time, the literature warns of the need to put certain constraints on decentralisation; otherwise, efficiency gains could vanish. See W.E. Oates, Fiscal Federalism (New York: Harcourt Brace Jovanovich, 1972) 4–11. 4 On the importance of vesting subnational entities with at least a marginal power over their financing, see, among others, M.F. Ambrosiano and M. Bordignon, “Normative vs Positive Theories of Revenue Assignments in Federations”, in E. Ahmad and G. Brosio (eds.), Handbook of Fiscal Federalism (Cheltenham: Edward Elgar, 2006) 306–338; and also R. Bird, “Tax Assignment Revisited”, in J.G. Head et al. (eds.), Tax Reform in the 21. Century: A Volume in Memory of Richard Musgrave (The Hague: Kluwer Law International, 2009) 441–470, at 453. 5 In this respect, see R.A. Musgrave, The Theory of Public Finance: A Study in Public Economy (New York: McGraw-Hill, 1959).

Defining Fiscal Federalism

13

federalism appear in the literature,6 and it was referred to as the “assignment of functions to different levels of government and the appropriate fiscal instruments for carrying out these functions”.7 The wording itself expresses the strong link between fiscal federalism and the federal state. In this respect, the United States of America represented the prototype. Due to the aggregative nature of the American federation, the debate focused primarily on the need to grant certain financial independence to the federal level.8 This perspective was only later reversed, when so-called holding-together federations (e.g. Belgium and Spain) or those systems that resulted from a combination of aggregative and devolutionary processes (e.g. India and Canada) were considered. In these cases, research has concentrated mostly on the margin of autonomy conferred upon the subnational levels of government. The first generation of studies on fiscal federalism addressed the topic by essentially adopting a theoretical approach. They laid out a general normative framework that considers decision makers to be “benevolent social planners” in the way that they allocate public functions and resources and that uses a standard three-branch model to describe this allocation.9 While these models prevailed in the 1950s and 1960s, they did not actually find any correspondence in practice. First of all, there is no single system that fully reflects all of the assumptions about the assignment of functions and resources. Second, it soon became clear that decision makers could not be treated as “custodian[s] of the public interest”.10 6 7 8 9

10

Emblematic is the work of Oates, Fiscal Federalism, supra. In this sense, see W.E. Oates, “An Essay on Fiscal Federalism”, Journal of Economic Literature, 37 (1999) 1120–1149, at 1121. See “Federalist No. 30–36”, in A. Hamilton et al., The Federalist Papers (New York: McLean, 1788). This refers to macroeconomic stabilisation, redistribution and allocation. In this regard, see Musgrave, The Theory of Public Finance, supra; and the famous “decentralization theorem” elaborated by Oates, Fiscal Federalism, supra, at 35–38 and 54, as later revisited by the author in Oates, “An Essay on Fiscal Federalism”, supra, at 1122–1123. Partially in contrast with the above-mentioned approach, the theory of Public Choice—traditionally developed in usa between the 1960s and the 1970s—considers decision makers as rational actors acting in their own interests. Accordingly, voters will be inclined to support those candidates that are more in line with their own interests and politicians will adopt decisions with the primary purpose of being re-elected. Among its most prominent scholars see: G. Brennan and J.M. Buchanan, The Power to Tax: Analytical Foundations of a Fiscal Constitution (Cambridge: Cambridge University Press, 1980). W.E. Oates, “Toward a Second-Generation Theory of Fiscal Federalism”, International Tax and Public Finance, 12 (2005) 349–373, at. 350.

14

Valdesalici

It is not by chance, therefore, that the field of investigation has expanded in recent decades. The second generation of studies on fiscal federalism progressively enriched the scope of analysis by taking into consideration systems of incentives generated by political and fiscal institutions.11 The underlying rationale is that forms and procedures of vertical and horizontal revenue distribution play a role in meeting the above-outlined objectives.12 This goes back to the assumption that different taxation systems and dissimilar political structures result in diverse incentives that influence public choices in different ways.13 Many factors appear to support this more expansive approach, including the tendency in the field of economics to address problems that are not necessarily confined to the borders of a single state,14 to apply a more and more interdisciplinary approach to the topic and to separate the notion of fiscal federalism from the federal state. The first factor is linked to the growing complexity of economic processes, which implies the need to redraw jurisdictional lines and to redefine fiscal responsibilities of different levels of government.15 The second is a consequence of the changing perspective of secondgeneration­scholars. Their new point of view has proven rather successful in expanding­the field of examination, as it entails different disciplines, such as political science. One of its major attractions is that it combines the strengths and weaknesses of a variety of approaches that can provide in-depth insights into the subject matter and its critiques. Varied approaches of investigation pursue diverse objectives and, most importantly, can find different solutions to common problems. Put simply, one can assume that economists tend to put a major emphasis on efficiency and equity issues, political scientists stress the importance of responsibilisation and coordination, while public-law scholars 11

12

13

14 15

Among others, see the following studies: B.R. Weingast, “Second Generation Fiscal Federalism: Political Aspects of Decentralization and Economic Development”, World Development, 53 (2014) 14–25; B.R. Weingast, “Second Generation Fiscal Federalism: The Implications­of Fiscal Incentives”, Journal of Urban Economics, 65 (2009) 279–293. A comparison of the different theoretical approaches to the topic is offered by Oates, “Toward a Second-Generation Theory”, supra; as well as by Weingast, “Second Generation Fiscal Federalism: Political Aspects”, supra. In this respect, see Weingast, “Second Generation Fiscal Federalism: The Implications”, supra; and J.A. Rodden et al., Fiscal Decentralization and the Challenge of Hard Budget Constraints (Cambridge: mit Press, 2003). In this case, the usa, as the place in which the economic notion of fiscal federalism the first came to light in the middle of the 19th century. In this sense, see Oates, Fiscal Federalism, supra, at 9.

Defining Fiscal Federalism

15

provide a critical assessment of existing legal frameworks and their functioning in light of the fundamental constitutional principles of autonomy and solidarity.16 The emancipation of the concept from the us paradigm—encouraged by the transformation of the role of the state—has itself contributed to the enrichment of the debate. Indeed, the importance of the issue and the wide reach of the phenomenon have championed the inclusion of systems exhibiting features of both federal and unitary states (so-called hybrids) in the analysis.17 The fact that fiscal federalism is no longer limited to classical federal states but now also involves all cases characterised by any form of “power de-concentration­ on a territorial base”18 has favoured the circulation of the related theories and the flourishing of comparative studies on the topic. The latter in particular has worked effectively in attracting more and more the interest on the part of other disciplines, including that of constitutional lawyers.19 Initially, legal studies on federalism focused almost exclusively on the vertical distribution of legislative and administrative competencies and on the institutional dimension. With a few exceptions of a non-comparative nature,20 financial regimes have only become subject to comparative legal analysis in the last few decades. A change was prompted by the fact that the reform of financial systems has become more frequent since the end of the 1990s, thus attracting the attention of legal scholars. It soon became clear that this subject matter represents an 16

17

18 19

20

See J. Edwin Kee, “Fiscal Decentralization: Theory as Reform”, in A. Khan and W.B. Hildreth (eds.), Financial Management Theory in the Public Sector (Westport: Praeger, 2004) 165–186, at 165 ff.; and also S.H. Beer, “A Political Scientist’s View of Fiscal Federalism”, in W.E. Oates (ed.), The Political Economy of Fiscal Federalism (Toronto: Lexington Books, 1997) 21–46, at 21 ff., where the political science standpoint is well portrayed. In this respect, see the all-encompassing classification of different types of states offered by R.L. Watts, “Federalism, Federal Political Systems, and Federations”, Annual Review of Political Science, 1 (1998) 117–137. F. Palermo, “Stato regionale”, in L. Pegoraro (ed.), Glossario di diritto pubblico comparato (Roma: Carocci, 2009) 252–255, at 253. G.G. Carboni, “Fiscal Federalism and Comparative Law”, Comparative Law Review, 5 (2014) 1–20, at 7. All in all, studies of comparative constitutional law have flourished starting in the 21st century thanks to the increasing weight assumed by constitutional courts as decision makers on key issues of a social and economic nature. This change in their role has awoken an ever more interdisciplinary interest (e.g. legal, political sociological and economic scholars are engaged in studies in this field). In this regard, see T. Ginsburg and R. Dixon, “Introduction”, in T. Ginsburg and R. Dixon (eds.), Comparative Constitutional Law (Cheltenham: Edward Elgar, 2011) 1–15. Among others, see K.W. Dam, “The American Fiscal Constitution”, The University of Chicago Law Review, 44 (1977) 271–320, in which the author deals with the us case.

16

Valdesalici

essential condition for the functioning of any federal system on the basis of its twofold nature. On the one hand, the financial endowment of the different levels of government is instrumental for the discharge of their competencies; on the other hand, there exists a strong mutual interdependence between financial and institutional components.21 The latter, in particular, is a two-way relationship. Assuming that revenue follows functions, financial arrangements are conditioned by the institutional structure of the state as a whole. But the contrary also holds true. The allocation of powers in financial and fiscal matters is instrumental for the allocation of legislative and administrative responsibilities, as it is intended to ensure that subnational entities can make use of their constitutionally guaranteed margin of autonomy. With regard to this second aspect, financial rules have frequently impacted institutional systems and their functioning, and in some cases can be considered responsible for their formal or informal change.22 This holds particularly true if one observes the integration process that is taking place within the eu member states that belong to the Eurozone and the impact it has on both the national and subnational tiers of government.23 Finally, since the early 1990s, the field of fiscal federalism has expanded greatly in the context of developing countries and has generated rampant interest in its understandings and explanations of the respective systems.24 The idea that fiscal federalism refers in very general terms to “the public finances of the various orders of government in a federal system”25 has been crucial in this respect. Even though there is no unanimous agreement on a common definition, this broader picture outlines a common object of investigation, which involves 21 22

23

24

25

Carboni, “Fiscal Federalism and Comparative Law”, supra, at 8. In this respect, see F. Palermo, “Comparare il federalismo fiscale: cosa, come, perché”, in F. Palermo and M. Nicolini (eds.), Federalismo fiscale in Europa: Esperienze straniere e spunti per il caso italiano (Napoli: Edizioni Scientifiche Italiane, 2012) 1–11, at 2 and 10; and also, G. Brosio, Equilibri instabili: politica ed economia nell’evoluzione dei sistemi federali (Torino: Bollati Boringhieri, 1994) 146. See G. Färber’s chapter “Taxing Powers of Subnational Entities: Between Domestic and Supranational Constraints” in this volume, particularly part 4, and the literature mentioned in the footnotes. D.E. Wildasin, “Fiscal Federalism”, in S.N. Durlauf and L.E. Blume (eds.), The New Palgrave Dictionary of Economics (2nd edn., Basingstoke: Palgrave Macmillan, 2008), http://www .dictionaryofeconomics.com/article?id=pde2008_F000118 (accessed 10 January 2017). A. Shah, “Comparative Conclusions on Fiscal Federalism”, in A. Shah (ed.), The Practice of Fiscal Federalism: Comparative Perspectives (Montreal: McGill-Queen’s University Press, 2007) 370–393, at 370.

Defining Fiscal Federalism

17

an analysis of “the respective roles and interaction of governments […] with a particular focus on the raising, borrowing and spending of revenue”.26 All in all, the examination thus includes the allocation of taxing and spending powers, as well as of regulatory and revenue responsibilities. In addition, the emphasis given to both “the respective roles and interaction of governments” redirects attention beyond the mere allocation of powers and responsibilities and takes into consideration the functioning of the system by means of intergovernmental relations. This enrichment can be connected to the theory of federalism as a process27 and appears to be particularly appropriate for addressing the phenomenon and its causes. At the same time, it is shown to be indispensable for understanding how different systems actually work. Beyond stressing the importance of a dynamic approach to this field of study, this evolution of the concept of fiscal federalism results in an inclusive spectrum of analysis that makes it feasible to embrace systems that have very few traits in common: among these, the existence of two or more orders of government and a vertical distribution of powers and responsibilities. 3

Revising Fiscal Federalism

The above-illustrated evolution has progressively led to a revision of the significance and the borders of the subject of fiscal federalism. On the one hand, the literal meaning of the words is far too simplistic and misleading, while, on the other hand, the need to rethink its significance as a result of growing worldwide complexity is emerging. Own taxes of the subnational entities exemplify this phenomenon. From the very beginning, studies on fiscal federalism have focused predominantly on the distribution of taxing powers in federal systems. The so-called tax assignment problem has been one of the focal points of discussion and analysis,28 as the substance of the issue has been explored by studies of both the first 26 27

28

G. Anderson, Fiscal Federalism: A Comparative Introduction (Oxford: Oxford University Press, 2010), at 2. See C.J. Friedrich, “Federal Constitutional Theory and Emergent Proposals”, in A.W. Macmahon (ed.), Federalism: Mature and Emergent (New York: Russell & Russell, 1962), at 528 ff. The first explicit reference to “tax assignment” can be found in R.A. Musgrave, “Who Should Tax, Where and What?” in C.E. Mclure (ed.), Tax Assignment in Federal Countries (Canberra: Australian National University, 1983) 2–19.

18

Valdesalici

and second generation, as well as by scholars of all disciplines.29 Emblematic is also the interest that legal scholars pay to this aspect. At an earlier stage, in fact, a minimalistic approach to fiscal federalism prevailed, and the analysis was limited to the distribution of the power to tax and to assessments of the existing margin of tax autonomy vested in the different tiers of government.30 However, this picture was soon revealed to be unsatisfactory, as it failed to capture the actual scope of the phenomenon.31 It is undeniable that the power to tax represents the milestone of intergovernmental financial relations, but the latter cannot be reduced to this aspect only. If it were for the degree of tax autonomy—in the strict sense—vested with subnational entities, systems of fiscal federalism would hardly be found anywhere in practice. In fact, the exclusive power of subnational entities to impose and regulate their own sources of taxation appears to be a marginal feature in most federal systems. The wide distribution of the welfare model has accentuated this pattern. Calling for the redistribution of wealth, a new pattern emerged in which taxing powers became progressively centralised, although each to a different extent.32 This overall picture holds true particularly when it comes to European case studies, with the sole exception of Switzerland.33 That said, this scheme is being adopted in emerging federations as well, in which subnational financing is more the result of revenue-sharing schemes rather than tax-base sharing.34 However, a centralising trend can also be found in mature non-European federations. Although Canada represents one of the most fiscally decentralised states in the world, its intergovernmental financial relations are based more and more on tax collection agreements,35 which de facto reduce the room for manoeuvre that is vested in the provincial level. From a formal point of view, Canadian provinces are vested with significant taxing power. For instance, each province has the power to set its own personal income tax rate and to determine its tax base. Despite this, the system has become more coordinated­ 29 30 31 32 33

34 35

A review of the tax assignment problem that bridges theory and practice can be found in Bird, “Tax Assignment Revisited”, supra. Carboni, “Fiscal Federalism and Comparative Law”, supra, at 2. In this regard, see Palermo, “Comparare il federalismo fiscale”, supra, at 6–7. For details on subnational taxes, see, among others, the examination conducted by D. King, “Allocation of Taxing Powers”, oecd Journal on Budgeting, 6 (2007). With reference to European federal systems, the scope of subnational tax autonomy is examined from a comparative and legal perspective in Alber and Valdesalici, “Reforming Fiscal Federalism in Europe”, supra, at 334 ff. Nigeria, Pakistan and South Africa could be examples. B. Alarie and R.M. Bird, “Canada”, in G. Bizioli and C. Sacchetto (eds.), Tax Aspects of Fiscal Federalism (Amsterdam: ibfd, 2011) 79–137, at 107 ff.

Defining Fiscal Federalism

19

than ever before. This is partly due to the practice of signing federal-provincial agreements for tax collection. The rationale behind this system of federalprovincial­diplomacy is to increase coordination. This goal, however, is pursued­ by means of tax harmonisation at the expense of territorial differentiation and subnational autonomy.36 Furthermore, formal and informal reforms of existing financial agreements are gaining momentum, and the need to cope with problems related to the democratic deficit and economic and financial stability, combining local players with global actors, is becoming more and more apparent. As it always involves a balance of opposites, the outcome is unpredictable. Nevertheless, these trends demonstrate the failure of traditional categories to describe the phenomenon. Comparative studies of federalism are familiar with this dynamic approach, yet changing contexts require new institutional solutions and approaches to cope with emerging trends and problems. This represents one of the major challenges for studies of fiscal federalism and requires a comparative approach and an interdisciplinary perspective. 4

Towards a Working Definition

Along the path paved by the overall expansive trend characterising the object of investigation and mostly inspired by practices from all over the world, this book adopts a broad conception of fiscal federalism. In this respect, the old minimalist meaning proves not to be very useful in understanding the contemporary world and its emerging issues. It was able to provide only far too partial insights that were limited to a handful of so-called classical federations, but it was unable to grasp the dynamics of today’s reality. However, an extended point of view has to cope with the lack of a global theory: as with the general concept of federalism, there is no one-size-fits-all definition of fiscal federalism. To some degree, the notion of financial constitution could be considered a common point of reference. The term is a literal t­ ranslation of

36

This is the case of Australia and the usa. Another example could be the German Federal Republic, where the power to tax is fully in the hands of the federal level, disregarding the fact that the Basic Law includes most of the taxing powers under the concurrent FederalLänder competence catalogue. In terms of how the system functions, in fact, these powers fall within the sphere of influence (and decision) of the federal tier. Although every federal law on tax matters requires the Länder’ consent via the Bundesrat if the law stipulates regulations concerning taxes whose revenue accrue fully or partly to them.

20

Valdesalici

the term Finanzverfassung,37 coined by Austrian and German scholars and referring to those constitutional provisions that establish the principles and rules of the system of public finance, and having particular regard to the determination, distribution and use made of financial resources by the different levels of government.38 The concept, in fact, could be considered an evolution of the notion of fiscal constitution that first appeared in 1977 in The University of Chicago Law Review thanks to a contribution by Professor Kenneth W. Dam (law and economics),39 and later used by two famous American economics scholars—James Buchanan and Richard Wagner—to refer to those written or unwritten rules that guide fiscal decisions in the United States.40 As such, the expression is linked to those theories that rely on the nexus between fiscal federalism and the federal state; however, it takes credit for including not only rules “formally incorporated in some legally binding and explicitly constitutional document”, but also unwritten rules like “customary, traditional, and widely accepted precepts”.41 This approach entails three key consequences that are relevant for our purposes. First of all, it gives relevance to political facts and economic rules that have an impact on the interpretation and implementation of the rules, as well as on the way in which a system functions and evolves.42 Second, it justifies the inclusion of all those sources of law that deal with the subject matter but do not have formal constitutional status. In this case, substance prevails over form, as these “deserve to be thought of as quasi-constitutisonal” in nature. Third, it stresses the importance of “[contemplating] the Constitution as a 37

38

39 40 41 42

This is the terminology used, with little variation, by Italian and Spanish scholars as well. See, among others, M. Salerno, “Riflessioni sulla nuova costituzione finanziaria”, Federalismo Fiscale, 1 (2007) 119–138, at 123; M. Medina Guerrero, “Financiación autonómica y control de constitucionalidad (algunas reflexiones sobre la stc 13/2007)”, Revista d’Estudis Autonòmics i Federals, 6 (2008) 92–124, at 98. The latter, indeed, makes specific reference to the “territorial financial constitution”. With regard to the Austrian system, see P. Pernthaler, Österreichische Finanzverfassung: Theorie—Praxis—Reform (Wien: Wilhelm Braumüller, 1984), at 21 ff. In a similar way, with reference to the case of Germany, see, among many others, J. Hellermann, “Artikel 104a”, in H. von Mangoldt et al. (eds.), Kommentar zum Grundgesetz (6th edn., München: Franz Vahlen, 2010), at 1099–1186. Along the same line: BVerfGE 55, 274 (300)—Berufsausbildungsabgabe. Dam, “The American Fiscal Constitution”, supra, 271–320. J.M. Buchanan and R.E. Wagner, Democracy in Deficit: Political Legacy of Lord Keynes (New York: Academic Press, 1977). In this sense, ibid at 24. Carboni, “Fiscal Federalism and Comparative Law”, supra, at 3.

Defining Fiscal Federalism

21

whole, considering provisions not specifically directed [at] fiscal matters and taking into account the federal structure created by the Constitution”.43 This advocates the idea of taking into consideration the general principles and values that inspire the constitutional order as a whole, as well as those provisions that tangentially intersect with financial rules. The principles of the national tax system or economic and financial policies, together with national budgetary policies, could represent examples in this regard. Only this broader picture can reveal the “imposing edifice of powers and limitations” that constrains fiscal decisions.44 The only precondition is found in the need to test the impact of these rules on the financial system and its functioning.45 The latter assumption supports a dynamic understanding of fiscal federalism as inclusive of intergovernmental financial relations. As a matter of fact, only if rules and actors are observed in their interactions within the frame of reference can such an extended scenario be deemed to result in a wide-ranging understanding of the phenomenon. This is even more relevant when it comes to constitutional areas, such as financial constitutions, which show a wide gap between the law in books and the law in action. 43 44 45

In this sense, Dam, “The American Fiscal Constitution”, supra, at 272. Ibid. This broader interpretation is supported by, among others, Anderson, Fiscal Federalism, supra, at 2.

chapter 2

From a Formal to a Substantial Approach: Sources of Law and Fiscal Federalism Sara Parolari 1

Introduction

In all multilevel states,1 the arrangements regarding the raising, sharing and spending of resources are critically important, both politically and economically, for the functioning of the system as a whole. In addition, the way in which financial powers are assigned can significantly influence the allocation and exercise of legislative and administrative powers. Thus, guarantees of adequate financial resources and distribution mechanisms, fiscal and monetary powers as well as equalisation and solidarity funds are important for the functioning of each level of government, as they may create or prevent intergovernmental conflicts. More recently, the overall responsibility of the central state vis-à-vis subnational entities has been emphasised due to financial and fiscal conditionality, in particular for those states that are members of the Eurozone.

1 In this chapter, the term “multilevel state” is used in order to describe all the legal systems characterised by two or more levels of government (national and subnational), including both federal and regional states. With regard to the debate on the nature of federalism as a notion, including both federal and regional states, see, among others, R. Watts, Comparing Federal Systems (Montreal and Kingston: McGill–Queen’s University Press, 1999); A. Gamper, “A ‘Global Theory of Federalism’: The Nature and Challenges of a Federal State”, German Law Journal, 6 (2005) 1297–1318; P. Häberle, Föderalismus, Regionalismus, Kleinstaaten in Europa (Baden-Baden: Nomos, 1994); K. Wheare, Federal Government (Oxford: Oxford University Press, 1963). With regard to the notion of fiscal federalism and the theories behind it, see, among others, E. Ahmad and G. Brosio, “Introduction: Fiscal Federalism—A Review of Developments in the Literature and Policy”, in E. Ahmad and G. Brosio (eds), Handbook of Fiscal Federalism (Cheltenham: Edward Edgar, 2006), 3–29; W.E. Oates, Fiscal Federalism (New York: Harcourt Brace & Co., 1972), and W.E. Oates, “An Essay on Fiscal Federalism”, Journal of Economic Literature, 37 (1999) 1120–1149; A. Shah (ed), The Practice of Fiscal Federalism: Comparative Perspectives (Montreal and Kingston, McGill–Queen’s University Press, 2007); R. Broadway and A. Shah (eds), Fiscal Federalism: Principles and Practice of Multiorder Governance (Cambridge: Cambridge University Press, 2009).

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_004

From a Formal to a Substantial Approach

23

This chapter does not focus on the content of the various financial arrangements, which will be dealt with in the following chapters of this volume. Instead, attention will be given to an often-underestimated aspect: the sources of law regulating the financial settings in a selection of case studies and their possible effects, if any, on the design of the financial measures adopted within each legal system. The final aim is to draw a conclusion on how the “form” can influence the “substance” of a legal system. 2

“Financial Constitutions”: A Difficult Compromise between Legal Certainty and the Need for Flexibility

While constitutional and legislative guarantees are fundamental for legal certainty and economic management, there is also a need for flexible adaptation to changes in the economic and financial context. Since responsibilities related to both revenues and expenditures tend to change over time, no “financial constitution” can be expected to remain permanent.2 It is therefore challenging to find a compromise between legal certainty and flexibility that establishes mechanisms that are capable of adjusting, from time to time, the system of fiscal relations. The expression “financial constitution” comes from the German notion of the Finanzverfassung as an evolution of the concept of a “fiscal constitution” elaborated by James Buchanan and Richard Wagner, which specifically relates to the discipline of fiscal decisions in the us federal system.3 The concept refers both to the rules defining the financial relationships between the different levels of government and the rules on the allocation of the tax authority between national and the subnational governments.4 With limited exceptions, financial relations in multilevel states are first settled in constitutional documents. Nonetheless, there are broad discrepancies within this common pattern. 2 On this aspect, see R.L. Watts, “Processes for Adjusting Federal Financial Relations: Examples from Australia and Canada”, in P. Boothe (ed), Fiscal Relations in Federal Countries: Four Essays (Ottawa: Forum of Federations, 2003) 17–39, at 17. 3 J.M. Buchanan and R.E. Wagner, Democracy in Deficit: The Political Legacy of Lord Keynes (New York: Academic Press, 1977), 21–23. On the notion of “fiscal constitution”, see, also, K.W. Dam, “The American Fiscal Constitution”, The University of Chicago Law Review, 44 (1977) 271–320, at 272. 4 For this definition, see G. Bizioli, “Italy”, in G. Bizioli and C. Sacchetto (eds), Tax Aspects of Fiscal Federalism: A Comparative Analysis (Amsterdam: ibfd, 2011) 403–454, at 404n5.

24

Parolari

On the one hand, there are states that have very detailed and structured “financial constitutions”, that dedicate several articles to the topic that are collected in one or more specific chapters. Among these, Switzerland and Germany are typical examples, where the constitutional text, beyond defining basic principles, regulates in detail the functioning of the system, even specifying the revenue-raising powers of the different levels of government. The German Constitution (the Basic Law, Grundgesetz of 1949) devotes its Title x, from Article 104a to 115, to the discipline of intergovernmental financial relations: it lists the taxes belonging exclusively to the Bund and the Länder, mentions the taxes that are shared among the different tiers of government and, even more interesting, fixes the percentage of the tax share for each level.5 This constitutional scenario therefore distinguishes itself in terms of its clarity, and it underlines the importance of the topic addressed.6 Similarly, the Swiss Constitution of 1999 (as subsequently amended) deals rather comprehensively with various aspects of the financial regime (Articles 126–135, Title iii), providing the basic principles, the attribution of the financial competences to the cantons and the confederation, as well as the use of direct democracy on cantonal fiscal legislation.7 On the other hand, the “financial constitutions” of other states are classified as open since the constitutional discipline consists of few provisions, mentioning only the basic fundamental principles of financial relations (such as the principle of autonomy, the principle of solidarity, the duty to maintain a balanced budget). In such cases, the detailed regulation of the subject is left

5 See E. Bertolini, “I rapporti finanziari intergovernativi nell’evoluzione dell’ordinamento federale tedesco”, in G.F. Ferrari (ed), Federalismo, sistema fiscale, autonomie. Modelli giuridici comparati (Roma: Donzelli Editore, 2010) 77–105; V. Losco, Il federalismo fiscale in Germania: il disegno costituzionale e le applicazioni giurisprudenziali (Milano: Egea, 2005); H.P. Schneider, “Il federalismo fiscale in Germania”, in A. Depretis (ed), Federalismo fiscale “Learning by doing”: modelli comparati di raccolta e distribuzione del gettito tra centro e periferia (Padova: Cedam, 2010) 9–24. 6 See J. Englisch and H. Tappe, “The Federal Republic of Germany”, in G. Bizioli and C. Sacchetto (eds), Tax Aspects, 273–326, at 274. On Germany’s “financial constitution”, see, also, H.V. Mangoldt et al (eds), Kommentar zum Grundgesetz (München: Franz Vahlen, 2010); and M. Sachs (ed), Grundgesetz Kommentar (München: C.H. Beck, 2011). On constitutional reforms, see A.B. Gunlicks, “German Federalism and Recent Reform Efforts”, German Law Journal, 6 (2005) 1283–1296. 7 On Switzerland’s “financial constitution”, see D.P. Rentzsch, “The Swiss Confederation”, in Bizioli and Sacchetto, Tax Aspects, supra, 223–272.

From a Formal to a Substantial Approach

25

either to implementing legislation—consisting of ordinary or special national laws—or to autonomy statutes or to intergovernmental agreements. Among these countries, typical examples include Belgium, Italy and Spain. The Belgian Constitution—dating back to 1831 but has been subsequently amended several times—deals with the topic of financial relations in Title v, where three articles (175, 176 and 177) explicitly attribute the discipline of the system of financing for communities and regions to a special law adopted with a reinforced procedure.8 Even the Italian system belongs to this second group, as the main aspects of fiscal federalism are summarised in a few constitutional provisions (Articles 81, 116, 117 and 119 of the 1948 Constitution, as subsequently amended), leaving the detailed regulation to ordinary legislation. In particular, according to Article 119 of the Constitution, all territorial entities shall enjoy autonomy on both the revenue and the expenditure side, pursuing financial and political accountability before the electorate. Indeed, financial autonomy has to be balanced with solidarity, cohesion and coordination, as the Constitution provides for the coordination of the systems of both public finance and taxation and for an equalisation mechanism in order to ensure equality among all territories.9 Similarly, Spain’s Constitution of 1978 contains only a few articles (the most relevant of which are Articles 156, 157 and 158) that refer to fiscal federalism, which are very general provisions declaring that the regional financing system has to be based on a series of principles (such as autonomy, solidarity, sufficiency and coordination, as well as the duty to maintain a balanced budget).10 In particular, Article 156 declares the autonomous communities’ financial autonomy, Article 157 lists their revenue sources, and Article 158 states how the financing guarantees for the provision of minimum services, as well as the

8

9

10

On Belgian’s path to fiscal federalism, see, among the others, M. Verdonck and K. Deschouwer, “Patterns and Principles of Fiscal Federalism in Belgium”, Regional and Federal Studies, 13 (2003) 91–110. See A. Valdesalici, “Features and Trajectories of Fiscal Federalism in Italy”, in S.A. Lütgenau (ed), Fiscal Federalism and Fiscal Decentralization in Europe: Comparative Case Studies on Spain, Austria, the United Kingdom and Italy (Innsbruck: Studien Vorlag, 2014) 73–101, and Bizioli, “Italy”, supra. For an explanation of the implications of these principles, see A. Herrero and J.M. Tránchez, “El Desarollo y Evolución del Sistema de Financiación Autonómica”, Presupuesto y Gasto Público, 62 (2011) 33–65. See, also, J. Linares Martín de Rosales, “Comentarios a la Ley Orgánica de Financiación de las Comunidades Autónomas”, Hacienda Pública Española, 65 (1980) 119–192.

26

Parolari

inter-regional compensation fund, are determined.11 The details are instead provided by a special law adopted according to a reinforced procedure.12 In the middle, there are multilevel states whose constitutional provisions on fiscal federalism, on the one hand, do not simply lay down the basic general principles and, on the other hand, do not comprehensively regulate the topic, leaving the determination of many of the detailed arrangements to political processes, judicial interpretation, governmental practice and intergovernmental relations.13 The “financial constitution” of the United States, for example, is characterised by the absence of a comprehensive body of explicit foundational principles governing fiscal relations between the national and the subnational levels. Despite this, the us Constitution (written in 1787, ratified in 1788, and in force since 1789) includes a set of provisions on the power to tax, to spend, to print money and to borrow. In particular, it grants Congress the power to impose and collect taxes, duties, imposts and excises (Article I(8)(1)), while also identifying its explicit limitations. It does not mention the power of the states to levy taxes, even if the general rule, according to which the states retain all the powers not explicitly delegated to the federal government in the Constitution, also applies to fiscal powers. Moreover, this power is implicitly recognised by stating that no state shall, without the consent of Congress, levy imposts on imports or exports (Article I(10)(2)). With regard to spending power, the Constitution does not contain any specific discipline distinguished by the one of the normative powers at either the federal or the national level. Similarly, it does not contain any provision on intergovernmental transfers, leaving their design to federal laws with the consequent development of an uncoordinated variety of possibilities. Other than these explicit constitutional provisions, an impressive structure of rules has evolved, mainly from Supreme Court jurisprudence and from key framework legislation that has adapted the system to changing conditions. Likewise, the Canadian federation is based on a supreme written constitution (adopted in 1867 with some significant amendments in 1982), which refers 11 12

13

See J. Zornoza, “New Trends in Fiscal Decentralization: a Spanish View”, in Lütgenau, Fiscal Federalism, supra, 105–126, at 107. For the main characteristics of the Spanish financing system, see, among others, A.H. Alcalde, “The Spanish Way of Fiscal Federalism”, in Lütgenau, Fiscal Federalism, supra, 15–39. See R. Broadway and R.L. Watts, “Fiscal Federalism in Canada, the usa, and Germany”, iigr–Queen’s University Working Paper, 6 (2004) 1–24, at 3, http://www.queensu.ca/iigr/ working-papers/watts-collection (accessed 25 November 2016).

From a Formal to a Substantial Approach

27

to the distribution of expenditure responsibilities and revenue sources among the different levels of government. The Constitution deals with the allocation of taxation powers mainly in Sections 91 and 92, stating that the federal government is entitled to establish essentially all kinds of taxes, while the provinces have the power to impose direct taxation and licence fees, explicitly specifying that this is an exclusive power. It also provides, although not in details, the requirement of federal equalisation payments to ensure that provincial governments have sufficient revenues to provide reasonably comparable levels of public services at reasonably comparable levels of taxation (Section 36.2). However, in practice, the rules concerning this topic are provided in ordinary laws, jurisprudence, conventions and, above all, agreements signed between the federal and provincial governments. Such agreements delineate an architecture of financial relations that are rather different from those outlined by the constitutional text, in many cases justifying the dominion of the federation against the federal principle of parity between federation and provinces. Similarly, Australia dedicates some articles of its Constitution (Commonwealth of Australia Constitution Act 1900) to financial matters. In particular, it provides the federal level with the power to enact laws with respect to taxation (Sections 51(ii) and 90), to appropriate monies out of consolidated revenue and to grant money to the states (Section 96).14 However, the processes for adjusting federal–state financial relations have been developed beyond the constitutional provisions as a result of non-constitutional intergovernmental negotiation and agreements, with a stronger tendency—compared to Canada—to establish formal institutions.15 As an exception to this dual approach, the United Kingdom has no written constitution, and most of its fiscal regulations are laid down either in instruments of soft law (such as the Barnett Formula) or in statute law (for example, the Scotland Act 2012). Therefore, while some systems offer broad constitutional guarantees for financial relations, including taxation powers, other regimes only provide for general principles in their constitutions, leaving detailed regulation either to specific legislation or to other sources of law. The more detailed the “financial constitution” is, the more the autonomy of subnational entities is guaranteed, as changes in public finances require constitutional revision, which usually requires the involvement of a different level of government. On the contrary, an open model means fewer guarantees, but it is more flexible, as changes do not require constitutional amendments (with their special procedures 14 15

M. Stewart, “Australia”, in Bizioli and Sacchetto, Tax Aspects, supra, 137–186, at 139. See Watts, “Processes for Adjusting”, supra, 17–39.

28

Parolari

and thresholds and, in some cases, popular consultation). Indeed, flexibility is extremely important because financial relations are very dynamic: a typical feature is their unstable nature, as public finances are scarce, limited and extremely volatile, evolving constantly. Resources—in particular, tax revenues, public spending priorities, as well as surpluses or deficits—can change dramatically in short periods. The economic and financial crisis of recent years is a tangible example in this regard. Consequently, financial issues are often at the heart of the political debate, and intergovernmental relations in this field are constantly engaged in reform processes. In countries with detailed constitutions, this implies the need to revise constitutional documents on many occasions. In Germany, for instance, Title x of the Grundgesetz has been revised more or less twenty times since its entry into force in 1949 (the most recent reform being the Föderalismusreform ii in 2009). The same trend can be seen with regard to Switzerland, where there was even a case where one constitutional reform was adopted before the previous revision had time to enter into force. On the contrary, constitutional amendments have been rare in open models, with adaptation being achieved mainly through other evolutionary processes.­ In the case of Italy, for example, this occurs mainly through the interpretative role of the Constitutional Court, which happens to exercise this role almost as a substitute for the legislative power. Outside Europe, the us system has adapted to changing conditions mainly by way of amending the statutes (ordinary laws) that provide the rules on taxation and expenditure, as well as through Supreme Court jurisprudence.16 In Canada, the evolution of the system of financial relations in response to varying social and economic circumstances has been achieved mainly through intergovernmental negotiations and agreements, as well as by means of amendments to federal legislation and through the judiciary. 3

Implementing Legislation: The Need for Reinforced Procedures as a Guarantee for Matters of Constitutional Relevance

As mentioned in the previous paragraph, some multilevel states that have adopted an open model of a “financial constitution”17 tend to determine their financial framework in a normative act that is, in most cases, vested with constitutional or quasi-constitutional binding force. 16 Dam, The American Fiscal Constitution, supra, at 273. 17 This can be affirmed especially with regard to European multilevel states.

From a Formal to a Substantial Approach

29

This is the case in the Spanish system, where the subject matter is thoroughly regulated in the Ley Orgánica de Financiación de las Comunidades Autónomas­(Organic Act on the Financing of the Autonomous Communities (lofca)), which is a source of law that has superior force in relation to an ordinary law passed by the Parliament (but inferior to the Constitution). In fact, the approval, amendment or repeal of this kind of law requires an overall majority of the members of Congress in a final vote on the bill as a whole (Article 81(2) of the Constitution). Organic laws are normative acts that typically relate to the implementation of fundamental rights and public liberties, those approving statutes of autonomy and the general electoral system and other laws provided for in the Constitution, such as the law for the financing of the autonomous communities (Article 81(1) of the Constitution). In this regard, the lofca, originally approved by Organic Law 8/80 and subsequently amended several times, regulates the relationships between the central state and the autonomous communities of the general regime, leaving aside the relationship with the autonomous communities having a special foral regime.18 Reforms of the lofca are de facto elaborated by a powerful intergovernmental body composed of ministers of the state and of the autonomous communities, the Consejo de Política Fiscal y Financiera (Council for Fiscal and Financial Policy (cpff)). It is important to remember that the autonomy statutes (together with Constitution and lofca) are sources of law that deal with the topic of financial relations in Spain. Like the lofca, they are approved by organic laws, but there is no clear relationship between the two in the cases of potential conflicts in the regulation of the financial structure of the autonomous communities.19 Even if there are no unique interpretative criteria, the Spanish Constitutional Court has stated in many of its judgments that the autonomy statutes have to be interpreted in conformity with the lofca, and not the other way round, so that the latter has to be considered a fundamental parameter of constitutionality.20 Nevertheless, it is worth mentioning that, in the past, the autonomy statutes had a strong political influence on the lofca: it often happened, in fact, that the latter was modified in order to transpose some of the contents of

18 19 20

See F.A. García Prats, “Spain”, in Bizioli and Sacchetto, Tax Aspects, supra, 361–402, at 365. Ibid. at 381. F. Pérez Royo, “Cuestiones sobre el poder tributario de las ccaa: la relación entre las disposiciones sobre financiación en los Estatutos y la Ley Orgánica del artículo 157.3”, in A.M. Pita Grandal (ed), La financiación autonómica en los Estatutos de Autonomía (Madrid: Marcial Pons, 2008), 231–238.

30

Parolari

the statutes when a reform had been introduced (this happened, for example, in the case of the recent amendments to the Catalan Autonomy Statute).21 The same applies in the case of Belgium, where more details on the financing system for the communities and regions are provided in a special law (loi spéciale relative au financement des Communautés et des Régions) that is passed by a majority of the votes cast in each linguistic group in each house of parliament on condition that a majority of the members of each group is present and provided that the total number of votes in favour that are cast in the two linguistic groups is equal to at least two-thirds of the total number of votes cast (Article 175 and Article 4(3) of the Constitution).22 A different case is that of Austria, where the “financial constitution” consists of one single article (Article 10(1)(4) of the 1920 Constitution) that has to be integrated by the constitutional finance law (the Finanz-Verfassungsgesetz (fvg)). This is a separate constitutional act that lays down the fundamental rules governing the powers of the Bund, Länder and municipalities, as well as the financial relations between them.23 Within this legal framework, the design of specific rules is left to the fiscal equalisation law (Finanzausgleichsgesetz (fag)), which is a federal law with a limited period of application that results from a negotiation process among the federal government, state governors and municipal councils.24 In this sense, financial relations in Austria are protected against amendments by the central level not through a constitutional provision but through the fact that the process of approval and amendment are inclusive and participatory. 21

22

23

24

On the relationship between the lofca and the autonomy statutes, see C. García Novoa, “El sistema de financiación en la reforma del Estatuto Catalán, ¿es compatible con el orden constitucional?” rcdp—Revista catalana de dret públic, 32 (2006) 75–118. See, also, the jurisprudence of the Tribunal Constitucional and, in particular, sentences 31/2010 and 247/2007. The first of these laws was approved in 1989, and it was subsequently amended in 1993 and 2001. See L. Primicerio, “Le relazioni finanziarie intergovernative in Belgio, tra logica della uniformità e valorizzazione delle differenze”, in V. Atripaldi and R. Bifulco (eds), Federalismi fiscali e Costituzioni (Torino: Giappichelli, 2001) 109–135. See, also, L. Gallez and B. Gors, “Aspetti recenti del federalismo regionale e del decentramento territoriale in Belgio”, Le Regioni, 6 (2005) 1055–1082, at 1057. In this sense, see G. Kofler, “Austria”, in Bizioli and Sacchetto, Tax Aspects, supra, 327–340. See, also, F. Palermo, “La Costituzione finanziaria austriaca. Tare genetiche di un sistema in evoluzione”, in V. Atripaldi and R. Bifulco (eds), Federalismi fiscali e Costituzioni, supra, 75–108. In this regard, see H. Pitlik, “Fiscal Federalism, Austrian Style: Fear of Competition”, in Lütgenau, Fiscal Federalism, supra, 41–59.

From a Formal to a Substantial Approach

31

A quite peculiar case is once again the British legal system: when it opened up to statute law in this field, it did so by using ordinary legislation (an act of the Westminster Parliament) adopted according to a special procedure. For instance, the new rules on the transfer to Scotland of the power to set income tax rates in Scotland are foreseen by the Scotland Act 2012, which is a normative act adopted according to the procedure provided by the so-called Sewel Convention. This means that when the uk Parliament legislates on a matter that is normally dealt with by the Scottish Parliament (such as fiscal matters), Westminster may legislate only if the Scottish Parliament has given its consent. It is interesting to note that the implementing phase in the United Kingdom is mainly left to constitutional conventions, which are stricto sensu non-legal rules.25 In fact, notwithstanding the criticism of distribution criteria, the Barnett Formula is still in force.26 This means that the majority of expenditures afforded by the devolved administrations (leaving aside local authority on finance) are derived from a block grant provided from Westminster and calculated under a formula based on a conventional rule, which may therefore be changed if the uk Government decides to renegotiate it. This means that financial arrangements in the uk are mainly non-statutory and are instead the result of negotiations through political channels. Since the formula is not based on legislation, but rather on its acceptance by the devolved administrations, there is no recourse to the judicial process in the event of dissatisfaction. However, it has been remarked that the devolution referenda held in various areas of the United Kingdom also provided legitimacy to the funding arrangements (and indirectly to the Barnett Formula), fostering a better safeguard than any uk statute could have offered.27 As an exception to the tendency towards reinforced sources of law, the new version of Article 119 of the Italian Constitution, which vests more financial autonomy in territorial entities, was implemented, on the one hand, through an ordinary normative act and, on the other hand, without any formal 25

26

27

On constitutional conventions as sources of law, see P. Leyland, The Constitution of the United Kingdom: A Contextual Analysis (Oxford and Portland: Hart Publishing, 2012), 32–36. The expression comes from the then-deputy finance minster, now Lord, Barnett, who invented it. See A. Trench, “The uk’s Devolution Finance Debates”, in Lütgenau, Fiscal Federalism, supra, 321–339. See S. Eden, “United Kingdom”, in Bizioli and Sacchetto, Tax Aspects, supra, 541–580, at 551. On the Barnett Formula, see, also, P. Leyland, “The Scottish Referendum, the Funding of Territorial Governance in the United Kingdom and the Legislative Role of the Westminster Parliament”, Le Istituzioni del Federalismo, 4 (2014) 857–884, at 862ff.

32

Parolari

participation­of territorial entities. Framework Law 42/2009, which provides for a new order of financial relations among all tiers of government,28 beyond not having constitutional binding force, is nothing other than a delegation from the Parliament to the executive of the power to adopt a number of bylaws (nine enactment decrees in two years) in order to provide for the effective functioning of the new financial regime.29 It should also be noted that the nine enactment decrees so far adopted are not self-executing:30 either they need further integration by means of executive rules or they postpone the determination of crucial aspects of the discipline. This is quite unusual, as, in most countries, financial regulations, even if the implementing rules do not have constitutional status, rely at least on a parliamentary law and not on a governmental act.31 In Italy, by contrast, even if financial relations have a constitutional nature, the Parliament has delegated

28

29 30

31

With regard to the content of this law, see: Bizioli, “Italy”, supra. For an in-depth analysis of Law 42/2009, see, also, E. Jorio et al, Il federalismo fiscale. Commento articolo per articolo alla legge 5 maggio 2009 n. 42, (Santarcangelo di Romagna: Maggioli, 2009); and A. Ferrara and G.M. Salerno, Il federalismo fiscale. Commento alla legge n. 42 del 2009 (Napoli: Jovene, 2010). On the consequences of the choice of this kind of law, see F. Scuto, “The Italian Parliament Paves the Way to ‘Fiscal Federalism’”, Perspectives on Federalism, 2 (2010) 67–88. The first step was taken with the adoption of the Decree on Public Property Federalism, which provides for the assignment of state properties to regional and local authorities (D.lgs. 85/2010). The same year, two other enactment measures were adopted: the first on “the Capital City of Rome” (D.lgs. 156/2010), which was followed by a decree aimed at transferring the related functions to the capital city (D.lgs. 61/2012, then modified and integrated through D.lgs. 51/2013); the second on “Standard Costs and Needs of Provinces and Municipalities”, which should have allowed the definition of the new financing criteria at the basis of equalisation (D.lgs. 216/2010). Then, in a fourth decree, a compromise was reached on “municipal federalism”, which concerns revenues for financing the municipal level (D.lgs. 23/2011). A fifth act governs the “regional financing system” and establishes the premises for defining the “standard costs in the health sector” (D.lgs. 68/2011). Another decree on “Infrastructural Equalisation and Special State Grants” was then adopted on 31 May 2011 (D.lgs. 88/2011). In June 2011, a decree that stipulates the “harmonisation of the budget models” among the different entities (D.lgs. 23/2011) was endorsed; while, in September 2011, the last decree (D.lgs. 149/2011) on “Awards and Sanctions” was finally adopted with the aim of fostering accountability and democratic oversight. However, the implementation process is far from being concluded. See F. Palermo and A. Valdesalici, “Italy: Autonomism, Decentralization, Federalism or What Else?” in J. Lluch (ed), Constitutionalism and the Politics of Accommodation in Multinational Democracies (London: Palgrave Macmillan, 2014) 180–199, at 187.

From a Formal to a Substantial Approach

33

its specific regulation to the government.32 This means that such questions will be decided by the political majority, without the need to reach the sort of broad consensus that would be expected for a matter of constitutional relevance. In this case, both the Parliament and the territorial entities have been granted only a marginal role. Indeed, the approval of executive decrees has to follow a complex and derogating process that aims to guarantee a minimum involvement of both territorial and parliamentary forces. The legislator (Article 2(3) of Law 42/2009) stipulates a prior agreement with the territorial entities, which are represented through a system of conferences and ad hoc bodies established by Law 42/2009 itself. The fact that there are no consequences for failed negotiations, apart from the need to motivate the dissent, hinders effective participation in the decision-making process. In addition, a parliamentary commission has to give a positive opinion about the governmental decree, but this, once again, is a mere consultative role. In addition, since the Italian Parliament is composed of two equal chambers, no territorial representation is guaranteed at this level either.33 The only exception within this framework are the financial regulations of the so-called special regions, which are based on sources of law that have a particular rank due to the procedural guarantees for their adoption. As a matter of fact, both the revision of the autonomy statutes, with particular regard to the part where regional finance is framed, and the related enactment measures need the consensus of all entities involved out of respect for the principle of parity. A particular procedure therefore has to be followed, consisting in the adoption of a law by the Parliament that reproduces a prior agreement between the executives at the national and territorial levels. As a consequence, it is considered an entrenched law that has a superior binding force with respect to ordinary laws. The same goes for the related enactment measures, which have to be adopted by the executive in the form of a decree, but all parties have to agree upon their content, which is de facto defined in joint commissions made up of representatives of both the state and the entity in question, even if such commissions are not elected bodies. Being entrenched sources of law in between ordinary and constitutional 32

33

On this aspect, see D. Cabras, “Il processo di attuazione della legge delega in materia di federalismo fiscale: il ruolo del Parlamento”, federalismi.it, 12 (2009), www.federalismi.it (accessed 25 November 2015); and R. Bifulco, “Il Parlamento nella tenaglia degli Esecutivi: il federalismo fiscale e la riforma del Senato”, nelMerito (2009), www.nelmerito.com (accessed 25 November 2015). See Valdesalici, “Features and Trajectories”, supra, 73–101. In this sense, the constitutional reform under discussion, if it finally enters into force, will somehow introduce a form of territorial representation, as the new Senate will be composed of members of the regional assemblies and mayors of municipalities in each region.

34

Parolari

acts, these governmental decrees are considered sources of law with a quasiconstitutional­ status.34 Therefore, it follows that (even if the participation of territorial entities is guaranteed more than in the case of ordinary regions) the phenomenon of the marginalisation of elected assemblies is also common to special regions. It is up to the executive branch to define the details of the financial relations system, which are then—only formally—adopted by the relevant legislatures. Indeed, the increasing role of the executives in reform processes in general and with regard to financial issues in particular is a phenomenon common to many legal systems, as is confirmed by the examples provided in the following section. 4

From Political Negotiation to Jurisprudence: The Contribution of Non-Legal Sources to the Design of Fiscal Federalism

Financial arrangements, as provided either at the constitutional level or in implementing legislation, typically leave some room for negotiation and agreement at both the multilateral (in the case of federal systems) and bilateral levels (in the case of single territories with differentiated status) in order to adapt to the continuous changing of social and economic circumstances. The recourse to political negotiation is more frequent when it is difficult, for various reasons, to implement formal constitutional amendments or when it is necessary to manage the asymmetrical framework. In Canada, where the financial system reflects the inherent nature of the multiethnic and multinational federal state,35 when the legal framework is highly asymmetrical, it becomes crucial to abandon uniform solutions in order to find flexible answers to accommodate particular needs. For this reason, negotiations between the executives at all levels of government are frequently used with the aim of producing adjustments to the arrangements for financial transfers from the federal government to the provinces. As both the federal and the provincial governments have taxation powers with regard to personal and corporate income taxes, as well as sales taxes, there is a risk of overlapping

34

35

On the role and binding force of the enactment decrees of the autonomy statutes see, among others, the following judgments of the Constitutional Court: No 51/2006, No 341/2001, No 213/1998, No 137/1998. On this aspect, see I. McLean, Fiscal Federalism in Canada (Oxford: Oxford University Press, 2003).

From a Formal to a Substantial Approach

35

tax jurisdictions.36 In order to reduce the risk of conflict, agreements between governments have been reached with the purpose of giving the federal level the possibility to pursue general objectives, while the provinces maintain the power to design the main traits of their tax policy.37 Unlike other countries, Canada has neither specific legislation nor a constitutional provision governing these agreements, which can therefore be formal or informal without having­a predefined format.38 At the same time, it is clear that these intergovernmental agreements are not intrinsically legal matters,39 but rather political ones, with their observance relying on the political leverage of the participants and upon reaching some sort of consensus.40 Similarly to Canada, many of the processes for adjusting financial relations in Australia are developed through political negotiation and bargaining even if, in this case, more formal institutions have being established (such as the Loan Council) in order to facilitate these processes.41 What is not legislated is left to soft law, informal meetings, practices and forums for negotiation. In both cases, it is clear that financial negotiations have taken on a form of “executive federalism” that is managed by the executives of each level of government within the federation (ministers, but also civil servants), but not involving the elected assemblies. Therefore, parliaments at both the national and the subnational level play a marginal role, leaving all powers in the hands of the executive branch, which has, in practice, come to dominate this relationship, although it is formally accountable to the legislature. In both cases, the “financial constitution” has thus been reinterpreted and modified without a formal reform process, sometimes up to the point of a complete loss of content of the normative provisions officially in force. Intergovernmental agreements have therefore been the primary arena for negotiating the evolution of financial systems, somehow adjusting legal provisions to reality. In addition, courts have a crucial role in defining the competences of the different levels of government, since it is natural that conflicts 36 37 38

39 40

41

Watts, “Processes for Adjusting”, supra, at 6. Ibid. at 7. See, also, Broadway and Watts, Fiscal Federalism, supra, at 13. On this aspect, see J. Parker, Comparative Federalism and Intergovernmental Agreements: Analyzing Australia, Canada, Germany, South Africa, Switzerland and the United States (New York: Routledge, 2015) 65–88, at 69. This was confirmed by the Supreme Court in the case Reference Re Canada Assistance Plan (b.c.), [1991] 2 s.c.r., 525. J. Poirier, “Intergovernmental Agreements in Canada: At the Crossroads Between Law and Politics”, in J.P. Meekison et al (eds), Reconsidering the Institutions of Canadian Federalism (Montreal and Kingston: McGill–Queen’s University Press, 2002) 425–462. R.L. Watts, “Processes for Adjusting”, supra, at 12.

36

Parolari

may arise regarding fiscal issues. Instead of frequently modifying constitutional documents, the evolutionary interpretation of the “financial constitution” and its implementing legislation has been attributed—together with negotiations between the executives from each order of government—to jurisprudence.­ Sometimes, this has even led to a distortion of its functions where judicial power has become a sort of substitute for the legislature and has therefore been called to play a different role than the one assigned to it by the Constitution itself. This has happened in Canada, among other countries, where one of the roles of the Supreme Court is to resolve conflicts involving the competences belonging to the provinces and the federation, with such disputes accounting for the majority of its judgments. The consequence is that the Court has often rewritten the constitutional text, a fact that has overwhelmingly contributed to the determination of taxation powers (direct or indirect) being assigned respectively to the federation and the provinces.42 This is also true with regard to spending powers: as an example, it is jurisprudence that has affirmed, over time, the existence of a general spending power of the federation that has allowed the federal government to become involved in areas of provincial jurisdiction, such as health care.43 Similarly, lacunae in the us financial constitution have been filled in by the interpretation of the Supreme Court. This offers an answer to the need to adjust a document drafted in a very different era to changed financial conditions. The Court’s interpretation of the Commerce Clause in the Constitution is somehow indicative of the substitute role of jurisprudence: instead of intervening to legislate on the subject, Congress has frequently adopted selfrestraint in waiting for a judgment of the Supreme Court, which sets the limits of state legislation on the matter. All in all, the Supreme Court has a crucial role in defining the extension of federal powers, giving content to general theoretical principles such as the one of uniformity.44 Italy represents another example where a constitutional court plays a strong interpretative role even to the extent that it substitutes for the legislature in setting the cornerstones of financial relations. An example can easily be found 42

43 44

In this regard, see T.E. Frosini and P.L. Petrillo, “Il federalismo fiscale ‘decostituzionalizzato’ canadese e la negoziazione tributaria tra livello federale e provinciale”, in G.F. Ferrari (ed), Federalismo, sistema fiscale, supra. See, also, B. Alaire and R.M. Bird, “Canada”, in Bizioli and Sacchetto (eds), Tax Aspects, supra, 79–136, at 91ff. In this sense, see P. Boothe, “Taxing, Spending and Sharing in Federations” in Boothe, Fiscal Relations, supra, 5–16, at 8. G.G. Carboni, Federalismo fiscale comparato (Napoli: Jovene, 2013).

From a Formal to a Substantial Approach

37

if we take into consideration regions’ financial autonomy. According to the Italian Constitution (Article 119), regions are vested with financial autonomy in terms of both revenues and expenditures. As to spending, regions rely, in theory, on fully fledged autonomy that results from the combined scope of legislative and administrative competences. De facto, the state has adopted a set of rules (e.g. Law Decrees 78/2010, 98/2011, 138/2011, 201/2011, 1/2012, 95/2012; Law 147/2013) that call for a review of decentralised spending competences. As such, they often interfere with regional autonomy, but the Constitutional Court has adjudicated them as a legitimate attempt to cope with the demanding European obligations under the aegis of the central state competence of the “fundamental principle of financial coordination” (among others, Constitutional Court judgments 430/2007, 341/2009, 11/2010, 82/2015).45 Generally speaking, the reasoning of the Constitutional Court relies on the following rationale: if the state sets overall rules, it considers them to be the principles of financial coordination, which is currently a concurrent (state-region) competence. This vests in the state the authority to determine the fundamental principles of the subject matter, while the regions regulate the details. Indeed, the Court has given a rather generous and extensive interpretation of what a fundamental principle in this field is, broadening the scope of state legislation and de facto nullifying the regional role (Constitutional Court judgments 68/2011, 108/2011, 182/2011, 77/2013, 89/2014). One could conclude that the more a “financial constitution” is open, the more room jurisprudence has to manoeuvre. This does not mean that, in countries with detailed financial constitutions, the jurisprudence of the constitutional courts has not become a crucial means of assessing and resolving the problems that frequently result from fiscal relations in multilevel states.46 Even in constitutional systems based on very detailed constitutional documents and a clear separation of powers, the contribution of the judiciary in defining the functioning and the development of financial intergovernmental relations has been quite relevant in recent years. In Germany, for example, the Constitutional Court (Bundesverfassungsgericht) has exercised in a substantial way the power to review the constitutionality of legislation even regarding the financial issues. In doing this, it has sometimes been criticised for “stretching” its judicial review power in a quasi-legislative or political way, notwithstanding­ 45 46

All judgments of the Italian Constitutional Court are available at http://www.giurcost .org/decisioni/index.html. With regard to Germany, the Constitutional Court (Bundesverfassungsgericht) has been fundamental in defining the concept of Bundestreue, being an unwritten constitutional principle. On the latter, see H. Bauer, Die Bundestreue (Tübingen: Mohr, 1992).

38

Parolari

the constitutional principle according to which it must exercise this power without entering the field of political activities and the fact that it has imposed on itself a principle of judicial self-restraint.47 5

Conclusion

Despite the attempt to classify the various sources of law regulating financial arrangements suggested in this chapter, one can conclude that it is hardly possible to determine general rules on the topic. This is due, among other factors, to the intrinsic nature of financial relations, which are extremely dynamic. Consequently, the way they are regulated may significantly differ from one state to another, and they may be subject to incessant variations. This is particularly true in the current economic crisis, which requires substantial modifications of legal systems in order to face the new challenges. Even the traditional distinction between civil law and common law is not helpful in this regard, as there are no common solutions that would suit all of the relevant legal systems. Therefore, the sources of law regulating financial arrangements vary very much even when considering case studies from the same legal family. Within this pattern, however, certain broad tendencies can be observed: on the one hand, there is an increasing role for executives in reform processes due to the need to introduce emergency measures as fast as possible; on the other hand, the judiciary has a crucial role in giving content to theoretical principles, sometimes to the extent of becoming a substitute for the legislature. Furthermore, new sources of law, even if not strictly legal, are increasing in number. This is a phenomenon common especially in asymmetrical contexts, where negotiation and political agreements have emerged in recent years as atypical sources of law that are essential for regulating the relationship between the centre and the peripheries of the state, as well as for addressing claims for secession. Indeed, in many cases (such as special regions in Italy, autonomous communities of foral regime in Spain, or Scotland in the uk), the final effect of such political negotiations has been the achievement of better financial conditions for the territorial entities involved. Since it is extremely difficult to single out common trends in the sources of law regulating financial matters, it is likewise hard to evaluate whether the chosen legal instruments have had an influence on the design of the financial measures adopted within each legal system. Taking the constitutional level as 47

See D. Brand, Financial Constitutional Law: A Comparison between Germany and South Africa (Johannesburg: Konrad-Adenauer Stiftung, 2006), at 223ff.

From a Formal to a Substantial Approach

39

an example, one may wonder whether there is a relationship between the degree of detail of the constitutional document and the degree of decentralisation of the financial system. If considering countries from both civil-law and the common-law traditions, it could be argued that the more open the model is, the less devolved the financial system is. However, it would be incorrect to conclude that this is a general rule, a fortiori, when examining the widespread trend in most multilevel states pointing to an increase in the number of claims for subnational financial autonomy. In conclusion, the question of whether and how the “form” has an influence on the “substance” of financial arrangements has no unequivocal, final answer due to the complexity of reality reflected in various legal systems.

chapter 3

Financial Autonomy vs. Solidarity: A Dialogue between Two Complementary Opposites Cheryl Saunders 1

Introduction

All federations or deeply devolved governmental systems necessarily make provision of some kind for the organisation and allocation of authority in relation to financial matters. Given the importance of what may be described as fiscal federalism some of these arrangements are likely to be protected in a Constitution. Fiscal arrangements include at least the allocation of competence in relation to taxation and spending and the principles that govern any redistribution of public funds between the orders of government. They potentially extend also to authority to borrow or to raise public funds by other means, including arrangements for the receipt and distribution of foreign aid. The distribution of competence in relation to taxation, spending and other financial aspects of government almost invariably mirrors the federal distribution of competence more generally. Any judicial doctrines developed in relation to the general allocation of competence thus are likely to apply in the fiscal context as well. In some respects, nevertheless, both the principles that govern the allocation of fiscal competence and the judicial doctrines that apply to it are likely to be distinctive. The competence to raise or receive public funds, together with the concomitant arrangements for revenue redistribution, is central to the capacity of the orders of government in a federation to perform their functions in the public interest. The manner in which governments raise, spend, and receive funds also is structured by the requirements of representative democracy for the accountability of elected officers to the public. Individually or collectively, these features of fiscal federalism offer a catalyst for its treatment as a special case. Just as there are different approaches to the allocation of competences amongst the federal systems of the world, so there are different approaches to federal fiscal design. The former may be conceived as ranged along a spectrum from dualism to integration, with a multitude of variations in between. Dualism for this purpose refers to federations in which each order of government makes and administers its own legislation in the areas of competence allocated

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_005

Financial Autonomy vs. Solidarity

41

to it. Under an integrated approach to federal design, on the other hand, the sub-state orders implement much central legislation as of right and may, correspondingly, have limited legislative competence of their own. By definition, orders of government in a dualist system have a high level of autonomy, which flows from the logic of the federal design. By contrast, governments in integrated federations necessarily are interdependent to a greater degree again, as a matter of design. To the extent that fiscal federalism is a sub-set of the approach to the division of competence, this same distinction between dualism and integration can be expected to apply in the fiscal context. Arrangements for fiscal federalism, similarly, can be expected to be ranged along a spectrum. On the other hand, the link between revenue raising and revenue redistribution suggests other considerations that need to be factored into the analysis as well. As a generalisation, as a matter of principle, the allocation of tax competence in conditions of dualism implies that each order of government is responsible for raising the revenues needed for its own purposes and obviates the need for revenue redistribution (whatever the situation in practice). The association with autonomy is reinforced in these circumstances. On the other hand, the interdependence created by integration necessitates a sharing of revenues, giving rise to a principle more accurately described in terms of solidarity. This chapter compares approaches to the design and operation of fiscal arrangements in federal-type systems by reference to the extent to which they reflect the principles of autonomy and solidarity. For this purpose, autonomy refers to those elements of the system that expect and protect independent action by each order of government and encourage self-reliance. Solidarity, on the other hand, refers to features that involve interdependence and anticipate the collective interest in effective government in a federated state. The chapter argues that, even though one or other of the principles of autonomy and solidarity may be more pronounced in some federations than in others, all federal-type systems necessarily have elements of both. A degree of autonomy in relation to taxing and spending is an inevitable concomitant of the federal requirement of self-rule. Equally, however, there are obvious pressures for solidarity between orders of government in a single polity that is shared. The chapter argues further that, whatever the initial mix of autonomy and solidarity, change is likely over time, in the functioning of the federation, if not in its design. Catalysts for change include the economic context in which the federation operates, internally and externally and political power-struggles between the orders of government over these important tools for policy implementation. Change may involve alteration in the federal constitutional or legal

42

Saunders

framework. Even where the framework remains static, however, functional change may occur through political practice reinforced, on critical points, by judicial interpretation. To make such a comparative study practicable, the chapter focusses on two case studies: the United States and Germany. The selection has the advantage of taking examples from different geographical regions, different systems of law and different constitutional traditions. Most importantly for present purposes, however, each of these federations exemplifies a significantly different approach to the design and operation of fiscal federalism from the perspective of the mix of autonomy and solidarity. The United States and Germany can be seen as paradigm cases for present purposes, in the sense that the former provides for a high degree of autonomy and the latter reflects considerable solidarity. To this extent they occupy opposite ends of the spectrum, subject to the caveat that each also incorporates at least some elements of both principles and that in each some movement along the spectrum is evident over time. These two cases thus are effective to illustrate what the principles of autonomy and solidarity may involve in functioning federations. The case studies are analysed separately below, in Parts 2 and 3 of this chapter. Each part begins by exploring the admixture of principles in the design and operation of the fiscal constitution, focussing on taxation and expenditure. Each also deals separately with the effects of the interpretation and application of the fiscal constitution by the courts, through the processes of judicial review, by the Supreme Court of the United States and the Federal Constitutional Court of Germany respectively. As will be seen, judicial review of the respective constitutions is significant not only in enforcing the boundaries of fiscal federalism but also because it yields further insight into the principles on which fiscal arrangements are conceived to rest. It should be acknowledged at the outset that these two cases, alone, are potentially misleading for comparative purposes. Most federal-type systems in the early 21st century involve a mixture of the principles of autonomy and solidarity in more equal proportions. Most therefore can be found somewhere along the spectrum of approaches to fiscal federalism between these two paradigm cases. A final concluding section of the chapter draws attention to the principal ways in which autonomy and solidarity may be combined elsewhere. As that section also suggests, however, the fiscal constitutions of most federations can be understood as variations on one or both of the two paradigm cases, which are useful for this reason as well. Paradoxically, nevertheless, the fiscal constitution of each federation is unique, at least in some degree. In some instances this is the result of the particular combination of features in the design of the fiscal constitution from the outset. In any event, it is likely to

Financial Autonomy vs. Solidarity

43

be reinforced by the interplay of autonomy and solidarity in the evolution of the federation over time. 2

United States

2.1 Autonomy and Solidarity The United States is included in this chapter as a federation with fiscal arrangements in which the autonomy of the several orders of government is particularly pronounced. This characteristic is the unsurprising consequence of the manner in which the federation was formed, by drawing together the original 13 states with developed systems of government of their own. It remains a significant feature of the United States federation, despite the dramatic stages that have occurred over time through its operation in practice, described by scholars as involving transition from co-ordinate federalism through cooperation­to coercion.1 In addition, however, as this part seeks to show, fiscal federalism in the United States as it now operates in practice also exhibits marked elements of solidarity, albeit of a kind that reflects the dualism of the design from which autonomy derives. In its current form and operation, fiscal federalism in the United States results from a combination of formal constitutional change, judicial interpretation and political practice. As with any federation, the features of the constitutional system that make it federal in form cannot adequately be understood without considering the institutional, legal and cultural context in which federalism is embedded. In the United States, these include distinctive arrangements for the separation of powers, a common law approach to constitutional adjudication and a political culture that tends to favour self-reliance. The scheme for the federal allocation of competence is to enumerate the legislative powers of the federal order of government, leaving the rest to the States, as confirmed by the Tenth Amendment.2 The system is dualist, in the sense that it is assumed that each order of government executes its own laws and provides for adjudication of disputes arising under them, subject to the requirements of the federal Constitution.3 The legislative powers of Congress are not numerous but they are potentially broad, even on their face, and 1 T.E. Smith, “Intergovernmental Relations in the United States in an Age of Partisanship and Executive Assertiveness, in J. Poirier et al (eds), Intergovernmental Relations in Federal Systems (Oxford: Oxford University Press, 2015) 411–439, at 505. 2 u.s. Const. Art. I(8). 3 u.s. Const. Art. II(4); Art. III(2).

44

Saunders

include incidental authority to make laws “which shall be necessary and proper for carrying into execution” powers vested in federal institutions. Although they are not specifically identified as such, the legislative powers of Congress are variously either exclusive or concurrent.4 Concurrency is understood to indicate competences also exercisable by the States, subject to any constitutional provision to the contrary and to the supremacy of federal law.5 The allocation of competence in relation to taxation accords with this overall scheme. What on the face of it is a general power to “lay and collect taxes, duties, imposts and excises” is conferred on Congress by Article 1(8), subject to a requirement for the uniformity of all “duties, imposts and excises”.6 Article 1(9) prohibits taxation of exports, however. With Section 2, it once also required the apportionment of direct taxation between the States in proportion to population, effectively precluding a federal income tax, some forms of which were held by the Supreme Court to comprise direct taxation.7 In its application to income tax, this constraint was removed by the Sixteenth Amendment in 1913, paving the way for federal income taxation “from whatever source derived” and providing an income stream to fund the expansion of federal activity into new fields.8 A subsequent decision of the Supreme Court held that much income taxation was indirect after all and so subject to the uniformity requirements of Article 1(8),9 as interpreted by the Supreme Court.10 The States also have a taxation power that is broadly general in terms of types of taxation, with the important exception of State duties on imports or exports without the consent of Congress.11 State taxation competence also is subject to other constraints, however, derived explicitly or implicitly, directly or indirectly from the federal Constitution, as well as from the Constitutions of 4

5 6

7

8 9 10 11

G.A. Tarr, “United States of America” in J. Kincaid and G.A. Tarr (eds), Constitutional Origins, Structure, and Change in Federal Countries (Montreal: McGill–Queen’s University Press, 2005) 381–408, at 382; 388. u.s. Const, Art. vi. The addition of the words “pay the debts and provide for the common defence and general welfare of the United States” provides a link to what now is the spending power and might have made the power purposive but has not done so in practice. This issue is taken up below. Pollock v. Farmers’ Loan and Trust Co, 157 u.s. 429 and 158 u.s. 601 (1895). For the link between the apportionment provisions and the original constitutional compromises over slavery see B. Ackerman, “Taxation and the Constitution”, Columbia Law Review, 99 (1999) 1–58. A.A. Ekirch, Jr., “The Sixteenth Amendment: The Historical Background” Cato Journal, 1 (1981) 161–182, at 161. Brushaber v. Union Pacific Railroad Co, 240 u.s. 1 (1916). Head Money Cases, 112 u.s. 580 (1884); United States v. Ptasynski, 462 u.s. 74 (1983). u.s. Const. Art. I(10).

Financial Autonomy vs. Solidarity

45

the several States themselves. The most significant of the federal constitutional constraints include the implications drawn from the dormant commerce clause where State taxation is deemed to distort interstate or foreign trade and federal laws that pre-empt aspects of State taxation.12 The impact of both on State tax systems has expanded over time, initially with the growth in the scope of federal competence and more recently in consequence of conditions of globalisation, including the spread of internet commerce.13 While both spheres of government constitutionally have access to most forms of taxation the federal sphere has greater flexibility and enjoys the advantage of the supremacy of valid federal law. Both spheres of government have a significant presence in the tax field. The federal sphere now raises more than half the total tax revenues, largely from individual and corporate tax and payroll tax.14 State practice varies considerably, but relies variously on income and sales taxes, while local government imposes property taxes. Autonomy remains a marked feature of the system, not only in the extent to which each order of government raises revenue for its own purposes but in the relative degree of independence that it enjoys in doing so. Over time, however, the operation in practice of the distribution of competence in taxation in the United States has developed features that also can be described in terms of solidarity, although this terminology is not used. These features of the system typically are driven by the demands of operating discrete and sometimes competing tax systems within what nevertheless is a single,­albeit federal, polity. They take a variety of forms, some of the most significant of which are mentioned here. The growth in federal tax revenues vis-à-vis those of the States, has been complemented by very substantial fiscal transfers from the centre to the States. The bases for these and the manner in which they occur will be elaborated further below; it suffices to note here that in 2010 they accounted for around 37 per cent of the revenues available collectively to the States.15 In an example of a different kind, while federal legislation can inhibit 12

13

14

15

W. Fox, “United States of America”, in A. Shah (ed), The Practice of Fiscal Federalism: Comparative Perspectives (Montreal: McGill–Queen’s University Press, 2007) 345–369, at 356–357. The impact of globalisation on State tax systems, through pre-emption and judicial interpretation, can be gleaned from: Federation of Tax Administrators, “Current fta Resolutions” (2015), http://www.taxadmin.org/current-resolutions (accessed 16 March 2016). Tax Policy Center, “The Numbers: What Are the Federal Government’s Sources of Revenue?” in The Tax Policy Briefing Book, http://www.taxpolicycenter.org/briefing-book/ background/numbers/revenue.cfm (accessed 16 March 2016), calculations for 2010. Tax Policy Center, “State and Local Tax Policy: What Are the Sources of revenue for State Governments?” in The Tax Policy Briefing Book, http://www.taxpolicycenter.org/briefing -book/state-local/revenues/state_revenue.cfm (accessed 16 March 2016).

46

Saunders

State taxation it can alleviate problems for the States as well, by authorising State action that otherwise would infringe the dormant commerce clause,16 by refraining from unnecessary pre-emption and by overcoming other limitations on State taxation authority, of which the imposition of State sales tax on outof-state vendors is one.17 Whether Congress chooses to exercise its authority in these ways depends on politics, rather than principle. In yet another example, the co-existence of federal and State tax systems inevitably prompts questions about their impact on each other. One consequence is the arrangements whereby taxpayers can deduct some State and local taxes when calculating income for federal purposes, at cost to the federal government’s own revenues.18 Another is that States may adopt or adapt elements of the federal tax system for their own taxation purposes by, for example, employing federal definitions of income.19 Despite the potential for considerable interstate tax competition, some practices suggest horizontal solidarity as well, including co-operation in the collection of State taxes.20 Similar patterns can be seen in the design and operation of government spending in the United States. In principle, each order of government spends funds that it raises for itself, for its own purposes. It is accountable to its own voters for both taxation and expenditure, in accordance with practices and procedures applicable under the federal or relevant State Constitution, including any requirements for the budget to be balanced, which are prevalent in State systems.21 The expansion of federal legislative power through judicial interpretation, which affected the distribution of competence over taxation in ways that have already been explored, affected spending as well, by dramatically increasing the matters on which the federal order of government might

16 17 18

19 20 21

For a critique, see N.R. Williams, “Why Congress May Not “Overrule” the Dormant Commerce Clause”, ucla Law Review, 53 (2005–2006) 153–238. Fox, “United States of America”, supra, at 358. S. Maguire and J.M. Stupak, “Federal Deductibility of State and Local Taxes” Congressional Research Service (2015), https://www.fas.org/sgp/crs/misc/RL32781.pdf (accessed 16 March 2016). Interest earned on State and local government securities also are exempt from federal tax: Fox, “United States of America”, supra, at 365. Fox, “United States of America”, supra, at 359. Fox, “United States of America”, supra, at 360, referring to the Multi-State Tax Compact and Commission and the Streamlined Sales and Use Tax Agreement. They take a wide variety of forms, however. Detail is available at: National Conference of State Legislatures, “State Balanced Budget Requirements: Executive Summary” (1999), http://www.ncsl.org/research/fiscal-policy/state-balanced-budget-requirements.aspx (accessed 16 March 2016).

Financial Autonomy vs. Solidarity

47

spend or exercise regulatory authority.22 In a parallel development, the federal power to tax to “pay the debts and provide for the common defence and general welfare of the United States” was held to be the source of a power to spend, for purposes outside federal regulatory authority.23 The stance of the judiciary in these two critical sets of decisions is considered in the next part. These developments were a spearhead for inroads into the autonomy of State spending. They enabled federal institutions to affect State action through mandates imposed on the States in the exercise of substantive powers or through conditions attached to grants made pursuant to the spending power. While the acceptance of grants is voluntary, as a matter of constitutional law, fiscal realities make them hard to reject. Both mandates and conditional grants have further implications for State spending where the mandates are underfunded or grant conditions require matching State funds. Both are common.24 Smith cites a study that calculated the cost to the States of unfunded mandates between 2004 and 2008 as 131 billion usd.25 In these circumstances, there is no doubt that at least some autonomy in spending has given way to interdependence. To what extent interdependence merits description as solidarity is another matter, however. The very fact of fiscal transfers may suggest solidarity, in the sense that transfers relieve State budgetary pressures and offer services to State communities that they would not otherwise receive. The use of mandates also creates in the United States some features of an integrated federation, in the sense that federal initiatives are in fact administered by the States. The common description of the process as “coercive”, however, reflects the extent to which decisions affecting one order of government are made unilaterally by another in a manner that is more adversarial than substantively co-operative.26 It is also relevant in this regard that intergovernmental fiscal transfers in the United States are not 22 23

24 25

26

Tarr, “United States of America”, supra, at 390–391. Frothingham v. Mellon, 262 u.s. 447 (1923); United States v. Butler, 297 u.s. 1 (1936). For more recent, important, authority see South Dakota v. Dole, 483 u.s. 203 (1987) and National Federation of Independent Business v. Sebelius, 567 u.s. (2012); 132 S.Ct. 2566. Smith, “Intergovernmental Relations”, supra, at 515; Fox, “United States of America”, supra, at 349–350. Smith, “Intergovernmental Relations”, supra, at 515, citing National Conference of State Legislatures, “Mandate Monitor Overview (2009), http://www.ncsl.org/ncsl-in-dc/ standing-committees/budgets-and-revenue/mandate-monitor-overview.aspx    (accessed 16 March 2016). The Unfunded Mandates Reform Act 1995 was a congressional response to criticism of unilateralism but has not precluded the practice: Smith, “Intergovernmental Relations”, supra, at 515.

48

Saunders

used as a vehicle for a general equalisation scheme, as occurs in most other federations.27 2.2 Constitutional Review The shifts that have taken place in the United States in the respective emphases on autonomy and solidarity are attributable to a range of causes, including constitutional amendment and political action.28 Judicial interpretation also has played a critical role, however. It has fallen to the courts to interpret and apply the meaning of constitutional provisions expressed in very general terms in a political landscape that has changed continually, sometimes dramatically, over time. For this purpose, courts necessarily draw on assumptions about the principles on which the federation is based, which may be so ingrained that they are not clearly articulated. This part considers whether and to what extent the principles in play in key decisions on fiscal federalism in the United States can be described in terms of autonomy and solidarity. To that end, it examines two groups of decisions: those that have expanded federal power in ways that have a bearing on fiscal federalism and those that reflect an understanding of federalism as a combination of self and shared rule. Judicial review of the meaning and application of the Constitution is responsible for a vast expansion of federal power over the course of the 20th century through vehicles that include the commerce clause, the necessary and proper clause, the Fourteenth Amendment and the spending power. The spending power may be put aside for the moment, as raising more complex questions for present purposes. The story of the other powers is well-known and need not be rehearsed in detail here.29 Suffice it to note that, between them, these powers came to provide a base on which Congress could make laws in relation to most matters, if it had the will to do so, without check by the courts. The explanation for this development, which on the face of the Constitution is surprising, lies in the long trajectory of the history of the United States, from the disappointments of the Articles of Confederation, through the civil war and its aftermath, to the New Deal and the civil rights movement, reinforced by a separation of powers that offers considerable checks and balances within the political branches of government.30 Doctrinal plausibility 27 28 29

30

Smith notes calculations that suggest that States with high per capita income tend to benefit most from federal grants: ibid, at 526. Relevant amendments include the Fourteenth, Sixteenth and Seventeenth: ibid, at 509. For an overview, see K.R. Thomas, “Federalism, State Sovereignty, and the Constitution: Basis and Limits of Congressional Power”, Congressional Research Service (2013), https:// www.fas.org/sgp/crs/misc/RL30315.pdf (accessed 16 March 2016). Garcia v. San Antonio Metropolitan Transit Authority, 469 u.s. 528 (1985).

Financial Autonomy vs. Solidarity

49

was provided by the supremacy clause, also as interpreted by the courts31 and given additional force by the augmentation of central authority in conditions of globalisation. In application to fiscal federalism, these developments underpinned limitations on State taxation power pursuant to federal law and the imposition of federal mandates on the States that directed the manner of the exercise of State authority, often without adequate resources to fund them. While in one sense these decisions might be explained in terms of a principle of autonomy, it is a version of autonomy that involved considerable erosion of the autonomy of the States in practice, while continuing to pay lip-service to State sovereignty in principle. In this regard, therefore, it also is relevant that Supreme Court decisions in relatively recent times have indicated that the legislative powers of Congress are not limitless and that their boundaries will be enforced by the courts.32 The significance of this line of decisions over the longer term is difficult to predict, given the narrow majorities that typically have decided them and the relatively narrow doctrinal bases on which these decisions can be distinguished from those that pull the other way. The spending power also has been a vehicle through which the authority of central institutions has expanded vis-à-vis the States. A general power to spend has been linked with the general power to tax to “provide for the common defence and general welfare of the United States”.33 The power has been held to enable Congress to provide for grants to the States on condition, in areas beyond the legislative powers of Congress. Decisions on the spending power also have determined its characteristics and limits, however, along lines that require joint action, at least in principle. Acceptance of grants is voluntary; the grants power does not provide authority to regulate. And there are some limits on the conditions that can be attached, in consequence either of the character of the grants power or the requirements of the Tenth Amendment. Following the reasons in Dole, conditions must relate to the national program for which the grant was made and must not amount, in effect, to coercion.34 Sibelius further reinforced the reality of limits, in terms that might be understood by reference to either of the Dole criteria, by severing the provision in the challenged legislation that would have withdrawn a State’s Medicaid funding if it 31 32

33 34

E. Chemerinsky, “Empowering States when It Matters: A Different Approach to Preemption”, Brooklyn Law Review, 69 (2004) 1313–1333. Significant decisions include United States v. Lopez, 514 u.s. 549 (1995), United States v. Morrison, 529 u.s. 598 (2000), National Federation of Independent Business v. Sebelius, 567 u.s. (2012); 132 S.Ct. 2566. u.s. Const, Art. I(8); United States v. Butler, 297 u.s. 1 (1936). South Dakota v. Dole, 483 u.s. 203, 207, 211 (1987).

50

Saunders

did not accept the extension to Medicaid in the Patient Protection and Affordable Care Act. These aspects of the doctrine more clearly reflect the Supreme Court grappling with complex questions of the scope of central power in a federal system and might be understood in terms of solidarity on this basis. And the practices to which they give rise are clear instances of interdependence, albeit of a kind that correspond to a dualist conception of federalism. The second group of decisions more explicitly involve the Court grappling with federal principle, as a means of understanding the federal design of the Constitution, reinforced by the Tenth Amendment. One significant line of doctrine recognised a degree of immunity of each order of government from legislation of the other at a very early stage.35 The original doctrine of McCulloch v Maryland36 has long since broken down, partly on the basis of the logic of federal supremacy and partly on the assumption that the political process was adequate to protect the States from inappropriate federal interference of a kind that the courts found it difficult to define.37 The immunities doctrine has at least two legacies that are relevant for present purposes, however. The first concerns taxation. The federal government is immune from State taxation and others dealing with the federal government also have a degree of immunity from State taxation, if they are singled out by State law on that basis.38 For their part, the States also have a limited immunity from federal taxation, to the extent to which it discriminates against them (or others dealing with them) or, perhaps, it threatens their existence or capacity to function.39 A second continuing legacy of the immunities doctrine concerns what in the United States has become known as “commandeering”. Federal legislation cannot compel State legislatures or executives to regulate or administer on behalf of the federal order of government.40 Insofar as the latter holding prevents the federation requiring States to administer federal legislation it inhibits the emergence of integrated federal arrangements of a kind that are considered in the next part. Judicial doctrines of this kind stem from the need for courts to resolve problems that arise from the operation of the laws of different orders of government, 35

Another concerns the extent to which plaintiffs can sue a State under federal law, consistently with the Eleventh Amendment: Seminole Tribe of Florida v. Florida, 517 u.s. 44 (1996); Alden v. Maine, 527 (u.s.) 706 (1999). 36 17 u.s. (4 Wheat.) 316 (1819). 37 Garcia v. San Antonio Metropolitan Transit Authority, 469 u.s. 528 (1985). 38 New York v. United States, 326 u.s. 572 (1946); Washington v. United States, 460 u.s. 536 (1983). 39 New York v. United States, 326 u.s. 572 (1946); South Carolina v. Baker, 485 u.s. 505 (1988). 40 New York v. United States, 505 u.s. 144 (1992); Printz v. United States, 521 u.s. 898 (1997).

Financial Autonomy vs. Solidarity

51

each with limited legislative competence, in the same polity. To that extent, they might be described in terms of solidarity. The form of solidarity is distinctive, however. It starts from the premise of autonomy and assumes the institutional dualism of federal design. It works through the imposition of constraints on self-interested behaviour, rather than through development of a notion of the collective interest. 3

Germany

3.1 Solidarity and Autonomy The German federation stands in contrast to that of the United States in terms of federal and other institutional design, the historical, legal and geopolitical context in which it operates and many of the principles on which it is based, including the weight that it accords to solidarity and autonomy respectively. Of central importance for present purposes is the federal division of authority along horizontal as well as vertical lines, causing Germany to be described as an “integrated” federation, in contrast to the dualism of the United States.41 The interdependence that integration necessarily involves is augmented further by the institutions of the Bundesrat, though which Land governments participate in central decision making that touches on Land interests and the execution of federal legislation by the Länder as of right, which complements the concentration of legislative power at the centre. Both these distinctive approaches to federal design have deep roots in the German history of the 19th century and in particular the second German Empire, from 1871.42 Integrated federalism is broadly compatible with the codified German legal system and with the German attachment to equivalence in living standards and the social state.43 The influence of the principle of solidarity is manifest. As the following analysis shows, however, the principle of autonomy also underpins German federalism, including fiscal federalism, in a symbiotic relationship with solidarity. The particular forms that autonomy takes are adapted to the context of German federal design and the relative emphasis on it has varied over time. 41

42 43

S. Oeter, “Federal Republic of Germany”, in K. LeRoy and C. Saunders (eds), Legislative, Executive and Judicial Governance in Federal Systems (Montreal: McGill-Queen’s University Press, 2006) 135–164, at 136, 141. W. Heun, Constitution of Germany: A Contextual Analysis (Oxford: Hart Publishing, 2011), 51. C. Waldhoff, “Federalism—Cooperative Federalism versus Competitive Federalism” in H. Punder and C. Waldhoff (eds), Debates in German Public Law (Oxford: Hart Publishing, 2014) 117–130, at 117, 120: Heun, Constitution of Germany, supra, at 60.

52

Saunders

The general scheme for the allocation of competencies in the German federation involves the enumeration of an extensive range of central legislative powers leaving the remainder to the Länder.44 Central power may be either exclusive or concurrent.45 With only a few exceptions, central legislation on a matter in the concurrent list removes power from the Länder. In consequence, in practice, the Länder have limited legislative authority of their own, primarily in relation to education, culture, law and order and regional planning.46 They may also collaborate with the federation in joint tasks.47 Consistently with the integrated design of the federation, however, and significantly for present purposes, the Länder have the authority to execute all federal laws in their own right, other than those explicitly denied to them by the Constitution. They may in any event be delegated to execute the latter by federal commission.48 It follows from this arrangement that significant responsibility for public expenditure falls to the Länder and to local government. How this is funded depends on the fiscal constitution. The fiscal constitution follows the overall scheme for the division of competencies, elaborated in ways that are necessitated by its distinctive characteristics, in a discrete chapter of the Basic Law.49 The requirements of the Basic Law, in turn, operate within the framework of European Union law. In contrast to the general competitive assignment of tax competence that operates in the United States and some other federations, the Basic Law is relatively clear on which order of government has responsibility for which taxes and deliberately avoids double taxation.50 Authority to legislate for the vast bulk of taxation is assigned to the federation, as either an exclusive or a concurrent power, leaving only minor local taxes to the Länder and local government.51 The Basic Law provides for a high degree of interdependence in taxation, however, so that this is by no means the end of the story.52 Most taxes for which the federation legislates are collected by the Länder, including personal and corporate income 44 45 46 47 48 49 50 51 52

Grundgesetz (gg), Art. 70. gg, Arts. 73, 74. L.P. Feld and J. von Hagen, “Federal Republic of Germany”, in Shah, The Practice of Fiscal Federalism, supra, 126–151, at 126, 133, 136. gg, Ch. viiia. gg, Arts. 83, 85. gg, Ch. x. gg, Art. 105. Literally, “local taxes on consumption and expenditure…not substantially similar to taxes regulated by federal law”: gg, Art. 105(2a). The results are nicely summarised in Table 4 in Feld and von Hagen, “Federal Republic of Germany”, supra, at 141–142.

Financial Autonomy vs. Solidarity

53

tax and the value added tax (vat). Legislative competence does not determine the right to the income from particular taxes, under arrangements considered below. The consent of the Bundesrat is required for federal legislation relating to taxation where the proceeds accrue to the Länder or municipalities.53 Federal legislation may provide the base for taxation at the sub-national level, even where there is autonomy in setting the rate.54 In such a system, the allocation of tax revenues is of critical importance. Appropriately, the framework of rules within which this occurs is provided by the Basic Law. The principal determinant of the allocation of revenue is need, in the sense of responsibility for expenditure, at least in theory.55 The proceeds of some taxes are assigned exclusively to one order of government or another. Others are shared, through a process that incorporates a degree of flexibility and thus involves negotiation, subject to criteria set out in the Constitution itself.56 The horizontal distribution of revenue between Länder varies between taxes but, as a generalisation, involves extensive equalisation directed to redressing differences in fiscal capacity.57 Vertical transfers further support equalisation.58 Transfers on condition are relatively rare, with financial assistance for investments as an exception.59 The complex interdependence of these arrangements overshadows the elements of autonomy, which nevertheless are significant. The very existence of the fiscal contribution underpins the autonomy of both spheres of government and in particular of the Länder, which necessarily are more vulnerable.60 Both orders of government are “autonomous” in budget management, subject to the strictures of the European Union.61 The Länder execute federal legislation as a matter of constitutional right.62 The oversight for which the Basic Law provides is limited and involves the Bundesrat on key points.63 The federation is constitutionally obliged to provide the resources when the Länder act on federal commission.64 Conditional transfers are limited. 53 gg, Art. 105(3). 54 Feld and von Hagen, “Federal Republic of Germany”, supra, at 134. 55 Heun, Constitution of Germany, supra, at 65; Feld and von Hagen, “Federal Republic of Germany”, supra, at 134. 56 See generally, on both these points, gg, Arts. 106, 106a, 106b. 57 Heun, Constitution of Germany, supra, at 66–67. 58 Feld and von Hagen, “Federal Republic of Germany”, supra, at 134–135. 59 gg, Art. 104b. 60 Heun, Constitution of Germany, supra, at 65. 61 gg, Art. 109. 62 gg, Art. 83. 63 gg, Art. 84. 64 gg, Art. 104a(2).

54

Saunders

Autonomy is never complete, however, but is always qualified by the assumptions of the German federal state and the interdependence of the several orders of government. Both orders of government are involved in most aspects of fiscal decision-making, separately or through the institution of the Bundesrat. The federation provides the framework within which State budgetary autonomy operates. There is some federal oversight of Land execution of federal law. The federation and the Länder have responsibilities to each other, under a federal constitution in which each has a measure of protected autonomy of its own. While the position of federal order is immeasurably stronger, a wide range of decisions affecting the interests of the Länder require the consent of the Länder collectively through the institution of the Bundesrat. Most federations experience shifts in the relative emphasis on solidarity and autonomy in fiscal arrangements over time. In Germany, a minor but significant shift towards autonomy took place in two waves of constitutional review in the first decade of the 21st century. One of the concerns that sparked the move was the range of matters that required Bundesrat consent, leading to stalemate between federal institutions. The logical quid pro quo for reducing the influence of the Bundesrat was to increase the autonomy of the Länder individually, which also had the additional, potential advantage of disentangling the responsibilities of the federation and the Länder, at least to some degree.65 The first round of changes in 2006 eliminated the federal competence for the enactment of framework legislation; increased the legislative competencies of the Länder in some significant but still limited respects; conferred authority on the Länder to deviate from some federal administrative requirements—in addition to federal legislation enacted pursuant to Article 72(3)—augmented aspects of federal power; and reduced the requirement for Bundesrat consent, which nevertheless remains significant.66 From the standpoint of the fiscal constitution, the changes extended the tax competency of the Länder by empowering them to determine the rate of property taxation;67 added the provision now in Article 104b to permit federal grants to the Länder for prescribed forms of investment; and removed the need for Bundesrat consent to certain forms of financial grants.68 A second round of changes, specifically designed to 65

66 67 68

The planned changes are summarised in D.P. Kommers and R.A. Miller, The Constitutional Jurisprudence of the Federal Republic of Germany (3rd ed., Durham: Duke University Press, 2012), 79. Ibid, at 121–123; Feld and von Hagen, “Federal Republic of Germany”, supra, at 127; gg, Art. 72. gg, Art. 105(2a). Kommers and Miller, The Constitutional Jurisprudence, supra, at 97.

Financial Autonomy vs. Solidarity

55

deal with fiscal issues, made relatively little difference for present purposes.69 The thrust of these changes has some interest for their focus on autonomy. Equally, however, their limited range offers insight into the German commitment to solidarity. 3.2 Constitutional Review By comparison with the Constitutions of the common law federations of the United States, Canada and Australia, the German Basic Law is relatively easy to amend, at least in technical terms. Appropriately, amendment is sufficiently difficult to require substantial political consensus.70 The amendment requirements are not insuperable, however, and have been used with success more than 50 times.71 Of the changes that have been made, some of the most frequent and significant have affected the fiscal constitution.72 It follows that the fiscal constitution in Germany is by no means as reliant on judicial review for its evolution over time and its adaptation to new conditions as is the case in the United States. Constitutional review nevertheless is and has been significant in Germany, generally and in relation to the fiscal constitution.73 On occasions, there has been creative interplay between the two. The history of the evolution of the form and operation of the concurrent powers of the federation offers an example. Article 72 of the Basic Law conditions the circumstances in which the concurrent powers are exercisable by the federation by reference, in essence, to the need for federal regulation. Prior to 1994, this criterion was treated as effectively non-justiciable by the Federal Constitutional Court.74 Following unification, however, the clause was revised with a view to strengthening the autonomy of the Länder and was held to be mandatory, enforceable through judicial review.75 In 2006 it was revised again, to its present form, which applies the requirement for legislation for the “establishment of equivalent living conditions […] or the maintenance of legal or economic unity” to make “federal regulation necessary in the national interest” to only ten of the concurrent powers, thus reducing the occasions for judicial review. 69

Waldhoff, “Federalism”, supra, at 125; Kommers and Miller, The Constitutional Jurisprudence, supra, at 99. 70 gg, Art. 79, requiring a two-thirds vote in both the Bundestag and the Bundesrat. 71 Heun, Constitution of Germany, supra, at 22. 72 Ibid, at 65. 73 Feld and von Hagen, “Federal Republic of Germany”, supra, at 131. 74 Waldhoff, “Federalism”, supra, at 122. 75 BVerfGE 106, 62, Geriatric Nursing Act case (2002). Kommers and Miller, The Constitutional Jurisprudence, supra, at 132.

56

Saunders

Decisions of the Federal Constitutional Court in Germany have been seminal in articulating the values of solidarity and autonomy in the German federation and elaborating the relations between them, in the context of the principle of comity. In the First Broadcasting case, in 1961, the Court identified a range of circumstances, common in federations, in which the principle of comity was attracted.76 These included the implications of variation in financial capacity, of which more below. Comity also created obligations, however, in situations where mutual agreement was needed; where the effects of a law transgressed Land boundaries; to require the Länder to implement international treaties; and in relation to the “procedure and style of negotiations” between the orders of government.77 It may be, as the run of cases indicate, that the principle is “accessory” in effect and that its various applications have had their primary impact on the Länder.78 Nevertheless, it offers a standard and sets expectations against which the behaviour of both orders of government can be measured. It also operates as a caution. A principle that has been invoked by the Court in the past may be invoked again in the future, if the necessary conditions are deemed to arise. The principle of comity has been used with express reference to autonomy and solidarity in the resolution of disputes over the meaning and application of the financial equalisation provisions of the fiscal constitution. Financial equalisation creates tension between federal units in most federations. In Germany, these were exacerbated by the fiscal disparities between Länder in the wake of unification.79 The successive financial equalisation cases highlighted the difficulty of striking a balance between the budgetary autonomy of the Länder as reflected in, for example, Article 109 of the Basic Law and the obligations created for the Länder by a “mutually supportive federal community”.80 In the Third Equalization Case the Federal Constitutional Court confirmed that these requirements stopped short of weakening “the ability of the contributing Länder to meet their obligations” or “levelling out the finances of the Länder”.81 Solidarity required the reduction of differences, but it could not “eliminate them” entirely, consistently with the maintenance of autonomy, as understood by the Court.82 76 77 78 79 80 81 82

BVerfGE 12, 205; extracted in Kommers and Miller, The Constitutional Jurisprudence, supra, at 90. Kommers and Miller, The Constitutional Jurisprudence, supra, at 92–93. Ibid, at 95. Kommers and Miller, The Constitutional Jurisprudence, supra, at 103–104. BVerfGE 101, 158, Financial Equalization iii case (1999), quoted in Kommers and Miller, The Constitutional Jurisprudence, supra, at 102. Kommers and Miller, The Constitutional Jurisprudence, supra, at 102. Ibid.

Financial Autonomy vs. Solidarity

4

57

Conclusions

The quite different examples of the United States and Germany suggest that the principles of autonomy and solidarity are likely to be immanent in the fiscal arrangements of all federal systems. Their influence, in combination, is the inevitable consequence of the federal form of the state, which by definition recognises and respects a sphere of constitutional authority for each order of government within a single polity. These observations are generalisations, however, to which several caveats apply. First, fiscal arrangements are not necessarily perceived in these terms in the discourse that prevails in different federal states. Thus, in the two cases canvassed in this chapter, the terminology of autonomy and solidarity is familiar in Germany but is not in common use in the United States. Secondly, as these cases also show, emphases on autonomy and solidarity differ between federations, sometimes in marked degree. Thirdly and importantly, the principles may manifest themselves differently in different federations, under the influence of a range of contextual factors that include the design of the federation itself. In the dualist federation of the United States, for example, the undoubted instances of solidarity are coloured by the competitive and adversarial context in which the federation operates. In Germany, on the other hand, understanding of the demands of autonomy are deliberately tailored to and limited by core expectations of solidarity. In either case, the principles of autonomy and solidarity may be examined by reference to the federal system overall or to the important subset of federal arrangements that comprises the fiscal constitution. The latter has additional significance, in the sense that the manner in which autonomy and solidarity are given effect in the fiscal constitution is central to procedures for the democratic responsibility of the institutions of government as well as to the effective functioning of the federation. The federal fiscal arrangements in any federation can be expected to be unique to some degree. For the reasons advanced earlier, the principles of autonomy and solidarity will be relevant to each. As a generalisation, however, most of the federations of the world rely on solidarity to a greater degree than the United States and place greater emphasis on autonomy than appears to be the case in Germany. In terms of the spectrum of federations to which reference was made in the introduction, therefore, the majority of the federal type systems of the world are distributed over the intermediate sections of the spectrum, between the two more extreme cases examined in this chapter. No attempt is made here to range fiscal arrangements along the spectrum, which would require a degree of analysis beyond a single book chapter. As a guide, however, at least three primary points of difference may distinguish

58

Saunders

other federations from either or both of the two paradigm cases examined here. These may be reflected in the design of a federation, in its operation in practice, or both. First, allocation of general taxing powers to both levels of government to be exercised on a competitive basis is relatively rare. Even a federation that seeks in principle to enable each order of government to tax for its own purposes is likely to attempt to identify the taxes most appropriately exercised at different levels, subject to the constraints that necessarily apply in a federation. Whether through the formal allocation of tax competence or through the operation of the system in practice, there is a tendency for taxation to become increasingly centralised as, indeed, also occurred in the United States. This is by no means invariably the case, however.83 And the allocation of authority to impose taxation is only part of the story, in any event. To adequately evaluate the interplay of autonomy and solidarity it may be useful to distinguish authority to set the rate as well as the base of taxation, to determine whether there are arrangements for harmonisation of the taxes of the sub-state order of government and to evaluate the extent to which these are consensual. Secondly, general and conditional revenue redistribution from the centre to the subnational entities is a common phenomenon in most federations. In many, it is the necessary and anticipated corollary of tax centralisation. Typically, it is used in any event to ensure a measure of equalisation.84 Federations formed over the latter part of the 20th century are likely to make specific constitutional provision for such transfers. Whether based on an express provision or derived through interpretation, however, the challenge is to inhibit the extent to which transfers are used to erode the autonomy of the sub-national order of government in the exercise of its constitutional authority. To this end, some federations create agencies at arms’ length from the central government to advise on key intergovernmental fiscal decisions. The Finance Commission of India is an example.85 Whether these are effective for the purpose depends on the political dynamics of the federation concerned. Where the meaning of a spending power is elaborated through judicial decisions, it is likely to be made clear that acceptance of grants is voluntary, even where the conditions

83

84 85

A. Shah, “Comparative Conclusions on Fiscal Federalism”, in Shah, The Practice of Fiscal Federalism, supra, 370–394, at 370, 380. Shah identifies Switzerland, Canada and Nigeria as federations with relatively decentralised taxation arrangements. Ibid, at 388. G. Rao, “Republic of India” in Shah, The Practice of Fiscal Federalism, supra, 152–178, at 152, 162.

Financial Autonomy vs. Solidarity

59

attached to them are binding. Where the fiscal imbalance is severe, however, as in Australia, this protection for State autonomy may be illusory.86 A third point of departure from the paradigm cases lies in the manner of the distribution of competences, including fiscal competences. A number of the federal systems of the world mix the vertical and horizontal distribution of competences to a greater degree than either the United States or Germany. In some cases, of which India is an example, the authority of the sub-state order of government to administer federal legislation is delineated by the allocation of concurrent legislative power. Typically, however, these features of an integrated federal design are unaccompanied by the features that make them work effectively in Germany: the Bundesrat in which Land governments participate in central decision-making; guarantees of Land autonomy in the implementation of federal legislation; secured funding to match constitutional responsibilities. Other federations, of which Australia is an example, are dualist in design but partly integrated in practice, through the effect of conditional grants in areas of sub-state responsibility. These arrangements erode effective sub-state autonomy while leaving the formalities of dualism in place. One consequence of these departures is that each of the federations of the world mixes autonomy and solidarity in different proportions and in each of them autonomy and solidarity may play out in distinctive ways. The paradigm cases remain instructive, nevertheless, in the sense that they offer contrasting models, each with a logic of its own, that assist in analysing and understanding the infinite variations in the design and operation of federal fiscal arrangements that can be found elsewhere.

86

C. Saunders, “Australia”, in Kincaid and Tarr, Constitutional Origins, supra, 12–47, at 13, 36. The imbalance in Australia is characterised by a mismatch between the revenues raised by the centre and the expenditure responsibilities of the States, making the States dependent on revenue transfers, for which the Constitution makes limited provision.

chapter 4

The Practicalities of Economic Federalism: A Critical Review of How to Apply the Lessons of Fiscal Autonomy in Practice Andrew Hughes Hallett 1

Introduction

The European Union, and more particularly the Eurozone, has always been a “would be, maybe” federal union; and since the financial crisis of 2008 has become a de facto emerging economic federation. However, the weak economic performance since the introduction of the Euro and a disastrous performance since the debt crisis of 2008–2012—more evident in the difficulty of getting out of that crisis, than in the fact that it fell into it in the first place—has revealed the eu to be an accidental and very incomplete economic federal union. The real difficulty, this paper will argue, is that, despite reams of technical analysis and political advice on how specific policies should be designed and conducted within the existing framework, there is no conceptual framework to guide the policymaking process or its implementation, and little coherence within or between the existing institutions. Some forward thinking is needed here. This paper is concerned with this conceptual framework issue. In the delegation literature, principally that concerned with monetary policy and the design of central bank operating procedures, but also in the literature on principal-agent models, this distinction is generally characterised as instrument independence vs. target independence. In the former, subnational decision makers (or decision makers delegated to manage a certain sector of the economy) act as agents for the central/federal government authorities and have no responsibility other than being accountable for having reached the criteria that others have set for them in the most efficient manner possible and at least cost—without wasting resources, creating additional uncertainties or absconding with the money. They have no responsibility for designing those criteria; or for considering whether they are the best criteria, the best targets, and best policy priorities; or whether the policymakers in question have been allocated the most effective policy instruments to do the job. Agents under instrument independence have the freedom to use their policy instruments as they see best and are held accountable for having done so © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_006

The Practicalities of Economic Federalism

61

in the most efficient manner as defined by specific pre-assigned criteria. But they are not responsible for the overall performance of their economy, or the wider economy of which it is a part, or spillovers on others, or whether those criteria are consistent and will lead to the best performance. In the absence of coordination, a coincidence of preferences and detailed knowledge at the federal level of the region’s exact circumstances, those criteria will not achieve that best performance. Target independence is quite different. Here the subnational policymakers set both the criteria and the priorities for a good performance, and pick the instrument values so as to produce the best performance for themselves, within the wider economy taking into account local conditions, preferences and spillovers. There is still a potential coordination problem if there is a conflict between regional optimality and optimality in the wider economy. But that is always the case, even for a fully independent economy within the world economy. Some safeguards may need to be inserted to deal with that problem; but we would need to do so anyway to ensure that we all benefit from the spillovers from a better performing federal economy and a better performing regional economy. In this case, subnational policymakers are held accountable for their instrument settings, and held responsible for the performance of their own economy. And they have the capacity to create better outcomes because a wider set of choices are available in a problem which is otherwise the same as the instrument independence case. To my knowledge, no-one has ever used the distinction between instrument independence and target independence in the area of fiscal policy; or used it to define and mark the differences between systems with full fiscal autonomy and those with partial autonomy but still heavily dependent on grants, assigned tax revenues, or centrally determined shares of domestically raised taxes. Examples of the former, with target independence and de facto devolution max include Canada, some Spanish autonomous communities and some Italian regions, the Channel Islands or Isle of Man; examples of the latter are Australia and Germany with their reconciliations for subnational revenues through the Australian grants commission and Germany’s Finanzausgleich process respectively.1 1.1 Scotland as a Specific Example The Scottish Government faces the challenge of supporting economic activity in the face of severely constrained public finances. This has opened up a 1 This discussion begs the question: why decentralise economic policymaking in the first place? Section 2.3 contains a detailed discussion of that point.

62

Hallett

debate with the uk Government over how much autonomy Scotland should be granted in order to be able to address the problems she faces. This argument forces us to draw a clear distinction between a funding mechanism in which ministers are held accountable (in a book keeping sense) for raising and spending a defined stream of money on a defined and pre-specified set of objectives; vs. a regime which gives the Scottish Parliament the capacity and the responsibility to raise and spend the sums of money that they think would most improve the performance of the economy and standard of living of its citizens. Many commentators have criticised the strategy currently used to finance the Scottish Parliament, both the block grant system and the very limited amount of fiscal autonomy devised in the Scotland Act of 2012 and its successors. Our argument is developed along three lines. First, after setting out the fiscal federalism framework within which the analysis is set, the paper offers a critique of the rules for a partial devolution of income tax competence and minimal borrowing powers.2 The second and principal theme of the paper is to elaborate a model of fiscal federalism where comprehensive powers over taxation and spending are devolved, but remain consistent within the macroeconomic framework and constitutional integrity of a Federal Union. Finally, we provide a review of the empirical evidence which indicates that while fiscal autonomy per se may not result in an increase in long run economic growth rate, it does imply that enhancing the fiscal competence and responsibility of subnational governments typically results in productivity gains and hence a higher level of gdp per head. That means the population is permanently richer than before, even if thereafter their incomes continue to grow at the same rate. During the adjustment period of course, the economy has to grow faster to reach that new higher level growth path. It turns out that these improvements can be achieved through devolved tax powers, but not through devolved spending powers or shared taxes.

2 In this paper, I use the term fiscal federalism to denote a system in which there are explicit risk sharing or fiscal transfer mechanisms between regional or national economies. And fiscal autonomy to mean a system in which subnational policymakers can make their own decisions using local policy instruments to improve the performance of their own economies—­ which may or may not include reducing the risk of adverse shocks or cyclical movements. But with limited borrowing and variable revenues among member governments in the union, institutional measures to support risk management, risk sharing, fiscal oversight and a central budget are essential for the ability to make credible, stabilising transfers.

The Practicalities of Economic Federalism

2

63

Alternative Funding Mechanisms

Since circumstances will be different in different countries, we continue with the Scottish example in this section. Scotland is an economy in a long standing currency and political union with the rest of the uk, but with no freedom to choose its own economic or financial policies. No federalism therefore. Growth has been lower than in the uk as a whole, and employment and productivity have been lower. The issue then is to find a financing regime that would help secure growth and jobs. Simple funding mechanisms, such as a block grant or straight forward tax assignments, cannot give the government the capacity to improve the economy on a systematic basis. Instead they are a means to secure Scotland a pre-assigned stream of funding. The funding options are: block grant funding (such as the existing Barnett formula); tax assignment schemes (including apportioned taxes); partial fiscal autonomy; fiscal federalism (i.e., where specific taxation and spending powers are allocated to subregional governments together with risk sharing and/or transfer arrangements); and finally full fiscal responsibility. The Scotland Act would imply a small amount of autonomy, were it not for the fact that spending will be tied to fluctuating revenues defined by a forecasting rule in the hands of the uk Treasury (thereby reducing the opportunity for independent decision making). Borrowing would be the usual way to smooth out public spending. However, that has effectively been ruled out because the maximum debt for borrowing for spending purposes has been restricted to less than 0.5 per cent of gdp. The first alternative is to apportion taxes (for example, value-added taxes (vat)). This was not the original intention behind the attempt to find an alternative and improved funding mechanism. Indeed the original instructions were to consider full fiscal autonomy and the implications of the Steel commission report in 2006.3 But after a time, those options were ruled out. This paper therefore brings fiscal devolution at different levels back into consideration. 2.1 Partial Fiscal Autonomy At this point any government faces a dilemma. The existing block grant system has been widely criticised, not least by the uk government.4 On the other 3 The Steel Commission, “Moving to Federalism—A New Settlement for Scotland: Final Report”, Liberal-Democrat Party (2006). 4 House of Lords—Select Committee on the Barnett Formula, The Barnett Formula: Report with Evidence, 1st Report of Session 2008–09, hl Paper 139 (2009), www.publications.parliament .uk/pa/ld200809/ldselect/ldbarnett /139/139.pdf (accessed 02 March 2016). See I. McLean

64

Hallett

hand, the Scotland Bill mechanism would be unworkable as it stands because it requires information on future tax revenues that no policymaker can possibly have when making spending allocations. And, by denying borrowing for current or entitlement spending, it contains no mechanism to reconcile contractual spending (most of the budget) with variable revenue flows until several years after the event. Proposals of this kind therefore retain the very defect which the Bill was intended to avoid (and which led them to reject the option of allocating North Sea revenues to the Scottish budget): namely, revenue volatility. Moreover, the attempt to fix this problem by using Treasury forecasts of future tax revenues, allowing borrowing, and reconciling the forecasts with actual revenues later, will inevitably introduce three further difficulties: new grounds for quarrels between the centre and subnational governments, a long term deflation bias, and a net loss of devolution.5 There is an additional problem, generic to any partial fiscal devolution. Since taking office in 2010, the uk government has announced that it will raise the tax-free allowance on income taxes—to be funded by increases in other taxes (Employee National Insurance contributions and increased Capital Gains taxes in this case). This will alter the balance of taxation in the economy and reduce the revenues raised by income taxes. That would have a major impact on the finances of Scotland because the current proposal only assigns revenues from income taxes to the subnational government. As a result, the new allowance will reduce the revenues going to the Scottish Government while the revenue from higher National Insurance and Capital Gains will go straight to the uk government. However, no such effect will be observed in the rest of the uk since the loss in income tax revenues is compensated by increases in other taxes. This situation illustrates an important general principle: revenues that depend on a shared tax base will vary up and down with decisions taken by another tax authority, for which the subnational government has neither responsibility, nor control. If a return to a block grant subsidy is not acceptable, the only way out is to devolve the entire tax code for a particular tax: that is, decisions on all the income tax rates, tax bands and the tax base. et al, Fair Shares? Barnett and the Politics of Public Expenditure (London: Institute for Public Policy Research, 2008). 5 A deflation bias because income taxes grow slower than the other sources of revenue that drive the spending allocations that make up the block grant, and because historically income tax forecasts have overestimated actual outcomes—so too much will be taken off the grant in lieu of own resources. A loss of devolution because spending decisions will be driven by what someone else says you can spend based on the status quo, not what you might have to spend to improve the economy’s performance.

The Practicalities of Economic Federalism

65

The other problem with partially devolved taxes, in the absence of borrowing powers, is that a shared tax implies a retreat from devolution. If Scotland is to continue to meet her obligations, the Scottish government will have to raise or lower its own tax rates to compensate for changes made elsewhere—even if that conflicts with the Scottish economic strategy. Similarly, if the economy is hit by any shock, internal or external, that lowers its revenues, in the absence of borrowing powers the Scottish government will have to adjust its spending plans or raise/lower its tax rates even if to do so conflicts with its wider economic goals. Either way devolution is de facto lost. The only way out for the Scottish government is to seek greater autonomy (more taxes under its control) since the problem is caused by having assigned too few other taxes to Scottish control which could be used to compensate unplanned and unwarranted changes in revenue. In other words, a small amount of autonomy will automatically generate pressure for further autonomy. The resolution to this dilemma is to add subsidies to ensure the outcomes remain revenue neutral. But that would take us straight back to a system where the revenue flow is determined by decisions made centrally on an underlying grant. It is hard to imagine such an arrangement could be open, transparent and verifiable. 2.2 Efficiency and Political Accountability More generally, more devolution is better than less for two reasons. First because wider devolution is better as a matter of system design (internal diversification to stabilise revenues, and insurance against external policy changes); second because it creates a responsibility as well as accountability for creating a better economic performance. Since both are desirable, more is better than less. But questions remain. How do we attain greater economic efficiency? Legal accountability alone creates no responsibility for extra efficiency. And are the spillovers to/from the economies in the rest of the Union consistent with a stable macroeconomic framework? 2.3 Fiscal Federalism6 There is an extensive literature on the economics of fiscal decentralisation, or “fiscal federalism”, starting with the work of Musgrave in the 1950s and Oates in the 1970s. The conclusion of that literature is that decentralisation, hence fiscal responsibility, typically leads to better economic performance—both in theory and in practice. The argument is usually put in terms of economic efficiency: decentralisation provides an efficient way to correct various forms of 6 See fn. 2, supra.

66

Hallett

market failure, ensure an equitable distribution of resources, and stabilise regional economies and employment. To deny that, one must show that a single, centralised, monolithic government could or indeed would succeed in maximising social welfare across all regions out of a sense of benevolence, despite electoral pressures and special interests in a multi-region democracy. In practice, where there are regional differences in structure or resources, or in the way economies respond to shocks or policies, or in a region’s position in the cycle, it would be very hard (if not impossible) for any one government to come up with one set of policies that satisfies everyone in the sense of maximising wellbeing. Different regions will require different solutions to suit their particular circumstances. It would be doubly difficult if: • the central government has less precise information on local needs/conditions; or • if its policies are helpful in one place, but have adverse spillovers on another; or • if central government is less accountable because of political distance from the regions; • is subject to special interest groups because of the electoral calculus of majority rule. As a result the provision of public goods, and stabilisation of employment would be inefficient, and the performance of the regional and aggregate economies below potential. This line of argument is sufficient to demonstrate the decentralisation theorem of Oates:7 in multilevel governments, each level of government (including central government) will maximise social and economic welfare within its own jurisdiction. That provides a higher level of economic welfare than can be attained in a regime in which central government provides a uniform set of policies and public goods for all—since subnational policymakers can always choose to replicate the central government’s common policies if they wish to do so. This then is the case for devolved policies: decentralisation will always produce better and more efficient (or at least as good) outcomes for all, including for the central government—subject to not devolving so far as to create diseconomies of scale in the delivery of public services. To make the point another way, decentralisation automatically implies a larger number of policy instruments/decisions are available to achieve the same policy goals and their component parts. That will lead to better outcomes since governments could 7 W.E. Oates, Fiscal Federalism (New York: Harcourt Brace Jovanovich, 1972).

The Practicalities of Economic Federalism

67

always have chosen the same policies that they had before decentralisation. But in general, with the extra instruments, they will be able to improve on that. Hence it is a mathematical certainty that you can do better with decentralised decisions; the only reason you might not is if there were severe diseconomies of small scale, or if sub-national entities (or countries in the Eurozone) found it worthwhile to impose serious spillovers on their neighbours. The standard response to this argument has been to say that the same result can be achieved by a common set of policies plus lump sum transfers (subsidies, grants, side-payments) to each region, chosen to allow the same local outcomes. This may be true: but it just reproduces the central grant type system that has been used until now. It also means that grants may have to be more generous in some regions than others; and it imposes no accountability on those who raise the grants, or on those who spend them. Moreover, if grants are to respond to local conditions, and if there are structural or cyclical differences between regions, or more accurate information at the local level, or if central government finds itself fiscally constrained, then we will have to ask subnational governments to decide on those grants for themselves—giving rise to even less accountability and more perverse incentives. It would therefore be better to ask subnational governments to raise and spend their own revenues. At least they are then accountable to their own electorate and must bear the pain of their spending decisions. But they can still profit from more precise information on local conditions, differences, preferences, without the blocking coalitions of a centrally determined system. The result would be more effective policies: higher growth and higher employment than is possible under a grant or assigned taxes. 3

Policy Allocations

This argument so far makes the case for fiscal autonomy in a federal system. The next question is to decide how far autonomy should go. Subnational governments will recognise that they have circumscribed abilities to influence local employment or prices, or play an active stabilisation role, or to borrow. Central governments therefore retain a defining role to ensure coordination, monetary stability, stabilisation; and in competition policy and providing financial regulation and financial stability (albeit with representation from the regions). 3.1 Allocating Policy Instruments and Responsibilities This argument suggests a natural way to allocate policy instruments between central and subnational governments. The power to tax immobile factors,

68

Hallett

property, natural resources should be allocated to subnational governments. They should also be able to set user fees, benefit taxes, and spending; i.e., have the power to raise income, sales, corporate or business taxes, and the social security taxes that affect mobile factors, production costs and competitiveness; and also control spending on health, education, police and justice, infrastructure, research and development (r&d), innovation, and development. The latter are all instruments that affect productivity growth in the short and long run, and unit labour costs—and hence employment and the cost of doing business. Notice that taxes on mobile factors are included because the ability to set taxes equal to the marginal cost of providing services at the subnational level is necessary if households and firms are to choose locations that provide the most efficient level of services—and to give governments a direct incentive to supply those services efficiently. But more importantly, they are the policy levers that allow subnational governments to promote growth, employment and a better economic performance in their region. In general, this will not be competitive with neighbouring regions since a better performance in one place will spillover positively to help the regions next door, just as it does between neighbouring countries. It is not a zero-sum game therefore. The union would benefit as a whole from these reallocations if they can be used to enhance regional growth and employment. By contrast, the “framework policies” that affect monetary conditions, price stability, financial stability, taxes/spending for revenue insurance/risk sharing, competition and regulation policy, mechanisms for internal/external coordination, and commercial policy (tariffs, trade barriers, exchange rates) are better left with the central government.8 This allocation of policy instruments is made according to comparative advantage for achieving the objectives of subnational and central governments respectively. 3.2 Size of Government The type of allocation scheme above emphasises the importance of creating “own sources” of finance in a devolved system. A regime that relies on grants or tax assignments, by contrast, provides a fatal incentive to expand public spending programmes beyond their efficient level by pressing the centre to shift more in their direction, or by asking the centre to expand the common debt issue in order to allow that to happen. The prospect of an easy bail-out, or easy guarantees for local debt issues, would have the same effect. Creating 8 There is a case for allocating regulation in a single market (such as competition policy, or for banking and financial services) to the federal level, but the regulation of a natural monopoly to subnational governments.

The Practicalities of Economic Federalism

69

revenue autonomy overcomes that problem by passing responsibility and accountability to the regions. There may be a concern that these arrangements could lead to an expansion of government, with overlapping or conflicting functions at different levels. This will be minimised if policies and responsibilities are chosen according to comparative advantage. In that way, the distortions and inefficiencies that might otherwise arise from excess fiscal competition will be kept below the inefficiencies which appear with central financing. In practice, it appears that this tendency for larger government is weak and depends on the form that devolution takes, not its existence.9 The key point is where decentralisation is built around taxes (tax devolution), it is typically associated with smaller government. But where it is financed by transfers or grants (spending devolution) there is a tendency to larger and less efficient government because there is continual political pressure to expand transfers and grants. 3.3 Devolving Fiscal Responsibilities It is now possible to set out a fiscal framework that contains tax autonomy, but which retains institutional links to central government to maintain coordination, to allow risk sharing, and to align subnational fiscal policies with the framework policies of central government. By maximising own source financing, subnational governments increase accountability and realise the efficiency gains. Institutional measures to support the management of risk, risk sharing, monitoring and fiscal oversight are then needed to bring credibility and stability to the system. These suggestions are in line with the Steel Commission’s recommendations.10 To see how that might work, we start from the example of the Basque Country and Navarre which operate very successfully in Spain. These areas pay the Spanish government about 9 per cent of their revenues as “rent” for common services: defence, security, diplomatic services, central administration, contributions to Spain’s eu dues and debt service. In 2008, a comparable sum for Scotland would mean a contribution of £6bn to the uk government. To this we might add another 1 per cent of gdp to allow the uk government to make its own redistribution and stabilisation payments for solidarity purposes.

9

10

See W.E. Oates, “Searching for Leviathan: An Empirical Study”, American Economic Review, 75 (1985) 748–757; W.E. Oates, “Searching for Leviathan: Reply”, American Economic Review, 79 (1989) 578–583; J. Rodden, “Fiscal Federalism and the Growth of Government”, International Organisation, 57 (2003) 695–729. The Steel Commission, “Moving to Federalism”, supra.

70

Hallett

Power over the remaining taxes and expenditures could then be allocated to the subnational government: that is those noted in Section 3.1, but also an assigned share of vat (the legislative authority over vat may not be devolved by eu law); plus other sales taxes; capital gains and inheritance taxes; fuel tax; social security, payroll taxes/nonwage costs (to give Scotland control over her own competitiveness); “cap and trade” (including auctions under the eu Emissions Trading System (eu-ets)); environmental taxes; financial market levies; council tax; tax concessions; operating surpluses of statutory bodies; and the smaller taxes (stamp duty; alcohol, tobacco and vehicle excise duties; passenger duties). New taxes could include landing fees for natural resources, licence fees for electricity generation, green taxes or licensing, or a land tax. As a practical matter, since we would be moving into a world of variable revenues, priority should be given to allowing borrowing powers; and then to devolving legislative power over the taxes that affect the rate of return on labour—specifically national insurance contributions and payroll taxes paid by employers, also business taxes and corporation tax. Since the administration of some of these taxes may be complicated (how do you determine how much of an integrated corporation’s profit actually arises in each differently located plants?), there are various simplified ways to devolve these taxes— such as r&d credits in a unified system; or corporation tax rate reductions in proportion to, and only when, the regional gdp falls below the rest of the national average; or by apportioning corporation tax liabilities in proportion to employment by plant and location. It is further proposed to take pensions out of the social security system. This would require social security contributions to be made into a separate tax, as it is in the eu, instead of funding social security from general taxation and national insurance contributions. This separation is needed in order to define the rents to central government and the contributions to grants. Pensions can then continue to be paid out of a common federal national insurance fund. Finally, the most important feature of fiscal autonomy is that it transfers the means to control the tax base and exemptions, as well as tax rates and tax bands, as a way to expand the revenues and scope of the devolved taxes, to the subnational government. This is founded on the simple observation that the elasticity of tax receipts is always larger to plausible variations in the aggregate level of taxable income, than it is to plausible variations in the average tax rate. For example, in a linear system, the actual elasticities are equal (they are both unity); but a 4 per cent annual rise in nominal taxable incomes would require annual increases of 1 per cent point in average tax rates to match the contributions to tax revenues that could be made by expansions of the tax base (when the average tax rate is 25 per cent). Tax increases on that scale are

The Practicalities of Economic Federalism

71

neither plausible, nor likely to be politically acceptable. In a progressive tax system, the advantage is not so marked: taking a quadratic approximation, the elasticities are 1+2t for changing tax rates and 1+t for changes in the tax base. That might cause the fiscal strategy to be modified at high tax rates (large values of t). Thus, in a progressive system, it will be more important to gain control of the base and tax bands, than it is to get control of tax rates. Note also that automatic transfers (that is, additional risk sharing elements) are built in to this arrangement to provide an automatically stabilising force for the economy via the central budget and/or federal solidarity funds. This will help reduce the subnational government’s immediate need to borrow for spending purposes.11 4

Institutional Arrangements

Some new institutional arrangements are needed if the region is to realise the gains of fiscal autonomy. Devolution reduces the inefficiencies that stem from a centralised one-size-fits-all set of policies, and so improves economic performance. But it may create conflicts in the form of free riding; or spillovers that damage performance elsewhere if not restrained by government. Consequently devolution is at its most effective, and the gains from fiscal autonomy largest, when there is diversity of structure, circumstances, preferences, and when the spillover effects of local action are small. Since central government has a comparative advantage in imposing coordination, discipline, and setting the general thrust of policy, the best way to realise those gains is to create a decentralised scheme in which there are small grants to, or rents from, the regions to the centre; with institutions interposed between the two to secure both coordination and the gains of accountability. We suggest a Grants Commission. This is to reassure the federal government that each region is operating in a macroeconomic framework which is consistent with the rest of the union, and which prevents a region doing anything that might otherwise distort, destabilise or damage the federal economy or its regions. This Grants Commission could have three functions. First, partnership 11

If the horizontal transfers between regions, and vertical ones between regions and centre above, are subject to choice, then we reach full fiscal autonomy encompassing both the advantages and risk sharing capabilities as analysed by C. Tiebout, “A Pure Theory of Local Expenditures”, Journal of Political Economy, 64 (1956), 416–424, and A. Breton and A. Scott, Economic Constitution of Federal States (Toronto: University of Toronto Press, 1978), respectively.

72

Hallett

and coordination: to ensure the federation continues to operate as a social and economic union; and that the underlying subnational/national policies are consistent and mutually supportive. Second, economic and political: that there is a mechanism that defines effective roles for both the subnational and federal governments; that there is an explicit redistribution mechanism between regions if need be, and also a central budget to provide automatic risk sharing transfers that are short term and reversible. Third, a federal monetary fund to distribute any payments, loans, grants or bailout funds to/from the centre for solidarity or stability, or as described in the debt management section below. The Grants Commission would need an Economic Policy Forum, with representatives from federal government and each devolved government, to reach agreed decisions on matters of joint interest and resolve conflicts in the overall macroeconomic framework. The Forum would have authority to recommend changes to the economic policies of any of the constituent regions. Such recommendations would be advisory by majority, or binding with unanimity, and would include discussion of potential retaliations in cases of severe disagreement. 5

Fiscal Discipline and Debt Control

Decentralising fiscal policy may lead to fears of weaker budgetary control. Although there is no reason why subnational governments should be worse in this regard than central government, there is always a temptation to overexpand and export the burden of financing, tax raising, and paying-off debt to others. Borrowers may calculate (perhaps correctly) that central government would then prefer to bail them out rather than risk the financial disruption that would follow should they default. Thus a real or perceived guarantee of a bail out creates moral hazard among both borrowers and lenders, increasing the risk of default. Four mechanisms can be used to contain such behaviour: • Increasing the accountability of the subnational governments through greater autonomy; • Risk sharing transfers, such as in fiscal federalism.12 This will depend on the existence of a central budget and automatic transfers to/from that budget; • Enforceable limits on the size deficit or debt; and • An independent fiscal policy commission charged with oversight and monitoring of the government’s fiscal plans. 12

See fn 2, supra.

The Practicalities of Economic Federalism

73

The need for greater accountability implies fiscal responsibility is the appropriate regime. It also depends on the policymakers’ concern for reputation. A funding collapse would suggest that the subnational government had failed, and leave them accountable to the censure of the voting public—the more so the more autonomy has been granted. Discipline therefore calls for more autonomy rather than less, backed by some visible restraint/punishment mechanism. Risk sharing however depends on the existence of a central budget and automatic transfers to/from that budget. Risk sharing also flows from the cross-border ownership of stocks, bonds or other income sources; and from cross-border lending and credit. So there will always be some risk sharing where there are integrated markets for capital, investment and short term financing. However the presence of central government raises a new problem: moral hazard, the perception that excessive deficits will be bailed out or otherwise “insured” by loans supplied by the central government. Moral hazard blunts the incentive to maintain fiscal discipline; to shrink deficits in good times or to prepare for bad times. For that reason, it is better to have regional stabilisation, rather than central insurance and soft budget constraints. In other words, it is better to have more devolution rather than less; and it is better to have fiscal federalism13 with reversible transfers than centrally determined lending. The third mechanism is to impose enforceable limits on the size of fiscal deficits and debt. In the Eurozone, the problem with such limits has been twofold. First they have proved difficult to enforce. Second, the ability to monitor deficits in real time to detect significant deteriorations is very limited. Early releases of deficit figures, and the data necessary to strip them of their cyclical components, are so imprecise that the ability to detect violations reliably is only achieved after four years; too late to take any corrective action or to induce such actions through the threat of fines. It is better to focus on limits to debt. A debt target is useful for many other reasons. First, debt is what has to be financed, and what causes default risk. Second, the debt burden is more clearly defined: what has and needs to be borrowed is known to the markets since they hold the paper. Third, debt is a moving total of past deficits and hence represents exactly the structural position we wish to monitor. Fourth, debt is a stock and not a flow. That means it is persistent, which will make policymakers forward looking in their plans and, by extension, make their plans more credible—or at least more easily tested for credibility. Moreover, persistence gives them an incentive to restrict debt in order to preserve freedom from financing constraints in the future. 13

See fn. 2, supra.

74

Hallett

However, an effective monitoring agency is still needed to overcome two defects in the existing monitoring schemes; first that they are partial, and second they are purely backward looking. They do not imply any pressure to modify fiscal plans in the light of future problems. To get round that, an independent Fiscal Policy Commission in each region could be proposed. Such a Commission would have the responsibility to review the fiscal outlook, the revenues likely to be available, the current structural position and likely consequences of current spending policies for the sustainability of public finances, including the impact of changing demography on pension and health costs. 5.1 An Excessive Debt Protocol To ensure that effective debt limits will be enforced the following mechanism is proposed: • The federal and subnational governments jointly operate a central mone­ tary fund as above; • The debt targeting system should be set up as a debt target value and an upper boundary; • The space between the target and upper boundary should be divided into three ranges. If the debt target was set at 45 per cent of gdp, and the ceiling at 60 per cent,14 the excessive debt protocol ranges would be from 45 per cent to 50 per cent; from 50 per cent to 55 per cent; and from 55 per cent to 60 per cent. The first range would be the range of normal fluctuation. If the debt ratio entered the second range, the subnational fiscal authority would be placed on the watch list and subject to comment and advice from the Fiscal Policy Commission. Any support from the Grants Commission, or advice from the Fiscal Policy Commission would now become conditional on improvements being made. If the level of debt entered the third debt range, this would trigger public warnings and mandatory policy changes. Finally, if the debt ratio rose above 60 per cent, all guarantees would be lifted. The subnational government’s fiscal budget would be placed in administration, with the subnational Fiscal Policy Commission running government spending and taxation until the 55 per cent limit was regained. 14 The eu Council of Ministers adopted the idea of formal debt limits in its declaration of 26 June 2012. Checherita-Westphal et al find the optimal level for the Eurozone is about 50% of gdp. So a range of 45% to 60% seems realistic, see C. Checherita-Westphal et al, “Fiscal Sustainability Using Growth-maximising Debt Targets”, Applied Economics, 46 (2013) 638–647.

The Practicalities of Economic Federalism

6

75

Advantages of Fiscal Autonomy

The proposals here are predicated on the value, to the federal government, of a central Grants Commission; of saving money by abandoning block grant type funding; of savings achieved by reallocating spending decisions; of being able to create better economic results via devolved taxation and hence a lower burden on the central budget; of having a coordinating mechanism to make the regional and national policies fit together in a coherent macro-economic strategy. This is evidently a superior constitutional arrangement with gains for both sides. 6.1 The Impact of Fiscal Autonomy on Growth and Productivity How do the gains in efficiency and economic management from decentralising fiscal policy decisions translate into better outcomes? The answer is via higher productivity. This is most easily seen through the distinction between a systematically higher growth rate and higher levels of national income per head. First we need to clarify that there is no robust evidence that greater fiscal devolution is associated with higher rates of gdp growth; and second to affirm that, to the extent that fiscal devolution produces higher levels of gdp per head, it does so through the devolution of tax and revenue raising powers— not through the devolution of spending powers. On the first point: there is no evidence of permanently higher long term growth rates, but fiscal devolution does increase the level of gdp per head.15 Permanent growth rate changes, by contrast, depend on other factors and cannot be the result of an increase in the degree of fiscal devolution alone. Other factors, such as an increase in the growth of innovation and technical progress, or in (the quality of) the labour force, would need to be involved. This distinction is made clear by the following: 15

By 3% on average across the oecd for a doubling of devolved tax powers according to the oecd Decentralisation and Economic Growth papers (Part 1); this is chiefly driven by an increase in productivity (Part 2) and educational achievements (Part 3); see H. Blöchliger, “Decentralisation and Economic Growth—Part 1: How Fiscal Federalism Affects LongTerm Development”, oecd Working Papers on Fiscal Federalism, 14 (2013), http://dx.doi .org/10.1787/5k4559gx1q8r-en (accessed 2 March 2016); H. Blöchliger and B. Égert, “Decentralisation and Economic Growth—Part 2: The Impact on Economic Activity, Productivity and Investment”, oecd Working Papers on Fiscal Federalism, 15 (2013), http://dx.doi .org/10.1787/ 5k4559gp7pzw-en (accessed 2 March 2016); K. Fredriksen, “Decentralisation and Economic Growth—Part 3: Decentralisation, Infrstructure Investment and Educational Performance”, oecd Working Papers on Fiscal Federalism, 16 (2013), http://dx.doi .org/10.1787/5k4559gg7wlw-en (accessed 2 March 2016).

76

Hallett

There are two reasons to be cautious with these results. First, although they are quoted as leading to higher growth rates, growth models in economics usually show that the result of a one-off increase of some growth creating factor is an increase in the level of output, not a permanent increase in growth rates. For a permanent increase in the growth rate, continuing increases in that factor would be needed every year. But the degree of fiscal decentralisation cannot go on increasing without limit; and even if the consequences of an increase in devolution took some time to be realised, they would eventually come to an end.16 Figure 4.1 illustrates this proposition. The growth rate of gdp per head is given by the slopes of the lines AC or AF. If there is to be any uplift in gdp per head, then logically the economy must expand to get there. That implies a short period of faster growth. If tax devolution were introduced, we would go from B to D (a short period of GDP/Head

F

G E D C

A

O

B

t0

Time

Figure 4.1 Income per head vs. growth rates under increased Federalism. 16

A. Hughes Hallett and A.G. Scott, Scotland: A New Fiscal Settlement (Edinburgh: Reform Scotland, 2010), 38.

The Practicalities of Economic Federalism

77

additional growth) and then to E (the old growth rate returns). We would be richer at every point than we would have been otherwise (i.e. along DE instead of BC), but we don’t grow faster than before (as on BG) unless we also adopt the additional productivity enhancing measures that the higher revenues just created would make possible. This is why we have devoted time to discussing what additional measures might be necessary and institutional arrangements needed to support them. 6.2 Why Should These Performance Improvements be Expected? A survey of the available academic research is given in a paper by Feld and Schnellenbach.17 This survey provides an argument from first principles, based on work by Brueckner,18 that greater fiscal autonomy will be unambiguously associated with higher output levels and higher productivity or steady state growth rates if taxes move to support the chosen levels of public spending; that is, if tax devolution is included along with spending devolution. This result continues to hold, and robustly so, if tax competition is allowed between regions. But it does not follow if the tax regime involves a shared tax base, as proposed in the Scotland Acts. There are also arguments that fiscal autonomy will deliver better economic performance when ignorance of regional conditions, or pork barrel spending, or central coercive power pose serious problems. Specific examples of what can be expected from greater autonomy in terms of improved performance are given in Section 7 below. These arguments therefore imply that greater fiscal autonomy can be expected to deliver a better economic performance, at least in terms of gdp per head. Nevertheless there is no guarantee if policymakers turn out to be incompetent or the economic institutions weak, or if tax parameters on a common tax base are shared. But that is true for any economy. And it is especially true in partial autonomy schemes since they, unlike full fiscal autonomy or general decentralisation, depend on shared tax bases. This again makes a clear case for why it is better to go for extended devolution rather than a partial scheme. 6.3 Spending Devolution vs. Revenue Devolution The second issue is what kind of fiscal instruments should be devolved? There is no evidence that, if fiscal devolution raises gdp per head, it does so through 17

18

L.P. Feld and J. Schnellenbach, “Fiscal Federalism and Long-Run Macroeconomic Performance”, International Studies Program, Working Paper 10-09, Andrew Young School of Policy Studies (2010). J.K. Brueckner, “Fiscal Federalism and Economic Growth”, Journal of Public Economics, 90 (2006) 2107–2120.

78

Hallett

the devolution of spending powers. If anything, the opposite is true. Existing evidence on that point is emphatic: expenditure devolution depresses gdp levels, while devolving revenue powers increases them: Revenue decentralisation has the expected positive effect on productivity, and is consistently highly significant. Expenditure decentralisation, however, has a robust and highly significant negative effect on productivity [...]19 It should also obvious that adding revenues to the devolved expenditure instruments will, other things equal, lead to a better economic performance because doing so adds to the number of choices the government can make (for example, by choosing the size of the budget as well as what that budget is spent upon). It would therefore be best to devolve both spending and tax raising powers. But if it is required to choose between the two, then, in the light of the evidence above, it will be better to devolve tax raising powers before spending powers. 7

Empirical Experience

7.1 Evidence on General Performance Indicators We have cited only generalised survey evidence so far. But the empirical evidence in favour of our position goes much further. Many other studies report results in line with ours. For example, there are several studies of oecd countries (e.g., Germany, Switzerland) that show that improvements in gdp per head come with tax and spending devolution—for example, two by Thiessen,20 a further six cited in Hallwood and MacDonald21 and a series of studies by the oecd in which a doubling of the degree of fiscal decentralisation in the average oecd economy would be associated with an increase of gdp per head of 3 per cent, and a 1 per cent increase in the level of productivity.22 Once again, 19

20 21

22

L.P. Feld, “Submission to the Scotland Bill Committee of the Scottish Parliament” (2011), http://archive.scottish.parliament.uk/s3/committees/scotBill/documents/49.SB ProfessorLarsPFeld.pdf (accessed 2 March 2016). U. Thiessen, “Fiscal Decentralisation and Economic Growth in High Income oecd Countries”, Fiscal Studies, 24 (2003) 237–274. P. Hallwood and R. MacDonald, The Political Economy of Financing Scottish Government: Considering a New Constitutional Settlement for Scotland (Cheltenham: Edward Elgar, 2009). oecd 2013, Parts 1 and 2, supra.

The Practicalities of Economic Federalism

79

this evidence shows that the link is between fiscal devolution and higher gdp levels, not systematically higher growth rates, although there have to be higher growth rates in the short term to reach higher gdp levels. Second, the outcomes for the autonomous communities of common regime in Spain, vs. the Basque Country and Navarra (the charter regime), shows the same result. In Navarra, gdp/head grew from 25.9 per cent above the nonautonomous average to 29.3 per cent above between 1995 and 2009. For the Basque Country, gdp per head rose from 19.3 per cent to 34.2 per cent above that average. This is a key “controlled” experiment because no other institutional changes were made in that period, apart from introducing full fiscal autonomy in the Basque Country and Navarra in 1995. The point is not that these two regions were richer than the rest of Spain. They were. But that they increased the degree by which they were richer after tax devolution was brought in. Note also that these gains represent increases in gdp per head of between 0.25 per cent and 1.06 per cent per year over the non-autonomous average in that period. Similarly unemployment rose in all of Spain through the recession to 2012, but the rates in the Basque country (10.9 per cent) and Navarra (11.6 per cent) remained at half the non-autonomous Spanish rate (20.3 per cent). Conversely, non-autonomous Catalonia suffered, with unemployment rising to 18 per cent in 2010.23 It appears that fiscal autonomy, sensibly managed, can help the economy weather a storm better than a centralised system. As a result the Basque credit rating remained AAA when that for Spain fell to AA. A study by Djankov et al. extends this evidence.24 For oecd countries, a 10 per cent reduction in corporation tax or factors that reduce corporate tax liability (such as r&d credits) typically increase the investment rate by 2 per cent; growth by 1–2 per cent; the number of entrepreneurs from 3 to 5 per 100 population; and the number of company registrations by 20 per cent. Finally, a paper by Arnold et al., based on a study of 21 oecd countries over 34 years, produces a range of new results on the effect of changes in tax structures on investment and productivity.25 Their most striking result, from a “growth ranking” of different tax policies, is that restructuring taxes to change the incidence of income, consumption and property taxes can “increase gdp 23 24 25

Spanish National Institute of Statistics (January 2011); see, territorial statistics section in www.ine.es. S. Djankov et al, “The Effect of Corporate Taxes on Investment and Entrepreneurship”, American Economic Journal: Macro Economics, 2 (2010) 31–64. J.M. Arnold et al, “Tax Policy for Economic Recovery and Growth”, Economic Journal, 120 (2011) 59–80.

80

Hallett

per head by 0.25 per cent to 1 per cent in the long run”. Although this is not driven by the share of decentralised revenues, it is important for three reasons: (a) It shows that the ability to vary taxes in revenue neutral, inequality minimising ways can create higher productivity. This is an option under fiscal autonomy; but not possible under partial devolution schemes. The explanation is that, given tax devolution, fiscal policies can be tailored to increase performance in any specific circumstances. (b) Although only three taxes are involved here, the improvements in gdp per head are only a little short of those quoted for Spain. That confirms these estimates are broadly reasonable. (c) These improvements come from changing the tax structure, an option which is possible under revenue devolution but not under spending devolution. Indicators of Performance, Fiscal Rules and Fiscal Autonomy in Europe A useful study by Foremny provides specific evidence on the incidence and impact of greater fiscal autonomy over the 15 original members of European Union in the interval 1995 to 2008.26 Using well defined oecd indicators of tax devolution shows that the degree of fiscal autonomy varies significantly over countries, but not much over time (except in Spain and Italy). Consolidated national accounts for the uk, Greece, Ireland, Austria and the Netherlands­ show the least decentralisation, with 10 per cent of taxes or less decentralised; Sweden (60 per cent), Denmark (50 per cent), and Finland, Spain and some Italian regions with about 40 per cent devolved on average; and France, Germany, Belgium in between.27 But with this comes a corresponding shift in the extent to which fiscal policy is restrained and borrowing regulated. Most countries experienced an increase in the application of fiscal rules in 2000–2002 (in many cases from no restraint, to half the policies or more becoming subject to restraint); with Germany, France, Belgium, Finland, Italy, Portugal, Spain having at least half their subnational fiscal policies subject to rules. Only the uk, Netherlands and Greece had very few rules (if any). The result is that the field divides in half: Germany, Austria, Ireland, Netherlands, uk, Belgium, Denmark and Spain have more spending autonomy than tax autonomy; while Greece, Portugal, Italy, Finland,

7.2

26 27

D. Foremny, “Sub-national Deficits in European Countries: The Impact of Fiscal Rules and Tax Autonomy” European Journal of Political Economy, 34 (2014) 86–110. Ibid.

The Practicalities of Economic Federalism

81

France and Sweden have more tax autonomy than spending autonomy.28 The spread along the tax autonomy axis is larger than the spread along the spending autonomy axis—the Nordics apart who have high levels of autonomy in both directions. This is in line with the high performance economies in that period and consistent with the predictions of Section 6—at least for growth and employment. The question is: did these arrangements lead to more fiscal expansion and excess borrowing in the decentralised, more successful tax autonomous economies? 7.3 Fiscal Restraint: Debt and Deficit Ratios under Fiscal Autonomy Pooled regressions show that fiscal sustainability and public borrowing was in fact better in the tax autonomy economies. That is, local tax autonomy lowered fiscal deficits, more so in federal states than in unitary states. These are strong and robust results; stronger where labour market participation is high, but weaker if the share of sub-national spending in general government spending (as opposed to infrastructure spending) is high. Fiscal rules play no role in these results, so tax autonomy is the defining factor here. To be sure of these results, one needs to examine not only the effects on deficits (or deficits per head), but also the impact on growth—and thus on the deficit to gdp ratio. Here increasing use of fiscal rules and of tax autonomy both reduce deficit ratios: most strongly in Italy and the Nordics with fiscal rules, and in Spain, Belgium and Germany for fiscal autonomy. These results are robust, and remain inside their 95 per cent confidence intervals as the degree of autonomy/strength of rules varies—most strongly if tax autonomy is combined with fiscal rules. 8

Conclusions

The implication of the arguments for fiscal autonomy in this review is that economic and fiscal decentralisation is advantageous for the following reasons: (a) The efficiency gains from tax or spending devolution are typically equivalent to an increase in gdp per head of around 1 per cent. But it is necessary to have revenue devolution to get these gains. Spending decentralisation on its own does not guarantee any such gains because of responsibility conflicts, and because there is no accountability from having had to generate the revenues spent. 28

Ibid.

82

Hallett

(b) An additional increase in gdp per head implies some short run growth effects. But there is no implication of a permanent increase in the growth rate, unless other factors, such as an increase in the growth rate of productivity, are present at the same time. (c) Any decentralisation scheme designed to secure these efficiency gains must be supported by suitable institutional arrangements. Those arrangements need to cover fiscal sustainability, risk sharing, inter-regional consistency, simplicity of revenue collection. Examples of how this could be done have been given. The next step is to build up a consistent and effective set of institutions to ensure that these properties are realised. (d) It is easy to build in simplicity and risk sharing if vertical remittances are made to the federal government, and not the other way around, because this ensures a common budget (the key element for stabilisation) but leaves responsibility/accountability for creating a better economic performance at the subnational level where it can be exercised most effectively. This suggests a division of responsibility by comparative advantage: federal authorities look after stability (variances); subnational authorities after performance (mean values). (e) The empirical evidence from Europe and other oecd economies shows that these kinds of gains can and have been realised in a number of advanced economies. The propositions we have put forward are ready for practical implementation.

chapter 5

The Principles of Separation and Correspondence, the Comparative Method, and the Problem of Semantic Change Matteo Nicolini What’s in a name? That which we call a rose By any other name would smell as sweet.1

…

Economics is a much more important factor than politics in motivating language change among the general population […]2

∵ 1

Naming the Variety: Financial Relations, Denominational Issues, and Comparative Studies

This essay analyses the principles of separation and correspondence, i.e., the main principles upon which financial arrangements are built and fiscal relations are shaped in different constitutional contexts and designs. Furthermore, it ascertains how the principles in question work under diverse federal arrangements and therefore govern intergovernmental financial relations in both federal and regional states.3 1 W. Shakespeare, Romeo and Juliet (ii, ii, 1–2). 2 D. Crystal, The Stories of English (London: Allen Lane, 2004), 243. 3 See Art. 104a(1) of the German Basic Law (Grundgesetz—hereinafter gg); para. 2 of the Financial Constitution Act of 1948 of Austria (Finanz-Verfassungsgesetz—hereinafter f-vg); Art. 2 of the Swiss Constitution; s. 92(2) of the Constitution Act 1867 of Canada; schedule vii to the Constitution of India. The rule also applies to regional constitutions, such as those of Spain and Italy: see Arts. 156(1) and 119(3), respectively. In the United States, the principles are not expressly mentioned in the Constitution. Despite this, they have been applied to the distribution of finance since the inception of the Federation.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_007

84

Nicolini

Before examining both principles, however, I will share some reflections on the multifarious institutes and mechanisms that scholars usually include under the umbrella of “intergovernmental financial relations”. In-depth analyses have already been dedicated to the topic, as well as to their applicability to different types of federalism4—and yet there is room left for comparative surveys. The principles of separation and correspondence must be scrutinised by taking into account the broader federal constitutional contexts that they operate within. To this extent, these principles link the two constitutive parts of financial relations (i.e., the revenue and expenditure sides) to what, paraphrasing Friedrich, we may call “federalising and regionalising processes”.5 Indeed, revenue and spending powers—as well as solidarity mechanisms—allow subnational units and local government to finance and carry out their functions. The different tiers of governments are thus enabled “in achieving their policy objectives within their constitutionally assigned legislative and executive responsibilities”.6 When considering the correspondence rule, proceeds transferred from the national level to constituent units under equalisation mechanisms count as a part of revenues. Solidarity complements revenue-raising mechanisms and provides subnational units with additional funds that are aimed at increasing their fiscal capacity, i.e., constituent units’ ability to generate revenue.7 To put 4 See, among others, C. Murray and R. Simeon, “South Africa’s Financial Constitution: Towards Better Delivery?” South African Public Law, 15 (2000) 477–504; R.L. Watts, Comparing Federal Systems (3rd edn., Montreal: McGill–Queen’s University Press, 2006), 106; A. Shah, “Introduction: Principles of Fiscal Federalism”, in A. Shah and J. Kincaid (eds.), The Practice of Fiscal Federalism: Comparative Perspectives (Montreal: McGill–Queen’s University Press, 2007) 3–43; P.W. Hogg, Constitutional Law of Canada (Toronto: Carswell, 2011), 6–2ff.; G. Anderson, Fiscal Federalism: A Comparative Introduction (Oxford: Oxford University Press, 2009). 5 See C.J. Friedrich, “New Dimensions of Federalism”, Proceedings of the American Society of International Law at Its Annual Meeting (1921–1969), 57 (1963) 238–240; C.J. Friedrich, Trends of Federalism in Theory and Practice (New York: Praeger, 1968), 24; M. Burgess, In Search of the Federal Spirit: New Comparative Empirical and Theoretical Perspectives (Oxford: Oxford University Press, 2012). 6 Watts, Comparing Federal Systems, supra, at 106. 7 See, among others H.H. Landreth, “The Measurement of Local Fiscal Capacity”, The Journal of Finance, 16 (1961) 105–106; J. Mikesell, “Changing State Fiscal Capacity and Tax Effort in an Era of Devolving Government, 1981–2003”, Publius, 37 (2007) 532–550; T. Besley et al., “Weak States and Steady States: The Dynamics of Fiscal Capacity”, American Economic Journal: Macroeconomics, 5 (2013) 205–235. Concerning the interrelations between equalisation mechanisms, revenues, and fiscal capacity is apparent in Canada, see B. Dahlby and L.S. Wilson, “Fiscal Capacity, Tax Effort, and Optimal Equalization Grants”, The Canadian Journal of Economics, 27 (1994) 657–672.

The Principles of Separation and Correspondence

85

it differently, levying taxes, imposing fees and equalisation allow constituent units to fully finance their constitutional responsibilities. A correspondence is thus established between revenues, expenditures and competences, and this enables constituent units to discharge such responsibilities without any interference from the central level of government, i.e., according to the principle of separation.8 Separation then enhances both fiscal responsibility and accountability of the constituent units, and it preserves their constitutional “sphere of guaranteed autonomy”.9 However, a rapid survey of the different federal (and regional) constitutional designs discloses a great variety of financial arrangements. It could be argued that each federalising process has it own forms of financial relations, which indeed vary from state to state and take different forms in different constitutional contexts. The linguistic factor also affects this huge variety of forms of financial relations: each constitutional design assigns its own label to the variety of financial forms, and a legal approach must take this into account.10 We have to examine the Babel of the naming, and then try to accommodate correspondence, separation, and linguistic variety. This is due to the fact that such variety reflects how constitutions term their “domestic” mechanisms for the allocation of fiscal powers and the distribution of the estimates of revenue to be raised throughout the country. 2

From Language to Law: Correspondence, Separation, and the Variation of Forms

On the one hand, linguistic variation is complemented by a great deal of “financial” forms; on the other hand, it is also extremely differentiated on domestic 8

9 10

On the manifold mechanisms through which federal and regional governments affect the autonomy of constituent units, see Shah, “Introduction: Principles of Fiscal Federalism”, supra, at 21. G. Sawer, Modern Federalism (London: Watts, 1969), 27. See also W.W. Riker, Federalism: Origin, Operation, Significance (Boston: Little, Brown & Co., 1964), 6. On “the rich variety of forms of legal language, with their various rhetorical, ethical, political, and logical implications”; on the functions of legal language, which is “therefore [that] of ordering in the sense of creating order as well as in the sense of giving orders”; on the “interrelationships of national languages with each other and with the particular types of languages of various […] disciplines or activities”; see H.J. Berman, Law and Language: Effective Symbols of Community, edited by J. Witte, Jr. (Cambridge: Cambridge University Press, 2012), 75, 77, and 63 respectively. For a comprehensive approach to global variation in English linguistics, see R. Hickey (ed.), Standards of English. Codified Varieties around the World (Cambridge: Cambridge University Press, 2012).

86

Nicolini

grounds. When it comes to comparing financial mechanisms, the process of naming raises the following questions: does the linguistic variation really reflect a variety of forms? Does this cross-referential constitutional vocabulary entail differentiated concepts of “financial relations”? Or, as I contend, does it merely “confound [the] language [of constitutions], [such] that they may not understand one another’s speech?”11 This is not to deny that variations in naming can really reflect different ways of interpreting financial relations in different constitutional contexts. This assumption is held when federations try to reconcile the allocation of financial powers and the distribution of responsibilities between tiers of government. Although federal and regional states share the rationale of financial arrangements, there is no homogeneity between the ways such arrangements are enhanced. The pendulum swings between two notions of “financial relations”: they can be conceived as a subject matter included in the list of specified heads of legislative powers assigned to the different levels of government, or they are the presupposition for the discharge of (legislative and administrative) responsibilities constitutionally assigned to units in order to meet the needs of their respective communities. The vast majority of constitutions consider financial relations as presuppositions that allow such responsibilities to work. Canada and South Africa are the most relevant exceptions to the rule. In Canada, financial powers are considered a subject matter to be distributed among the different levels of government. Hence, Section 92(2) of the Canadian Constitution Act 1867 expressly limits provincial taxing power to “the raising of a revenue for provincial purposes”.12 This holds true as far as South African intergovernmental financial relations are concerned. First, Sections 214–215 and 228 of the 1996 Constitution refer to several acts to be passed by the national parliament, and provincial taxing and revenue power have been progressively reduced by national legislation.13 Second, subject matters like “public finance”, the “general­ 11 12

13

Ge. 11:7, in Holy Bible: King James Version (Cambridge: Cambridge University Press, 2006), 10. See Hogg, Constitutional Law of Canada, supra, at 6–21. See also Reference Re Legislative Jurisdiction of Parliament of Canada to Enact the Employment and Social Insurance Act (1935, c. 48), [1936] s.c.r. 427, at 434–435, per Duff C.J. Dissenting. The words “for provincial purposes” might be given a restrictive operation: “the power to legislate for taxation […] is concerned with taxation for the purpose of raising money for the exclusive disposition of the local legislature”. Under s. 228(2) of the Constitution, provincial taxing powers are limited by the Provincial Tax Regulation Process Act (No. 53 of 2001), which was passed upon the advice of the Financial and Fiscal Commission. The limitation of taxing powers is compensated by the distribution of an equitable share of the revenues raised nationally and provincially

The Principles of Separation and Correspondence

87

budgetary process”, and an “equitable share of revenue raised during the financial year” fall under the umbrella of Section 214 of the Constitution, and therefore are reserved to the exclusive competence of the national government. Finally, the South African Constitutional Court has interpreted revenueraising powers as a legislative subject matter. In two leading cases,14 it stated that only the national parliament enjoys plenary legislative power within the bounds of the Constitution, whereas: the legislative authority of provinces is circumscribed […] Provinces have no power to legislate on a matter falling outside Schedules 4 and 5 unless it is a matter […] expressly assigned to the province by national legislation […] Financial management of provincial legislatures is a matter that is listed neither in Schedule 4 nor in Schedule 5 to the Constitution. It follows, therefore, that it is a matter that falls within the legislative competence of Parliament.15 It does not follow from this that the allocation of finance, in general, and of revenue-raising powers, in particular, are immune from how constitutions distribute responsibilities. First, financial autonomy entails fiscal responsibility—­ each level of government must rely on appropriate proceeds in order to finance its own powers. Second, taxing powers are the major—but not the sole— sources of proceeds. In order to carry out their functions, units usually combine different revenues: licence fees, direct and indirect taxation, equalisation mechanisms, grants and so on. Third, the different levels of government do not recognise any limitations on their spending powers, which cannot therefore be considered as a subject matter, and this assumption is held both in Canadian and us case law.16 under the annual Division of Revenue Act (s. 214 of the Constitution), whose adoption is conditioned by the Intergovernmental Fiscal Relations Act (no. 97 of 1997). Finally, the budgetary process falls under the umbrella of the Public Finance Management Act (no. 1 of 1999), see s. 215 of the Constitution. See P. de Vos and W. Freedman (eds.), South African Constitutional Law in Context (Oxford: Oxford University Press, 2014), 305ff. 14 Premier: Limpopo Province v. Speaker of the Limpopo Provincial Legislature 2011 (6) s.a. 396 (c.c.) (“Limpopo i”) and Premier: Limpopo Province v. Speaker of the Limpopo Provincial Legislature 2012 (4) s.a. 58 (c.c.) (“Limpopo ii”). 15 Limpopo i, at para. 21. See N. Steytler and R.F. Williams, “Squeezing out Provinces’ Legislative Competence in Premier: Limpopo Province v. Speaker: Limpopo Provincial Legislature & Others i and ii”, South African Law Journal, 129 (2012) 621–637. 16 Smylie v. The Queen, [1900] 27 o.a.r. 172 (onca); Ontario v. Board of Transport Commissioners, [1968] s.c.r. 118; Alberta v. Canada (Transport Commission), [1978] 1 s.c.r. 61. For u.s. case law, see United States v. Butler, 297 u.s. 1 (1936), at 65. See also Steward Machine

88

Nicolini

Fourth, financial relations, on the one hand, and legislative and administrative powers, on the other, are interrelated under the principle of correspondence. On the one hand, legislative powers indicate which responsibilities must be financed through revenue powers; on the other hand, legislation is essential for enacting spending policies and raising funds by taxation. Both appropriation of funds and taxing must necessarily be exercised through legislation: the principle of correspondence thus governs the distribution of both tax bases and legislative powers related thereto, since taxes must be levied or collected under the authority of law.17 This interrelation between legislation and revenue-raising powers is particularly apparent when it comes to the principle of separation. Because of the correspondence between proceeds and legislative powers, each level of government should in principle levy taxes and collect proceeds within the scope of the responsibilities it has to discharge. However, the application of the principle of separation varies according to the different constitutional contexts. Dual federalism still reflects the abovementioned correspondence between taxation powers, the scope of the responsibilities to be discharged and legislative powers. As these powers are allocated at different levels, they are separated as if they were in watertight compartments: the different levels of government can only impose taxes in those fields the constitution assigns to them, and national and subnational taxing powers are limited to the raising of revenues only for their respective purposes.18 This was the case in the usa prior the 16th Amendment (1913), in Canada and in Australia, where financial autonomy and the discharge of responsibilities matched the distribution of legislative powers. This approach is still the operational rule

Co v. Davis, 301 u.s. 548 (1937); Oklahoma v. United States Civil Service Commission, 330 u.s. 127 (1947); Fullilove v. Klutznick, 448 u.s. 448 (1980). See W.F. Fox, “The United States of America”, in Shah and Kincaid, The Practice of Fiscal Federalism, supra, 344–368, at 349ff.; J.V. Corbelli, “Tower of Power: South Dakota v. Dole and the Strength of the Spending Power”, University of Pittsburgh Law Review, 49 (1988) 1097–1126, at 1101ff.; T.H. Verhuizen, “United States v. American Library Association: The Supreme Court Fails to Make the South Dakota v. Dole Standard a Meaningful Limitation on the Congressional Spending Power”, South Dakota Law Review, 52 (2007) 565–604. 17 See, among others, ss 91(3), 92(2) and 92(9) of the Constitution Act 1867; s. 83 of the Commonwealth of Australia Constitution Act 1900; s. 265 of the Indian Constitution; s. 96 of the Constitution of Malaysia. 18 See Pollock v. Farmers’ Loan & Trust Company, 158 u.s. 601 (1895). See L.H. Tribe, American Constitutional Law (New York: Foundation Press, 2000), 842–843; Fox, “The United States of America”, supra, at 349ff.

The Principles of Separation and Correspondence

89

in Switzerland,19 where the federal constitution resorts to a “mixed rule”, where competition, solidarity, and cooperation merge. The 2004 Swiss constitutional reform combined competition and solidarity: the three pillars of this “new” Swiss fiscal dualism rest on a stringent distribution of taxing powers between the federation and the cantons (Article 127ff.), on the application of fiscal responsibility, and correspondence between revenues and expenditures (Article 47(2)), as well as on equalisation mechanisms, which are aimed at correcting the competition among cantons (Article 135).20 Further, how the principles of correspondence and separation work is due to the tremendous changes that federations underwent in the course of the twentieth century. Whereas in some federations these changes were arranged without having recourse to formal constitutional amendments, such as in Canada and Australia,21 in others countries, constitutional amendments altered the forms of correspondence and separation vis-à-vis dual federalist contexts.22 But the major constitutional reflexes of the passing of dual federalism23 were due to the advent of cooperative intergovernmental financial relations. This entailed an alteration of the same meaning of correspondence and separation that we may call “social federalism”: the national government had to ensure minimum standards of income, health, education and welfare to all citizens, and the possibility of enjoying comparable services must not be submitted to excessively different tax rates.24 19

20 21

22 23 24

See Art. 47(2) of the Federal Constitution (amended in 2004 and entered into force in 2008): “the Confederation shall leave the Cantons with sufficient sources of finance and contributes towards ensuring that they have the financial resources required to fulfil their tasks”. See G. Kirchgässner, “Swiss Confederation”, in Shah and Kincaid, The Practice of Fiscal Federalism, supra, 318–343, at 329; G. Kirchgässner, “Finanzföderalismus in der Schweiz”, in Peter Bußjäger (ed.), Perspektiven des Finanzföderalismus in Österreich (Innsbruck: Studienverlag, 2013) 39–70, at 46ff; R. Kägi-Diener, “Art. 47”, in B. Ehrenzeller et al (eds.), Die schweizerische Bundesverfassung (Zürich: Dike-Schulthess, 2008) 875–881. See Kirchgässner, “Finanzföderalismus in der Schweiz”, supra, at 56ff. The centralising effects of World War ii favoured the creation of a new set of intergovernmental financial relations and the abandonment of dual federalism. In Canada and Australia, “it is true that the framers of the Constitution could hardly have foreseen the rise of the welfare state with its enormous growth in [subnational] responsibilities”. See Hogg, Constitutional Law of Canada, supra, 6–18. For Australia, see A. Morris, “Commonwealth of Australia”, in Shah and Kincaid, The Practice of Fiscal Federalism, supra, 43–72, at 71 note 2. For example, the 16th Amendment to the u.s. Constitution and the 1955 constitutional amendments to the German Basic Law. See below, Section 4. See E.S. Corwin, “The Passing of Dual Federalism”, Virginia Law Review, 36 (1950) 1–24. See Watts, Comparing Federal Systems, supra, at 108; Morris, “Commonwealth of Australia”, supra, at 55.

90

Nicolini

As a consequence, centralisation narrowed units’ taxes in those fields the constitution assigned to them, and triggered a continuation of highly centralised fiscal arrangements. Since shared costs assure a high minimum level of some important social services, units’ taxing powers must be inevitably complemented by equalisation mechanisms (grants, tax rental agreements, vertical and horizontal equalisation). This means that all “[r]evenues raised from the above-mentioned sources shall enable [constituent units] to fully finance the public functions attributed to them”, and shall count as proceeds in accordance with the principles of correspondence and separation.25 Variation is even greater in Austria, where the allocation of revenue-related legislative powers matches neither the distribution of responsibilities to be financed nor the revenue powers. Whereas the distribution of legislative powers is enshrined in the federal Constitution (Bundes-Verfassungsgesetz—hereinafter b-vg), revenue powers are set forth in the f-vg—to which Article 13 b-vg expressly refers—according to hugely differentiated criteria.26 3

“What’s in a Name?” Comparative Legal Studies and the Unitary Function Underpinning “Financial Linguistic Variation”

Comparative legal studies are usually entrusted with the goal of dealing with the variety of forms and denominations that characterise intergovernmental financial relations. On the one hand, these studies contribute to filling in the gaps between written provisions and the practice of law—and the flaw between black-letter constitutions and the living ones is clear when it comes to financial relations. On the other hand, this method makes it possible to examine a vast array of constitutional regimes and operational rules, and therefore to propose classifications that are the outcome of a cross-national analysis of federal systems. Comparative legal scholars are indeed accustomed to examining a vast array of 25

26

See Art. 119 of the Italian Constitution. The same holds true in Spain. See Art. 156 of the 1978 Constitution. For an examination of the principle of correspondence in regional countries (as well as in times of social federalism), see M. Nicolini, “Principio di connessione e metodo comparato”, in F. Palermo et al (eds.), Federalismo fiscale: una sfida comparata (Padua: cedam, 2011) 97–120; E. Aja, Estado Autonómico y reforma federal (Madrid: Alianza Editorial, 2014), 235ff. See P. Bußjäger, “Reforms on Fiscal Federalism in Austria”, in G. Robbers (ed.), Reforming Federalism—Foreign Experiences for a Reform in Germany (Frankfurt am Main: Peter Lang, 2005) 59–67; M. Stelzer, The Constitution of the Republic of Austria: A Contextual Analysis (Oxford: Hart Publishing, 2011) 153.

The Principles of Separation and Correspondence

91

forms of financial relations, to grouping them on the grounds of their common traits and to devising “prescriptive models” that are “a synthesis of complexity by logical categories” useful for the advancement of comparative legal studies.27 Further, the comparative method has proven to be extremely useful, e.g., it has made it possible to include the multifarious forms of financial relations under three basic categories: “revenue powers”, “equalisation mechanisms”, and “spending powers”.28 The comparative method thus exhibits an impressive ability to deal with legal complexity; furthermore, it proves capable of achieving the goal of accommodating the variety of forms that financial relations take under different constitutional designs. It does not follow from this, however, that this method is aimed at reducing and oversimplifying such variety; by contrast, it is capable of comparing and contrasting different forms of financial relations, detecting analogies and differences between them and devising models through which subnational responsibilities are financed. This means that the comparative method not only presupposes both legal and linguistic variation, but that linguistic variation cannot be considered an obstacle to comparative studies. The method indeed allows scholars to get beneath linguistic labels and grasp the commonalities among them. The variety of languages, then, does not confound the interpreter: paraphrasing Shakespeare, “that which we call distribution of finance by any other name would perform the same function”. What does make this possible is termed the functional approach. This approach goes beyond the financial and fiscal machineries that characterise a specific financial constitutional regime. Instead of focusing on variation, the approach in question is centred on the function that the principles of separation and correspondence tend to attain under different constitutional 27

28

See L. Pegoraro, “The Comparative Method and Constitutional Legal Science: New Trends”, in A. Mordechai Rabello and A. Zanotti (eds.), Developments in European, Italian and Israeli Law (Milan: Giuffrè, 2001) 113–129, at 117 and 126. See also R. David, Les grands systèmes de droit contemporains (Paris: Dalloz, 1974); R. Sacco, “A Dynamic Approach to Comparative Law (Installment i of ii)”, The American Journal of Comparative Law, 39 (1991) 1–34; R. Sacco, “A Dynamic Approach to Comparative Law (Installment ii of ii)” The American Journal of Comparative Law, 39 (1991) 343–402; K. Zweigert and H. Kötz, Einführung in die Rechtsvergleichung auf dem Gebiete des Privatrechts (Tübingen: J.C.B. Mohr, 1996), 31ff.; P.G. Monateri (ed.), Methods of Comparative Law: An Intellectual Overview (Cheltenham: Edward Elgar Publishing, 2012). Scholars consider these three features as the constitutive parts of financial relations. See Watts, Comparing Federal Systems, supra, 95ff.; R.L. Watts and J. Kincaid, “Introduction”, in R.L. Watts and R. Chattopadhyay (eds.), Emerging Issues in Fiscal Federalism (New Delhi: Viva Books, 2008) xi–xvii.

92

Nicolini

frameworks: “Institutions, both legal and non-legal, even doctrinally different ones, are comparable if they are functionally equivalent, if they fulfill similar functions in different legal systems”.29 In this respect, the search for commonalities and analogies may help in the process of highlighting mutual borrowings, the historical evolution of financial relations in specific constitutional contexts, as well as transplants.30 In other words, this analysis is aimed at detecting the dissemination of principles and mechanisms that govern financial relations, as well as their subsequent adaptation. However, the devising of prescriptive models does not support the homogenisation–unification of these diverse solutions adopted under different federal constitutions. I have already contended that comparative legal studies presuppose linguistic and legal variety; furthermore, and looking at how financial relations actually work, the functional approach encourages the practice of the “comparative law as a subversive discipline”.31 The subversive character of comparative studies reveals the fallacy of the presence of “new universals” in what has become a globalised world,32 and therefore shows how legacies and transplants exhibit an elevated degree of resilience in federal contexts that act as recipients of a transplanted solution in the field of financial relations. To sum up, the comparative method thus highlights the unitary function of financial relations: levying taxes, imposing fees and equalisation mechanisms are functionally oriented to provide constituent units with the appropriate proceeds in order to fully discharge their constitutional responsibilities.33 Once more, what unifies the different forms of financial relations and makes them functionally equivalent—what, poetically, makes them “smell as sweet”, notwithstanding the naming—is the principle of correspondence, whose constitutional (linguistic and legal) regimes I will examine in the following sections through the lenses of the comparative legal method. 29

Zweigert and Kötz, Einführung, supra, at 33ff. See also R. Michaels, “The Functional Method of Comparative Law”, in M. Reimann and R. Zimmermann (eds.), The Oxford Handbook of Comparative Law (Oxford: Oxford University Press, 2006) 339–382, at 340. 30 A. Watson, Legal Transplants: An Approach to Comparative Law (Athens: The University of Georgia Press, 1993); D. Berkowitz et al., “The Transplant Effect”, The American Journal of Comparative Law, 51 (2003) 163–203. 31 G. Fetcher, “Comparative Law as a Subversive Discipline”, American Journal of Comparative Law, 46 (1988) 683–700; H. Muir Watt, “Further Terrains for Subversive Comparison: The Field of Global Governance and the Public/Private Divide”, in P.G. Monateri (ed.), Methods of Comparative Law, supra, 270–288, at 270. 32 Muir Watt, “Further Terrains for Subversive Comparison”, supra, at 272. 33 Watts, Comparing Federal Systems, supra, at 106; Hogg, Constitutional Law of Canada, supra, at 6–21ff.

The Principles of Separation and Correspondence

4

93

The Correspondence Rule: Its Constitutional Regime, Its Operational Rule and the Principle of Fiscal Equivalence

The comparative method is also aimed at outlining the constitutional regime of financial relations, which is indeed inherent to every federalising process. As Alexander Hamilton stated: A complete power, therefore, to procure a regular and adequate supply of revenue, as far as the resources of the community will permit, may be regarded as an indispensable ingredient in every constitution.34 This holds true as far as intergovernmental financial relations are concerned. Questions arise, however, when constituent units do not possess the amount of resources that is necessary to fully finance (and therefore discharge) their constitutionally assigned responsibilities. This is due to the centralising effects that the advent of social federalism has triggered, as well as to the presence of de facto imbalances—both “vertical” and “horizontal”35—which call for the correction of “the fiscal inefficiencies and regional inequities arising from the differential fiscal capacities of various jurisdictions” by providing them with additional or further sources.36 The outcome is twofold: on the one hand, such changes cause a mismatch between subnational revenue-raising capacity and expenditure obligations;37 on the other hand, they can alter the way principles of correspondence and separation function, and therefore “break the nexus between taxing and 34

35

36 37

A. Hamilton, “No. 30 (Concerning Taxation)”, in A. Hamilton et al., The Federalist: The Gideon Edition, edited by G.W. Carey and J. McClellan (Indianapolis: Liberty Fund, 2001) 145–149, at 146. Vertical imbalances (or the “vertical fiscal gap”) highlight that “the revenues of the constituent units are inadequate to fulfill their constitutionally assigned expenditures responsibilities”. See Watts, Comparing Federal Systems, supra, at 103. Horizontal imbalances concern the relations between the different units: the variation in their revenueraising capacities makes them “not able to provide their citizens with services at the same level on the basis of comparable tax levels”. See Hogg, Constitutional Law of Canada, supra, at 6–2. As a consequence, there are differential net fiscal benefits across the country, which can determine “unequal treatment of citizens with identical private incomes depending on their place of residence”. See Shah, “Introduction: Principles of Fiscal Federalism”, supra, at 10. See Hogg, Constitutional Law of Canada, supra, at 6–8; Watts, Comparing Federal Systems, supra, at 98; Shah, “Introduction: Principles of Fiscal Federalism”, supra, at 13. As in Australia. See C. Walsh, “Federal Reform and the Politics of Vertical Fiscal Imbalance”, Australian Journal of Political Science, 27 (1992) 19–37.

94

Nicolini

spending, with implications for accountability with both spheres of government”, i.e., national and subnational.38 This is the case of aggregative federations, such as the usa, Canada, and Australia, which have rarely modified their constitutional texts in order to accommodate the law in the books to the law in action concerning financial relations. As a consequence, the design of the Constitution [according to which] each sphere of government [would] raise taxes for their own purposes and [would] be democratically accountable for both taxing and spending39 was progressively modified without altering the constitutional framework. Fiscal dualism was thus superseded by innovative, fluid intergovernmental financial relations but that are not constitutionally enshrined, which both confer a new role on the federation. The federation has “revenue resources that far outstrip its expenditure responsibilities” because it levies taxes throughout its federal and state territories and transfers part of the total amount to the subnational (and local) levels of government.40 This means that a rigorous approach to financial relations is not confined to a mere sketch of the relevant constitutional provisions; by contrast, it tries to accommodate the disparities between the black-letter constitution and the operational rules governing their respective constitutional regimes. Pointing out discrepancies between the law in the books and the law in action of financial relations41 expands the variety of forms. The practice of comparative law discloses that such flaws between the constitutional regime and the operational rule of financial relations are also inherent to those countries that have modified the constitutional provisions by which the principle of correspondence is governed. As already mentioned, comparative law as a subversive discipline reveals that correspondence can be considered neither a “new universal” nor a “prescriptive 38

39 40 41

C. Saunders, “The Interdependence of Federalism and Democracy in Australia”, in F. Palermo and E. Alber (eds.) Federalism as Decision-Making: Changes in Structures, Procedures and Policies (Leiden: Brill-Nijhoff, 2015) 20–39, at 33. Saunders, “The Interdependence of Federalism and Democracy”, supra, at 32. Ibid., at 33. These discrepancies are caused by “the tension between black-letter law and rules ‘in action’”. See J.-L. Halpérin, “Law in Books and Law in Action: the Problem of Legal Change”, Maine Law Review, 64 (2011) 46–76, at 47.

The Principles of Separation and Correspondence

95

grammar” in the field of financial relations; by contrast, it has multifarious functional equivalents. It is true that in the United States of America, Canada, Australia and Switzerland, the correspondence rule still entails that federal and units’ expenditures match revenues, and their respective constitutions require correspondence between revenues, expenditures and responsibilities. In other countries, however, this principle is no longer considered the backbone of the system, and its application is confined to the mere correspondence between expenditures and responsibilities, such as in Germany.42 This is the constitutional reflex of the passing of dual federalism and of the advent of cooperative, intergovernmental financial relations. The original text of the Basic Law did not set down any exception to the correspondence rule, which was contemplated by former Article 106 gg—the framers of the Basic Law adopted fiscal dualism. This was then overridden by the 1955 constitutional amendments, which modified Articles 106 and 107 gg and introduced revenue-sharing mechanisms for income tax and corporate tax. The trend towards a more cooperative set of financial relations culminated in the 1969 constitutional reform, which set down manifold exceptions to the principle of correspondence. In the 1990s and 2000s, the incorporation of Länder carved out from the former Democratic Republic of Germany, as well as the fiscal distress of several Länder led to the 2006 and 2009 constitutional reforms, which deeply altered Germany’s “fiscal constitution”.43 The German Basic Law provides us with the most accurate and precise definition of the principle of correspondence, which Article 104a gg articulates in two distinct principles. The first is the “principle of separation”. On the one hand, the federation and Länder separately finance the expenditures resulting from the discharge of their respective responsibilities (Article 104a(1) gg). This means that Länder must bear the costs of their own activities. On the other hand, the discharge of the constitutionally assigned responsibilities prescribes that Länder be attributed an appropriate amount of funds.44 42

43

44

See H. von Armin, “Finanzzuständigkeit” in J. Isensee and P. Kirchhof (eds.), Handbuch des Staatsrecht der Bundesrepublik Deutschland, vol. vi, Bundesstaat (3rd edn., Heidelberg: C.F. Müller, 2008) 843–874; U. Häde, Finanzausgleich (Tübingen: Mohr Siebeck, 1996), 48–62; C. Pielke, Das Konnexitätsprinzip in der deutschen Finanzverfassung (Hamburg: Dr. Kovac, 2010). See the Gesetz zur Änderung des Grundgesetzes, passed on 28 August 2006 (BGBl 2006, Teil I, Nr. 41, S.2034), also known as “Föderalismusreform”. See L.P. Feld and J. von Hagen, “Federal Republic of Germany”, in Shah and Kincaid, The Practice of Fiscal Federalism, supra, 126–149, at 127. Among them, see BVerfGE 72, 330.

96

Nicolini

The second principle is the principle of connection, which imposes stringent correspondence between revenues, responsibilities and expenditures related to the same responsibilities.45 An application of this principle can be found in Article 104a(2) gg: when Länder act on behalf of the Bund, the latter will finance the expenditures Länder are called on to implement.46 Article 104a(2) gg refers to the administrative federalism mechanisms: Länder generally execute federal laws in their own right (Article 84 gg); but, as the case may be, the Basic Law provides that administrative responsibilities shall be vested in the Bund, unless it entrusts the execution of federal laws to Länder (Article 85 gg). As a consequence, the federal government may issue general administrative rules and send Land authorities instructions for the accurate implementation of federal laws. Furthermore, the Bund grants Länder governments appropriate funds, with the purpose of bearing the costs deriving from the implementation of federal legislation. The Basic Law permits administrative inter-delegation, which, however, cannot affect constituent units’ financial responsibility and political responsibility, i.e., the principles of separation and connection.47 Not only does the principle of correspondence not match the “prescriptive grammar” stemming from a “universal” correspondence rule, but it does not correspond to the principle “revenue follow functions” either, a principle developed by us economists in the second half of the twentieth century.48 In this regard, the Basic Law establishes so many exceptions to the principle of correspondence,49 so that equivalence between spending powers and revenue powers is not ensured: indeed, both types of powers are allocated at different levels of government.50 This also means that the correspondence does not match the economic principle of fiscal equivalence, according to which the quest for optimality examines the “logically possible relationships between the ‘boundaries’ of a collective good and the boundaries of the government that provides it”.

45

46 47 48

49 50

This expression is a translation of the German Konnexität von Aufgaben und Ausgabenwerantwortung. See H.-U. Erichsen, Die Konnexität von Aufgaben und Finanzierungskompetenz im Bundes-Länder-Verhältnis (Berlin: Bad Hamburg, 1968). See J. Hellermann, “Artikel 104a”, in H. von Mangoldt et al (eds.), Kommentar zum Grundgesetz, Band 3: Artikel 83 bis 146 (München: Verlag Franz Valen, 2010) 1089–1186. See Hellermann “Artikel 104a”, supra, at 1117. See M. Olson Jr, “The Principle of ‘Fiscal Equivalence’: The Division of Responsibilities among Different Levels of Government”, Papers and Proceedings of the Eighty-first Annual Meeting of the American Economic Association, 59 (1969) 479–487. See below, Section 5. See Hellermann “Artikel 104a”, supra, at 1117.

The Principles of Separation and Correspondence

97

Further, it is evident that: [t]here will be no modification of the principle of fiscal equivalence when the government unit with boundaries that match those of the collective good it provides happens to be of just the right size to produce the collective good at the lowest point on the average cost curve. Neither is there any objection to fiscal equivalence when a government whose boundaries are determined by this principle produces its collective good under conditions of decreasing costs.51 Such “equivalence”, however, corresponds neither to the constitutional regime nor to the operational rule of financial relations in Germany. The constitutional reforms that altered the principles of correspondence and separation broke the nexus between taxing, spending and responsibilities, which are allocated according to the discretional public choice of the political branches. Finally, it should be mentioned that the principle of correspondence also affects the financial responsibilities of local governments: it indeed governs financial relations at the level of local government in Germany (see Articles 6 and 83(3) of the Constitution of Bavaria), in Belgium (Articles 41, 162 and 170 of the Constitution), and in South Africa. As the Constitutional Court of South Africa held: The purpose of a municipality’s revenue-raising powers is to finance a municipality’s performance of its constitutional and statutory objects and duties as set out in Sections 152(1) and 153 of the Constitution. These include the provision of services to communities in a sustainable manner, promoting social and economic development and providing for the basic needs of the community. These objects are integral in the task of constructing society in the functional areas of local government.52 5

Exceptions to the Correspondence Rule and the Shift towards Administrative Connection

The flaws between the constitutional regime and the operational rule of the correspondence principle are even more troublesome when we consider that 51 52

Olson Jr, “The Principle of ‘Fiscal Equivalence’”, supra, at 483 and 485, respectively. Liebenberg no v Bergrivier Municipality 2013 (5) sa 246 (cc), at 40. See also, Democratic Alliance v Masondo no 2003 (2) sa 413 (cc), at 17.

98

Nicolini

federal and regional constitutions establish numerous exceptions to the applicability of the same correspondence rule. In Germany, for example, the Basic Law establishes four relevant exceptions. First, the principle of correspondence applies to both legislative responsibilities and administrative inter-delegation: if a federal competence is delegated to the Länder, the federation has to cover the related costs (Auftragverwaltung: Article 104a(2) gg). Second, a federal law assigning money grants (Geldleistungsgesetze) can also force Länder to assume the costs deriving from the implementation of federal legislation (Article 104a(3) gg). Third, federal laws can oblige the Länder to provide grants, benefits or comparable services to third parties only with the approval of the Bundesrat (Article 104a(4) gg). Fourth, conditional federal grants (Finanzhilfen) can be assigned to Länder provided that: (1) they avoid disturbing the overall economic equilibrium; (2) they equalise differing economic capacities within the federal territory; (3) they promote economic growth (Article 104b gg).53 Articles 91a–91e gg establish additional exceptions. The Bund co-finances the responsibilities of Länder in several areas (improvement of regional economic structures, agrarian structure, coastal preservation, research), which are termed “common tasks”.54 The exceptions to the correspondence rule alter the constitutional distribution of powers, but federal laws allowing such derogations require the consent of the Bundesrat. The same assumption holds true as far as Austria is concerned. Article 4 f-vg expressly requires that Austrian Länder must have the appropriate amount of funds in order to finance their tasks. However, Länder do not have any form of fiscal responsibility, and exceptions to the principle of correspondence derive from the criteria distributing legislative, administrative and financial powers between Bund and Länder. To this extent, legislative and administrative powers do not follow the criteria that allocate revenue powers: “the allocation of powers […] is done in a very complicated way, being, therefore, sometimes subject to [a] highly controversial interpretation”.55 We have already noticed that criteria allocating revenue-raising and taxing powers are set forth in the f-vg. Under the f-vg, “the federal parliament has to assign each tax […] and thus decides who is responsible for legislation and execution” thereover.56 Furthermore, “federal law must assign taxes within 53

On these exceptions, see Hellermann “Artikel 104a”, supra, at 1132ff.; J. Hellerman, “Artikel 104b”, in von Mangoldt et al, Kommentar zum Grundgesetz, Band 3, supra, at 1187ff. 54 See Feld and von Hagen, “Federal Republic of Germany”, supra, at 132. 55 Stelzer, The Constitution of the Republic of Austria, supra, at 153. 56 For further references, see ibid., at 159.

The Principles of Separation and Correspondence

99

an abstract constitutional framework of various types of shared or exclusive tax allocation”: exclusive federal taxes, taxes divided between federal, state and local level, exclusive state taxes and exclusive municipal taxes.57 On the one hand, fiscal powers are in the hands of the federation; on the other hand, the distribution of taxing powers does not match the distribution of legislative and administrative responsibilities. The former is ultimately determined by a federal legislative act, the “Fiscal Equalisation Act” (Finanzausgleichsgesetz, hereinafter fag), which is periodically renewed. Before passing the act, however, the federation must try to enter into an agreement with the Länder. This is the sole mechanism through which Länder can influence the federal decision-making process over the distribution of finance: “it is remarkable that the second chamber of the Austrian federal parliament, the federal council, has in these essential matters for Länder only a suspensive veto”.58 To sum up, the federation is responsible for enacting the majority of tax laws, whereas Länder have limited taxing powers and only participate in the distribution of revenues raised. The outcome is a system of intergovernmental financial relations, where the principle of correspondence is thoroughly disregarded. Relevant exceptions to the correspondence rule characterise regional countries, such as Italy and Spain, as well as devolutionary federal countries, such as Belgium and South Africa. Although the constitutions of Italy, Spain, Belgium and South Africa vest constituent units with taxing and expenditure powers, these provisions require implementation through national legislation.59 As far as South Africa is concerned, we have already noticed how national legislation limits provincial limited revenue-raising powers: “allowing provinces to choose applicable tax rates and tax bases could result in tax competition […] thus reinforcing economic disparities”.60 Further, Section 220 of the Constitution establishes a Financial and Fiscal Commission, which has to be consulted, for example, for approving the national legislation regarding: 57 58 59

60

Bußjäger, “Reforms on Fiscal Federalism in Austria”, supra, at 69. Ibid., at 60. Arts. 175–177 of the Belgian Constitution require an entrenched federal law for the implementation of the constitutional provisions over finances. The Italian national parliament shall have recourse to an ordinary act at the legislative level (Art. 119(2)). The Spanish national parliament does the same approving the lofca financial organic law passed by a majority of votes cast in the lower chamber in the final vote on the bill as a whole (Art. 157). See J. López-Laborda et al, “Kingdom of Spain”, in Shah and Kincaid, The Practice of Fiscal Federalism, supra, 288–316, at 296ff. See above, Section 2.

100

Nicolini

the equitable division of revenue raised nationally among the national, provincial and local spheres of government […] the determination of each province’s equitable share of the provincial share of that revenue: and any other allocations to provinces, local government or municipalities from the national government’s share of that revenue, and any conditions on which those allocations may be made.61 This upholds the idea of the mismatch between the legal and the economic fiscal federalism, and therefore the possibility of applying an economic-oriented concept of the correspondence rule. Finally, exceptions to the correspondence rule originate from administrative federalism, which indeed derogates to the dual-federalist principle, since the distribution of finance does not reflect the allocation of legislative functions. In federal and regional countries, the administration of a substantial portion of federal legislation is now constitutionally assigned to the government of the constituent units, such as in Switzerland, Austria, Germany, India and Malaysia.62 Hence, the functioning of the principle of correspondence shifts from legislative responsibilities to administrative powers: the latter must then be fully financed in order to discharge the functions related thereto.63 If we consider that taxes are levied or collected under the authority of law, the shift from legislative responsibilities to the administrative powers can be considered both as an outcome of Pound’s “discrepancy between doctrine in books and empirical data about law”,64 and as an acquisition of the practice of comparative law as a subversive discipline. 6

“How and Why Words Change Meaning”: The Shift From (Legal) Correspondence to (Economic) Equivalence

The principle of correspondence has been undergoing an even more remarkable shift that is having an impact not only on its constitutional regime and operational rule—and therefore the nexus between taxing, spending and responsibilities—but also the meaning of the principle at stake. This is what

61 62 63 64

B. Khumalo and R. Mokate “Republic of South Africa”, in Shah and Kincaid, The Practice of Fiscal Federalism, supra, 263–285, at 265. See, for example, s. 83 of the Constitution of India. See Watts, Comparing Federal Systems, supra, at 100. Halpérin, “Law in Books”, supra, at 46.

The Principles of Separation and Correspondence

101

linguists call “semantic change”:65 the shift is not related to the linguistic variety through which correspondence and separation display their function but, instead, impacts the lexical semantics of the legal language, i.e., the function of the correspondence rule itself. We have already highlighted how the subversive character of comparative legal studies presupposes the complexity and diversity of (both linguistic and legal) forms. Furthermore, we have pointed out that the comparative method accommodates diversity by grouping such a variety of forms because they are functionally equivalent, i.e., they fulfil similar functions in different legal systems.66 To put it another way, comparative legal studies reveal a fallacy in current research related to financial relations. This fallacy is not concerned with the variety and heterogeneity of the forms such relations take under federal constitutions; rather, it concerns the idea that there is “at the least a global legal language” capable of expressing such variety, and that “global legal language [is necessarily] based on economic models”.67 Such a shift in semantics is due to the intrinsic feature of legal language: the language of laws and statutes is characterised by neutrality and generality; it avoids subjective and personal attitudes and […] regional ­marking. To ensure correct and unambiguous transmission of information it must be conservative in its choice of structure and lexis and hostile to stylistic variation.68 And yet, these features “represent anonymous authority and power”:69 external factors, such as economic ones, may thus challenge the meaning of the legal lexicon related to the distribution of finance, and therefore impose a new sense (the language of economics) on existing lexical items. How this change in meaning occurs is due to pressure of economic models, which focus on “the problem of allocative efficiency rather than [on] questions of stabilization and income redistribution”.70 The process of semantic change affects the existing lexemes, because these adopt an additional meaning that 65 66 67 68

69 70

See C. Kay and K. Allan, English Historical Semantics (Edinburgh: Edinburgh University Press, 2015), 70ff. See above, Section 3. Muir Watt, Further Terrains for Subversive Comparison, supra, at 272. M. Rissanen, “Standardisation and the Language of Early Statutes”, in L. Wright (ed.), The Development of Standard English 1300–1800: Theories Descriptions Conflicts (Cambridge: Cambridge University Press, 2000) 117–130, at 120. Ibid., at 121. See Olson Jr, “The Principle of ‘Fiscal Equivalence’”, supra, at 482.

102

Nicolini

complements the old one.71 This entails a shift from the “legal” principle of correspondence towards the “economic” principle of fiscal equivalence,72 whose application requires that funds and proceeds be definitively allocated at the central/national level of government, which is capable of ensuring such efficiency. The shift is even more significant when the substitution of meaning is complemented by the substitution of the word, i.e., when economic lexemes replace legal jargon. This is apparent when it comes to the allocation of revenue powers, which is termed the “tax-assignment problem”.73 The linguistic change is even more apparent as far as the expression “financial relations” is concerned. To this extent, scholars also resort to a widespread selection of terms: “distribution of finance”, “fiscal relations”, “financial (or fiscal) arrangements”, “fiscal federalism”, “fiscal” or “financial constitution” and so on.74 The meanings of the expressions in question depend on whether financial relations are examined through the lens of economics or from a legal perspective. This is evident in the case of the expression “fiscal federalism”.75 Although it is very common among political scientists and jurists,76 its use is very 71

See D. Kastovsky, “Vocabulary”, in R. Hogg and D. Denison, A History of the English Language (Cambridge: Cambridge University Press, 2006) 199–270, at 215. 72 See above, Section 3. 73 See C.E. McLure (ed.), Tax Assignment in Federal Countries (Canberra: Australian National University, 1983); R.A. Musgrave, “Who Should Tax, Where and What?” in McLure, Tax Assignment in Federal Countries, supra, 2–19. For a critical approach to the tax-assignment problem, see C.E. McLure, “The Tax Assignment Problem: Ends, Means, and Constraints”, Public Budgeting and Financial Management, 9 (1998) 652–683; R. Bird, “Tax Assignment Revisited”, in J.G. Head and R.E. Krever (eds.), The 21st Century: A Volume in Memory of Richard Musgrave (The Hague: Kluwer Law International, 2009) 441–470. 74 Watts, Comparing Federal Systems, supra, at 95, employs the expression “distribution of finance”; Hogg, Constitutional Law of Canada, supra, at 6–1, uses the expression “financial arrangements”; H. Klug, The Constitution of South Africa: A Contextual Analysis (Oxford: Hart Publishing, 2010), 216, uses “fiscal constitution”. 75 See W.E. Oates, “An Essay on Fiscal Federalism”, Journal of Economic Literature, 37 (1999) 1120–1149, at 1120–1121. The expression was first used in R.A. Musgrave, The Theory of Public Finance (New York: McGraw-Hill, 1959) and then in subsequent studies. See R.A. Musgrave, “Economics of Fiscal Federalism”, Nebraska Journal of Economics and Business, 10 (1971) 3–13; W.E. Oates, Fiscal Federalism (New York: Harcourt Brace Jovanovich, 1972). See also G. Färber, “Was ist Finanzföderalismus?” in Bußjäger, Perspektiven des Finanzföderalismus, supra, 11–38. 76 Among others, see Shah and Kincaid, The Practice of Fiscal Federalism, supra; Watts and Kincaid, “Introduction”, supra, at ix; E. Ahmad and G. Brosio, Handbook of Fiscal

The Principles of Separation and Correspondence

103

contentious. “Fiscal federalism” pertains to a “subfield of public finance”, and therefore denotes “which functions and instruments are best centralized and which are best placed in the sphere of decentralized levels of governments”. Fiscal economists consider “state and local finance as a field of research”77 for the allocation of revenue and expenditure powers, and address the following questions: “which taxes are best suited for use at the different levels of government?”; “placed in a context of a federalist government, should distribution be a national or a local responsibility?”78 These questions denote fields of research that are very different from those characterising comparative federal studies. Public finance indeed has a normative approach, and “fiscal federalism” is therefore an expression that entails a non-legal definition of financial relations: there is a need for every collective good with a unique boundary, so that there can be a match between those who receive the benefits of a collective good and those who pay for it. This match we define as “fiscal equivalence”.79 The expression does not consider the legal features—in particular, the constitutional ones—related to the allocation of financial powers. Fiscal federal studies are aimed at maximising the efficiency of public policies and services by establishing an appropriate distribution of revenue powers. On the contrary, comparative federal studies have a different approach to financial relations. “Legal” fiscal federalism analyses the constitutional foundations of financial powers and the mechanisms through which national and subnational levels appropriate funds in order to finance their constitutional responsibilities. It also examines revenue raising and revenue distribution, as well as equalisation and the scope of spending powers. Financial autonomy,

77 78

79

Federalism­ (Cheltenham-Northhampton: Edward Elgar, 2006); R. Boadway and A. Shah (eds.), Fiscal Federalism: Principles and Practice of Multi-order Governance (Cambridge: Cambridge University Press, 2009). Musgrave, “Economics of Fiscal Federalism”, supra, at 3. See Oates, “An Essay on Fiscal Federalism”, supra, at 1125, and R.A. Musgrave, “Devolution, Grants, and Fiscal Competition”, The Journal of Economic Perspectives, 11 (1997), 65–72, at 67. See also R.M. Bird and F. Vaillancourt (eds.), Perspectives on Fiscal Federalism (Washington: World Bank Institute, 2006). Quotations are from Olson Jr, “The Principle of ‘Fiscal Equivalence’”, supra, at 482, 483, 485, respectively.

104

Nicolini

solidarity and spending powers are the most relevant fields of research in “legal” fiscal federalism.80 In this regard, the expression “fiscal federalism” seems a mere projection of the above-mentioned principle of fiscal equivalence, and underpins the idea that legal financial relations must be governed by budgetary policies and therefore by economic theories. Public finance, in general, and fiscal equivalence, in particular, express what I have already termed the “efficiency rule”, that is, the optimisation of the allocation of proceeds, the strengthening of fiscal responsibility, as well as the promotion of an equilibrium between the estimates of revenues and expenditures. But such an efficiency rule does not match “legal” fiscal federalism, which upholds the idea that the budgetary position of federal and regional states should be balanced or in surplus. The normative (i.e., theoretical) approach of public finance budgetary is not the result of the economic reflexes of the legal correspondence rule: as for the revenue side, de facto imbalances, social federalism and democratic accountability are not considered as necessary presuppositions that allow the allocative efficiency model to work. If we turn to the expenditure side, the public finance model is challenged by the lack of legal limitation in the exercise of spending powers. To sum up, both taxing and spending powers are in the hands of the political branches, and the effective amount of money to be raised by taxation and collected under the authority of law depends on the economic situation. The “legal” applicability of the “economic” principle of balanced budgets in financial relations thus depends on whether it is enshrined in federal and regional constitutions, such as in Switzerland. The budget must to be balanced both at the federal level (Article 126 of the Constitution)81 and at the cantonal level. For instance, the cantonal constitutions of St Gall, Solothurn, Grison and Appenzell Outer-Rhodes prescribe that governments “accumulate savings in order to equalize revenue fluctuations over the business cycle [and] to build up reserves in good times and to eliminate structural deficits”.82 This holds true in the United States of America, where state constitutions require that expenditures match revenues: the “golden” budgetary rule imposes that a surplus of revenues be accumulated in “good economic times” as “‘rainy day’ 80 81 82

See Watts and Kincaid, “Introduction”, supra, at ix ff. T. Stauffer, “Art. 126”, in Ehrenzeller et al, Die schweizerische Bundesverfassung, supra, 1944–1957. Kirchgässner, “Swiss Confederation”, supra, at 328. See also Kirchgässner, “Finanzföderalismus in der Schweiz”, supra, at 46ff.

The Principles of Separation and Correspondence

105

funds that can cover what would otherwise be fiscal deficits in bad economic times”.83 The same occurs in Brazil, where the Fiscal Responsibility Act 2000 established several mechanisms that are aimed at enforcing constituent units’ fiscal discipline.84 Changes in financial relations are due to the fact that economic budgetary principles are offshoots of economic-oriented representations of law, which promotes the global economic legal language “spoken” by international financial actors, i.e., the above-mentioned process of semantic change that is affecting comparative legal studies. Against this background, additional efforts in the practice of subversive comparison are required, and scholars should highlight the change in legal lexical semantics, and therefore warn against the application of public finance budgetary principles to financial relations. Finally, the shift from “legal” correspondence to “economic” fiscal equivalence was triggered by the ongoing economic and financial crisis. The assumption is held when we consider how, in the course of this crisis, several eu member­states have amended their constitutions to introduce the principle of a balance between revenue and expenditure. This occurred in nearly all eu federal and regional countries, such as Austria, Germany, Spain and Italy.85 Thus, limitations on subnational financial powers are the outcome of the entrenchment of economic budgetary policies, since the above-mentioned constitutional reforms ensure that the budgetary position of the member states is balanced or in surplus according to Article 136 tfeu and to the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union’ (tscg) signed in Brussels on 2 February 2010 also known as the “fiscal compact”. In this respect, eu constraints on budgetary rules deriving from the tscg can be conceived as part of the federal-like integration process resting upon economic principles that progressively deprive member states of their constitutional Kompetenz-Kompetenz. The outcome is twofold. 83 84 85

M. Tushnet, The Constitution of the United States of America: A Contextual Analysis (Oxford: Hart Publishing, 2015), 161. F. Rezende, “Federal Republic of Brazil”, in Shah and Kincaid, The Practice of Fiscal Federalism, supra, 74–97, at 82–84. For Austria, see Art. 13(2) b-vg, as inserted by the 2008 constitutional reform (BGBl. i 1/1008). For Germany, see Arts. 109 and 110 gg, as amended by the 2009 constitutional reform. For Spain, see Art. 135 of the Constitution, amended in 2011. For Italy, see Arts. 81 and 119(1) of the Constitution, as amended by constitutional (amendment) act no. 1/2012. See T. Groppi, “The Impact of the Financial Crisis on the Italian Written Constitution”, Italian Journal of Public Law, 4 (2012) 1–14.; D. Delgado Ramos, “La reciente constitucionalización de la estabilidad presupuestaria ¿Una reforma necesaria?” Revista de Derecho Político, 87 (2013) 317–352.

106

Nicolini

On the one hand, the eu has not yet fostered the elaboration, transmission or adaptation of a common thick constitutional identity. This is particularly true when it comes to those abstract commitments to human rights, rule of law and democracy that are enshrined in Article 2 of the eu Treaty: in this regard, the eu is merely indicating allusive, vague circumlocutions that do not contribute to the formation of a thick, common constitutional identity. In this thin commitment, “there is nothing specifically European”.86 Allusiveness is thus the allegory of the present of the eu. The creation of a thick cultural, social, literary identity cannot merely resort to abstract commitments and to machineries governing financial governance; public discourse cannot be limited to those financial elites supporting an overturn of the foundations of the European integration process. When addressing a financial crisis, a constitutional identity-building process cannot rest on the mere creation of mechanisms for financial governance, such as those established by the tsgc. The us constitutional/federal experience might certainly provide Europe with additional solutions to face the economic crisis. The financial divide between the eu member states reveals the lack of a sole demos, and impedes narratives similar to those leading to the establishment of an ever more perfect union in the United States, i.e., those narratives that persuaded Alexander Hamilton to “successfully [restructure] America’s crippling sovereign debt in the 1790s by ‘federalizing’ the states’ debt”.87 On the other hand, the shift not only entails a change in the nexus between democracy, accountability and financial responsibilities but also the deletion of constructs purporting one of the main core values upon which constitutionalism was erected, that is, the idea of the supremacy of those same constitutions: The stability treaty not only requires … constitutional changes in each of the signatory states, but also raises significant questions about its relationship with eu law and the extent of the discretion left to member states to make fundamental decisions about taxation and spending.88 86

87 88

M. Kumm, “The Idea of Thick Constitutional Patriotism and its Implications for the Role and Structure of European Legal History” in H. Porsdam and T. Elholm (eds.), Dialogues on Justice: European Perspectives on Law and Literature (Berlin: Walter de Gruyter, 2012) 108–137, at 109. A. Loubert, “Sovereign Debt Threatens the Union: the Genesis of a Federation”, European Constitutional Law Review, 8 (2012) 442–455, at 442. S. Peers, “The Stability Treaty: Permanent Austerity or Gesture Politics?” European Constitutional Law Review, 8 (October 2012) 404–441, at 404.

The Principles of Separation and Correspondence

7

107

The Third Shift: From Economic Equivalence to Multilayered Government

Intergovernmental financial relations are not limited to the federal and intermediate tiers of government; they also encompass the regulation of local financial powers and their interrelation with the federal and intermediate levels. The shift towards a principle of correspondence inclusive of all the tiers of government is evident in aggregative federations and in regional states, where constitutions entrench the sphere of guaranteed autonomy attributed to local authorities and set down an exhaustive regulation that national and subnational legislation has to subsequently enforce. As a consequence, economic “fiscal federalism” does not adequately outline the role attributed to the different tiers of government in the distribution of finance. The “economic” approach to the correspondence rule merely focuses on the “efficiency rule” for the best allocation of functions between the central and the decentralised levels of governments, but it does not legally define which are the decentralised levels of government concerned. Moreover: the traditional theory of fiscal federalism lays out a general normative framework for the assignment of functions to different levels of governments and the appropriate fiscal instruments for carrying out these functions.89 On the contrary, comparative legal studies do differentiate between federal, state and local governments. In this respect, the origins—aggregative or devolutionary—­and the basic features of federal and regional constitutions assign different institutional responsibilities to constituent units and local authorities. In aggregative federations, member states have exclusive jurisdiction over municipal institutions: the idea of federal intervention into municipal … affairs seems inconsistent with the conception of municipalities as creatures of the state of which they are political subdivisions.90 89 90

Oates, “An Essay on Fiscal Federalism”, supra, at 1120–1121. C.P. Gillette, “Fiscal Federalism, Political Will, and Strategic Use of Municipal Bankruptcy”, The University of Chicago Law Review, 79 (2012) 281–330, at 298. The assumption according to which local authorities are “creatures” of the constituent units can be traced back to several constitutional provisions: see the 10th Amendment (1791) to the u.s. Constitution,

108

Nicolini

In Latin American federations and in regional states, constitutions expressly entrench the local autonomy and constrain the legislative powers over constituent units.91 In some cases—as in Italy and South Africa—the central state has the major legislative power, and constituent units have a limited jurisdiction over local government.92 The outcome is an intricate concurrency between national and subnational jurisdictions, and therefore national and subnational legislative powers affecting the principle of correspondence at the local level of government frequently overlap. In the United States of America, the federal order of government has recourse to several instruments—such as conditional and unconditional grants, tax agreements, etc.—in order to bypass the state level and allocate funds to the local one, thus “influencing” state–local relations.93 In Germany, the proceeds of local governments are part of the total amount of money the Bund transfers to each Land as a share of total revenue from joint taxes (Article 106(7) gg). As a consequence, revenues and expenditures of municipalities are deemed to be revenues and expenditures of the Länder (Article 106(9) gg). In most cases, financial relations give rise to a “three-layered federalism” and a multi-layered principle of correspondence,94 such as in Belgium, Italy, Spain, South Africa, Russia, Brazil and Austria. In Austria, Article 3 fag allows the Bund to have recourse to grants in order to transfer money to municipalities for specific purposes. This assumption holds true for Spain, where Article 142 of the Constitution and law No. 39/1998 on local fiscal management establish that the central level and the autonomous Communities must provide

91

92

93 94

s. 92(8) Constitution Act 1867 of Canada, Art. 50(1) of the Swiss Constitution, Arts. 28 and 104a gg, Art. 115(2) b-vg. Articles 115 and 115-I of the Mexican Constitution; Art. 123 of the Argentinian Constitution; Arts. 1, 18, 29, 30, 35 of the Brazilian Constitution; Arts. 243 and 243Q of the Constitution­of India; Arts. 137, 101, 141, 149(1)(18) of the Constitution of Spain; Art. 162 of the Constitution of Belgium, as implemented by the “special” law of 13 July 2001, which enables regions to set the basic organisational features of Belgian local government. See Arts. 117(2)(p), 118(1)(2) and 123(4) of the Italian Constitution; s. 40 of the Constitution of the Republic of South Africa 1996. It should be recalled that the five Italian “autonomous” regions (Trentino-Alto Adige/Südtirol, Friuli-Venezia Giulia, Aosta Valley/ Vallée d’Aoste, Sicily and Sardinia: Art. 116(1) of the Constitution) have exclusive jurisdiction over local authorities situated within their boundaries. See F. Palermo, “Asymmetric ‘Quasi-Federal’ Regionalism and the Protection of Minorities: the Case of Italy”, in G.A. Tarr et al (eds.) Federalism, Subnational Constitutions, and Minority Rights (Westport, ct: Praeger, 2005) 107–131, at 109. See also Chapter 14 of this volume, “Local Governments in African Federal and Devolved Systems of Government”, by N. Steytler and Z. Ayele. See Fox, “The United States of America”, supra, at 364. Bußjäger, “Reforms on Fiscal Federalism in Austria”, supra, at 61.

The Principles of Separation and Correspondence

109

local governments with an adequate amount of funds to fully finance their responsibilities.95 Both federal and regional constitutions allow local authorities to levy taxes in order to finance their responsibilities. In Canada, “municipalities levied taxes before any provincial tax was enacted”.96 In the United States of America, local authorities have been imposing property taxes in order to generate their own revenues from the inception of the federation.97 Under Article 106(6) gg, revenues from taxes on real estate and commerce, from local taxes on consumption accrue to the municipalities. Furthermore, local authorities levy taxes, impose fees, have recourse to public debt and receive financial transfers from national and intermediate levels of government: these mechanisms grant local entities the appropriate amount of resources in order to address the needs of their communities. Questions arise, however, when we try to specify the basic features—established by either constitutional or legislative provisions—of local taxing powers. The first point at issue is represented by the fact that, when it comes to local taxing policies, the principle of correspondence requires the intervention of the national or intermediate legislative branch. Indeed, the principles of “no taxation without representation” and of legality prescribe that a legislative act allows municipalities to appropriate funds through taxation and to raise proceeds. This limits local financial autonomy: indeed, local government may levy, collect, appropriate taxes and determine additional rates in accordance with the principles set forth by federal and subnational law.98 The second point at stake is related to the constitutional regime of the principle of correspondence as far as local financial powers are concerned. To this extent, the constitutions establish highly differentiated rules regarding local financial responsibility and accountability. In aggregative federations, the constitutions only establish the basic features of local financial powers, and then confer upon subnational legislatures the duty of specifying which powers are actually assigned to local authorities. The German Basic Law establishes the principle of local financial autonomy (Article 28); it sets several mechanisms for federal intervention in financial relations between Länder and municipalities 95

F. Velasco Caballero, “Kingdom of Spain”, in N. Steytler (ed.), Local Government and Metropolitan Regions in Federals Systems (Montreal: McGill–Queen’s University Press, 2009) 299–328, at 314ff. 96 Hogg, Constitutional Law of Canada, supra, at 6–2. 97 See Fox, “The United States of America”, supra, at 354. 98 See Arts. 106 gg, 7(5) and 8(5) f-vg, s. 243H of the Constitution of India, Arts. 117 and 119 of the Italian Constitution.

110

Nicolini

(Articles 106, 107 gg);99 it provides that a share of the revenue from income tax shall directly accrue to the municipalities (Article 106(5) gg). The implementation of this constitutional provision is then committed to paragraphs 6(1), 7(1)(2), 8, 9(3), 13 and 17 of the Gesetz über den Finanzausgleich zwischen Bund und Ländern (fag), i.e., the federal law on equalisation.100 Finally, several federal constitutions entrench the sphere of guaranteed autonomy attributed to local authorities, and then set down an exhaustive regulation of local government that national and subnational legislation has to enforce.101 This is apparent in South Africa. In Liebenberg, the court states: A municipality’s authority to impose rates and levies is derived from Section 229 of the Constitution. The purpose of a municipality’s revenueraising powers is to finance a municipality’s performance of its constitutional and statutory objects and duties as set out in Sections 152(1) and 153 of the Constitution. These include the provision of services to communities in a sustainable manner, promoting social and economic development and providing for the basic needs of the community. These objects are integral in the task of constructing society in the functional areas of local government.102 8

Crossing the Public–Private Divide: The Principle of Correspondence as a Guarantee to be Traded in Equity Markets?

Like the overlapping tiles of a roof, the three shifts in the operational rule of the correspondence principle uphold a huge transfiguration of the same rationale of financial relations.

99

See P.J. Tettinger and K.-A. Schwarz, “Artikel 28”, in H. von Mangoldt et al. (eds.), Kommentar zum Grundgesetz, Band 2: Artikel 20 bis 82 (München: Verlag Franz Valen, 2010) 567–641, at 629ff. 100 See M. Burgi, “Federal Republic of Germany”, in Steytler, Local Government and Metropolitan Regions, supra, 137–165, at 148ff. 101 See ss. 40 and 220 of the South African Constitution, Arts. 156 and 159, para. 3 of the Brazilian Constitution; s. 243H of the Indian Constitution for Panchayats. As for Brazil, see Rezende, “Federal Republic of Brazil”, supra, at 80. Under the 73rd and the 74th constitutional amendments (1992), “each of the [Indian] state governments has devolved powers to levy certain taxes and fees [on] local bodies”: see G. Rao, “Republic of India”, in Shah and Kincaid, The Practice of Fiscal Federalism, supra, 152–177, 156. 102 Liebenberg no v Bergrivier Municipality, supra, at 40.

The Principles of Separation and Correspondence

111

As already mentioned, this transfiguration was caused by the financial crisis, and the outcome has been a paradox: the efficiency rule and the economic models are now causing centralisation in financial relations. However, fiscal equivalence policies still presuppose multilayered federalism, where units count as financial recipients: it is already evident that both the “centralizing” and “decentralizing” ideologies are wrong, or at any event entail inefficiency. Only if there are several levels of government and a large number of governments can immense disparities between the boundaries of jurisdictions and the boundaries of collective goods be avoided.103 The shift from legal to economic principle of correspondence allows make “comparative law by numbers”.104 On the one hand, there is the idea that “law matters”: legal institutions have an impact on economic growth. This is in tune with neoclassical law and economics, which is based essentially on the idea that law should be measured by the incentives it sets for welfaremaximizing conduct.105 On the other hand, the Doing Business Reports of the Word Bank, i.e., “crosscountry comparisons including rankings of the attractiveness of different legal systems for doing business”,106 have led to the idea that it is possible to evaluate the economic performance of legal systems by applying quantitative methodologies and numeric indicators: The promise of evidence-based policy-making is that it is not only more objective and less prone to misuse, but also more transparent, more democratic, and more open to public debate than decisions taken by politicians and business leaders with references to qualitative forms of knowing.107 103 Olson Jr, “The Principle of ‘Fiscal Equivalence’”, supra, at 483. 104 R. Michaels, “Comparative Law by Numbers? Legal Origins Thesis, Doing Business Reports, and the Silence of Traditional Comparative Law”, The American Journal of Comparative Law, 57 (2009) 765–795. 105 Ibid., at 768. 106 Ibid., at 766. 107 R. Rottenburg and S.E. Merry, “A World of Indicators: The Making of Governmental Knowledge through Quantification”, in R. Rottenburg et al (eds.), The World of Indicators:

112

Nicolini

And “yet, the creation of these systems is rarely transparent of public”.108 The application of indicators complements the public finance budgetary principles, the economic-oriented representation of the law, the global legal– economic language and requires highly centralised and efficient decisionmaking processes as regards fiscal powers. Such a trend has been accentuated by the steady increase of the spread that upset the European Union, in general, and Spain and Italy, in particular, in recent years. These two countries are the only eu member states with a regional government framework. The growing spread between their bonds and those issued by Germany entailed an increase in the interest to be paid for refinancing the public debt. This led to the approval of extraordinary measures undermining Italian and Spanish regional financial powers. Indeed, the necessity of restricting any increase in the public debt resulted in the adoption of severe cuts on the amount of grants transferred to regions, provinces and municipalities, thus limiting regional and local spending powers.109 Needless to say, the increase in interest rates had a severe backlash throughout the European economic and monetary union,110 and the eu adopted several measures that gave legal relevance to the budgetary policies we examined above. This change in the meaning of financial relations, in general, in the applicability of the principle of correspondence, in particular, is threefold. First, there is a shift between centralisation, which seems to reflect the anxiety towards the creation of a global legal language and common regulation of financial relations in decentralised countries and markets. Second, there is an even more remarkable shift towards the efficiency rule, whose application requires that funds and proceeds be definitively allocated at the central/national level of government, which is financially responsible vis-à-vis financial markets and international investors—to put it differently, the national government is accountable as far as global economic governance is concerned. Third, financial global dominance causes a shift from the political to the economic sphere: and The Making of Governmental Knowledge through Quantification (Cambridge: Cambridge University Press, 2015) 1–14. 108 See ibid., at 1–2. See also H. Blöchliger, “Measuring Decentralisation: The oecd Fiscal Decentralisation Database”, in J. Kim et al (eds.), Measuring Fiscal Decentralization. Concepts and Policies (Paris: oecd Publishing, 2013) 15–36. 109 See F. Palermo and A. Valdesalici, “Italy: Autonomism, Decentralization, Federalism, or What Else?” in J. Lluch (ed.) Constitutionalism and the Accommodation in Multinational Democracies (Oxford: Palgrave MacMillan, 2014) 180–199. 110 See V. Borger, “How the Debt Crisis Exposes the Development of Solidarity in the Euro Area”, European Constitutional Law Review, 9 (2013) 7–36.

The Principles of Separation and Correspondence

113

the discretionary powers of political branches in levying taxes and appropriating funds rest on a new form of “confidence” between the political power and the sovereign financial market, where “distressed sovereign debt can be sold on private equity markets and the debtor [is] subjected to the harsh economics of private law”.111 It follows that there has been a complete overturn in the political debate and in the meaning of accountability, responsibility and capability of appropriating funds in federal states. The current financial crisis is thus undermining the same concepts of intergovernmental financial relations, as well as the same presuppositions of the federal-like eu integration process. The lack of a legal narrative representing the supranational body politic is manifest when it comes to the current economic and financial crisis. This is undermining the same presuppositions of the eu integration process and, to a bigger extent, the same identity-building process. In this regard, the crisis gives rise to issues that are related to global economic governance. Hence, international financial actors such as the World Bank and the International Monetary Fund,112 as well as private-sector investors, endorse the transformation of the legal and economic premises of financial relations. In particular, international financial actors suggest the adoption of a model capable of supporting a capitalist socioeconomic model, but that totally departs from the model of Soziale Marktwirtschaft (i.e., social market economy) enshrined in Article 3 teu.113 This is caused by international investment law, which “shifts power and authority from states to investors, tribunals and other decision-makers”, and “[t]hese shifts produce outcomes that only partially support global policies”, as well as the transfer of power and authority to decision-makers who are not democratically accountable.114 As this essay has highlighted, the principle of correspondence has both an economic-oriented significance and a new, subversive meaning, which crosses 111 Muir Watt, Further Terrains for Subversive Comparison, supra, at 286. 112 It is “remarkable” that the institutions of the World Bank Group “have reached their present status as the premier source of both development finance and economic research and information without introducing any major change in their constituent charters”. See I. Shihata et al (eds.), The World Bank in a Changing World. Selected Essays (Dordrecht: Martinus Nijhoff Publishers, 1991), 15. 113 On the negation of the Soziale Marktwirtschaft model in Germany, the country where the model was invented, see M. Ruffert, “Public Law and the Economy: A Comparative View from the German Perspective”, International Journal of Constitutional Law, 11 (2013) 925–939. 114 T.-C. Cheng, “Power, Authority and International Investment Law”, American University International Law Review, 20 (2005) 465–520, at 469.

114

Nicolini

the public/private divide, and is subsequently enforced under private-law mechanisms. In this respect, centralising trends must now cope with two necessities in economic-oriented systems of financial relations. On the one hand, the “correspondence rule” requires the application of the efficiency rule to the revenue–responsibilities equation and to the necessity of ensuring the homogeneity of economic and social conditions. On the other hand, this rule must cope with the new “legal” budgetary policy—budgetary positions must be balanced or in surplus—that percolates through the whole federal constitutional design. This entails a reconfiguration of federal and regional financial relations: the principle of correspondence now rests on the sustainability of what we may call “financial equivalence”, which can be traded in global equity markets. Further, this imposes an extreme connection between the powers of levying taxes, of imposing fees and equalisation mechanisms in order to allow constituent units to fully discharge their constitutionally assigned responsibilities under the budgetary guardianship of an ever more invasive central level of government.

Part 2 Foundations

∵

section 1 The Distribution of Powers

∵

chapter 6

Accountability and Revenue Assignment across Levels of Government: Rules, Practices, and Challenges Maria Flavia Ambrosanio, Paolo Balduzzi, and Claudia Peiti 1

Introduction

Economic theory suggests that revenue assignment to subnational governments (sngs) should also be designed with the aim of increasing the accountability of subnational policymakers. Thus, a certain degree of tax autonomy plays a central role, as it places great pressure on subnational administrators to manage public spending efficiently. This chapter focuses on a number of questions concerning the revenue assignment problem and is organised as follows. Section 2 briefly reviews the literature concerning the link between subnational governments’ accountability and the different sources of financing for sngs. Section 3 describes the practices of revenue assignment in a set of oecd countries, paying particular attention to the issue of the vertical fiscal gap that arises when the decentralisation of spending responsibilities is not matched with an adequate decentralisation of revenue autonomy. In many countries, a significant amount of subnational revenues comes from shared taxes, and this gives rise to certain problems, especially horizontal and vertical externalities, and to practical difficulties in measuring the taxing power of sngs. Sections 4 and 5 examine these issues in depth. Finally, Section 6 describes four case studies that exemplify all the possible combinations between the taxing power and the vertical fiscal gap. 2

Accountability and Revenue Assignment

Economic theory suggests that, in general, some degree of financial autonomy for sngs renders policymakers more accountable and could improve economic efficiency. Nonetheless, it is important to stress both the merits and the demerits of financial autonomy. In this section, we briefly discuss these issues, focusing on the relationship between taxing power and accountability.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_008

120

Ambrosanio, Balduzzi, and Peiti

Financial autonomy relates to the ability of an sng both to raise revenues from the subnational economy and to decide how to spend those revenues. Thus, it concerns both the revenue assignment and the expenditure assignment problem. For the purposes of this paper, we focus on the former. According to the most relevant literature on fiscal federalism, the revenue autonomy (or taxing power) of sngs—namely, a certain degree of control over independent sources of revenue—has different kinds of benefits, both economic and political, which are often interrelated. First of all, tax autonomy allows sngs to determine the level and quality of public services, according to local preferences, so as to link the benefits of public expenditures to the taxes paid by citizens. In other words, residents of sngs can choose the level of public services they desire by choosing to pay higher or lower taxes. It is very important that sngs be assigned marginal sources of own revenues, so that they can marginally increase (or reduce), and to some degree control, the amount of resources for the provision of public goods and services. Thus, tax autonomy places great pressure on subnational administrators to manage subnational government spending efficiently, because it is “not apparent, at the margin, that an additional pound spent in subnational services will be equal to the benefit of an equivalent reduction in taxation”.1 Second, tax autonomy places greater responsibility on subnational politicians and could help increase their accountability for using resources appropriately. This refers to a form of vertical-upward accountability in that citizens would be able to monitor the behaviour of subnational politicians, managers and administrators and “hold them to account for their failure or success to provide information and justifications for their decisions”.2 If citizens were dissatisfied, they could, for example, punish subnational politicians at election time. Following Bentham, “the more strictly we are watched, the better we behave”.3 Tiebout advanced a very general accountability argument in his classic “voting with your feet” model:4 if voters are free to move across jurisdictions, autonomous decentralised governments can compete to attract (or reject) them, both on the expenditure side and on the revenue side. 1 J. Darby et al., “Fiscal Federalism and Fiscal Autonomy: Lessons for the uk from Other ­Industrialized Countries”, Scottish Affairs, 41 (2002) 26–55, at 14. 2 S.I. Lindberg, “Mapping Accountability: Core Concept and Subtypes”, International Review of Administrative Sciences, 79 (2013) 202–226, at 15. 3 As quoted by J.E. Crimmins, “Jeremy Bentham”, in E.N Zalta (ed), The Stanford Encyclopedia of Philosophy (Spring 2015), http://plato.stanford.edu/archives/spr2015/entries/bentham/ (accessed 26 November 2015). 4 C.M. Tiebout, “A Pure Theory of Local Expenditures”, Journal of Political Economy, 64 (1956) 416–424.

Rules, Practices, and Challenges

121

More recently, Gadenne provided an explicit theoretical model underpinning this mechanism, even in the absence of mobile voters.5 Following Besley and Smart on the role of asymmetric information,6 Gadenne argues that voters observe local taxes more closely than inter-governmental grants; hence, local politicians have less incentive to divert public money to unproductive expenditures (e.g. private rents, corruption and so on) when these resources come from local taxes.7 Third, if sngs are assigned own sources of revenue and the ability to set tax rates, to a certain extent, this may also increase accountability thanks to the positive effects of tax competition, which refers to the strategic interaction of tax policy among sngs. Tax competition works in different ways. sngs may compete to attract businesses and mobile activities, thus promoting local and regional economic development and increasing the tax base and hence tax revenues.8 On the other hand, if subnational politicians are not benevolent, as in the Brennan-Buchanan approach, and instead they behave like Leviathans, and “taxes are used to maximise total revenue from private sector as this allows politicians and bureaucrats to maximize their spending power”,9 then excessive increases in tax rates would be punished by the loss of the tax base, so tax competition can protect citizens from the rapaciousness of politicians and bureaucrats.10 Fourth, subnational taxation may enhance the accountability of subnational politicians through so-called yardstick competition, which may act as a ­mechanism to reduce political rent-seeking. By observing tax policy in neighbouring jurisdictions, citizens may be better informed, can monitor the performance of subnational incumbents (it is easier to compare tax rates instead 5

6 7 8

9

10

L. Gadenne, “Tax Me, But Spend Wisely? Sources of Public Finance and Government Accountability”, Working Paper, (2015), https://sites.google.com/site/lgadenne/home/­ research (accessed 25 November 2015). T. Besley and M. Smart, “Fiscal Restraints and Voter Welfare”, Journal of Public Economics, 91 (2007) 755–773. Gadenne, “Tax Me, But Spend Wisely?” supra. H. Blöchliger and J. Pinero Campos, “Tax Competition between Sub-central Governments”, oecd Working Papers on Fiscal Federalism, 13 (2011), http://dx.doi.org/10.1787/ 5k97b1120t6b-en (accessed 12 December 2015). M.F. Ambrosanio and M. Bordignon, “Normative versus Positive Theories of Tax Assignment in Federations”, in E. Ahmad and G. Brosio (eds), Handbook of Multilevel Finance (Cheltenham: Edward Elgar Publishing, 2015) 231–263, at 240. R.M. Bird and F. Vaillancourt (eds), Perspectives on Fiscal Federalism (Washington, dc: World Bank, 2006); R.D. Ebel and J. Petersen (eds), The Oxford Handbook of State and Local Government Finance (Oxford: Oxford University Press, 2012); C.E. McLure Jr., The Tax Assignment Problem: Conceptual and Administrative Considerations in Achieving Subnational Fiscal Autonomy, (Stanford: Hoover Institution, 2007).

122

Ambrosanio, Balduzzi and Peiti

of ­complicated expenditure structures) and have greater incentive to check how tax revenue is spent than they would if public goods and services were financed by grants.11 Finally, tax autonomy gives subnational politicians and managers strong incentives to enforce a hard budget constraint, with positive effects on the fiscal balance of general government, given the direct link between subnational taxes and subnational expenditures. As argued by Shah, “to ensure fiscal discipline, governments at all levels must be made to face [the] financial consequences of their decisions”.12 However, tax autonomy also has certain demerits that are worth mentioning. First, sngs may have different fiscal capacities as a result of their different levels of income, wealth and economic activity. Given the same fiscal effort, poorer jurisdictions collect less revenues than richer ones, and this results in lower levels of services offered to citizens. Hence, tax autonomy may create equity issues among different sngs. Second, autonomous sngs may have an incentive to shift the burden of taxation onto non-residents (so-called tax exporting); thus, the costs of public services are not internalised by residents and subnational politicians.13 Third, tax competition could have negative effects (predatory tax competition) if it were to lead to the undesired migration of factors of production (especially capital) and mobile tax bases. An additional argument against full tax autonomy is that it can give rise to far too complex administrative structures for the smallest subnational authorities; moreover, it requires a necessary duplication of bureaucratic bodies, which increases the general cost of tax collection. A final argument relates to the higher likelihood of corruption or capture by interest groups; sngs may be more affected by the actions or requests of subnational elites and lobbies than national governments. One natural way to tackle these drawbacks within a financial autonomy framework would be to use intergovernmental transfers funded by general taxation as a supplementary means of finance.14 Following Charbit, grants are a natural instrument for internalising externalities in the provision of public 11 12

13 14

Ambrosanio and Bordignon, “Normative versus Positive Theories”, supra. A. Shah, “Fiscal Federalism and Macroeconomic Governance: For Better or Worse?” World Bank Policy Research Working Paper, 2005 (1999), http://ssrn.com/abstract=597215 ­(accessed 10 December 2015), at 34. C.E. McLure, “The Interstate Exporting of State and Local Taxes: Estimates for 1962”, ­National Tax Journal, 20 (1967) 49–77. H. Blöchliger and O. Petzold, “Finding the Dividing Line between Tax Sharing and Grants: a Statistical Investigation”, oecd Working Papers on Fiscal Federalism, 10 (2009); H. Blöchliger and O. Petzold, “Taxes and Grants: on the Revenue Mix of Sub-central

Rules, Practices, and Challenges

123

services at the subnational level, redistributing revenue across regional governments within a country, instigating development and growth in certain regions and taking advantage of economies of scale in tax collection by the central government.15 A grant system must, however, be designed in order to avoid well-known, undesired drawbacks. Grants may decrease sngs’ incentives on tax effort,16 to avoid excessive spending and to control deficits and debts, hence leading to problems of soft-budget constraint and to bailout interventions by the central government. They also weaken accountability by eliminating the direct link between those who benefit from subnational public services and those who pay for them. To sum up, subnational tax autonomy has many advantages and a few disadvantages, but their relative importance depends on the design of local subnational taxation (see Section 4). 3

Practices of Revenue Assignment around the World

How are taxes and grants combined in various systems around the world? In Table 6.1 and Table 6.2, we present data for a set of federal and unitary oecd countries, and, for federal countries, we discriminate between the local and the state level. Table 6.1 shows that sngs’ revenues as a share of general government revenues have, on average, not changed significantly in the last two decades (from 18.7 per cent in 1995 to 19.8 per cent in 2005 and 2013), although there have been differences within individual countries. For instance, the share decreased in Austria, Denmark, Ireland, Luxembourg and the Netherlands, while it increased in Belgium, Finland, Germany and Italy. Moreover, in each of the years considered, we see a great deal of variability, as measured by the coefficient of variation (around 70 per cent). As a matter of fact, in 2013 the share of subnational revenues over general government revenues varied from

15

16

Governments”, oecd Working Papers on Fiscal Federalism, 7 (2009), http://dx.doi.org/ 10.1787/5k97b11972bn-en (accessed 12 December 2015). C. Charbit, “Explaining the Sub-national Tax-grants Balance in oecd Countries”, oecd Working Papers on Fiscal Federalism, 11 (2009), http://www.oecd-ilibrary.org/taxation/ explaining-the-sub-national-tax-grants-balance-in-oecd-countries_5k97b10s1lq4-en? crawler=true (accessed 12 December 2015). Tax effort concerns sngs decisions to increase tax rates, within the limits decided by ­national government.

124 Table 6.1

Ambrosanio, Balduzzi and Peiti Subnational government revenues as a percentage of general government revenuesa

Country

1995

2005

2013

Austria

13.3 (3.7) 13.3 (6.0) 54.1 (42.3) 32 25.4 13.7 33.2 (22.2) 10.5 9.8 8 11.3 8.7 12 9.2 6.1 10.2 17.1 (5.6) 32.3 48.1 (27.7) 7.9

10.4 (3.5) 16.6 (9.7) 53.4 (43) 31.3 25.3 15.0 34.4 (23.4) 14.5 8.1 8.4 19.6 6.6 11.3 9.9 5.5 10.2 25.5 (14.1) 34.8 48.2 (29.5) 9.1

9.3 (2.8) 17.5 (10.7) 54.6 (43.7) 27.2 29.1 15.5 35.6 (24.4) 9.8 6.5 9.5 18.8 5.5 9.4 11.0 5.8 13.1 26.9 (14.2) 34.6 48.1 (29.5) 9.0

Belgium Canada Denmark Finland France Germany Hungary Ireland Israel Italy Luxembourg Netherlands Portugal Slovakia Slovenia Spain Sweden Switzerland United Kingdom

a Excluding transfers paid to other levels of government. Figures in parentheses represent state revenues for federal countries. The data for the Netherlands is from 2013. Source: Our calculations are based on the oecd Fiscal Decentralization Database and imf Government Finance Statistics.

5.8 per cent in Slovakia to 26.9 per cent in Spain to 54.6 per cent in Canada. Focusing on federal countries, excluding Austria, most of the subnational revenues are assigned to states. Table 6.2 focuses on subnational tax revenues as a percentage of total ­general government tax revenues. At state level, almost 40 per cent of general

125

Rules, Practices, and Challenges

Table 6.2 Subnational tax revenue as a percentage of total general government tax revenuesa

Country

1995

2005

2013

1.6 5.1 37.9 21.4 14.1 24.8

1.6 5.5 39.9 21.9 13.8 24.6

3.3 4.8 9.2 33.2 20.7 11.5 7.8 6.3 2.3 7.1 16.6 4.4 3.9 6.8 2.7 7.4 8.7 32.2 15.8 4.7

3.2 4.8 9.4 26.5 23.4 12.9 8.2 5.8 3.5 7.4 16.3 3.5 3.7 7.0 2.9 11.0 10.1 37.1 15.2 4.9

State Austria Belgium Canada Germany Spain Switzerland

1.8 1.8 37.1 21.6 4.8 23.8 Local

Austria Belgium Canada Denmark Finland France Germany Hungary Ireland Israel Italy Luxembourg Netherlands Portugal Slovakia Slovenia Spain Sweden Switzerland United Kingdom

4.1 4.8 9.8 31.9 22.3 11.0 7.4 2.5 2.7 5.9 5.4 6.4 2.7 4.2 1.3 6.3 8.5 30.9 17.6 3.7

a The data for the Netherlands is from 2012. Source: Our calculations are based on the oecd Fiscal Decentralization Database and imf Government Finance Statistics.

126

Ambrosanio, Balduzzi and Peiti

government tax revenues in Canada in 2013 was assigned to provinces, whereas this was the case for less than 2 per cent of such revenues in Austria. Apart from Spain, no particular change occurred in the period considered. A great heterogeneity also characterises the local level in both unitary and federal countries: in 2013, ratios ranged from 3.2 per cent in Austria to 37.1 per cent in Sweden. The most relevant change occurred in Italy, where the share of local taxes increased from 5.4 per cent in 1995 to 16.3 per cent in 2013. The composition of subnational revenues—taxes, grants, user fees and other revenue—is even more interesting. Table 6.3 shows, yet again, a very heterogeneous picture. Looking at 2013, the share of local taxes assigned to states in federal countries varied from 7.6 per cent in Austria to 62.6 per cent in Germany; the share of taxes assigned to local governments varied from 9.1 per cent in the Netherlands to 63 per cent in Sweden. Thus, we can see an impressive variability across countries in the share of grants. In 2013, this amounted, on average, to about 44 per cent of local revenues, but it ranged from 15.2 per cent in Germany to 84.4 per cent in Austria for the state level, and from 11.2 per cent in Switzerland to 70 per cent in the Netherlands for local governments. If we look at the time trend, two countries stand out: in Italy, the share of grants decreased from 60.5 per cent in 2005 to 39.3 per cent in 2013, while, in Slovakia, this share increased from 11 per cent to 66.6 per cent. As regards user fees, they are commonly defined as public prices and service charges, meaning that they basically correspond to prices paid by individuals for specific services provided by the subnational government on a private on-demand basis. In Table 6.3, these are measured as the sum of two items: “Market output and output for one’s own final use” and “payments for nonmarket output” (following the oecd). Apart from some exceptions (Austria, the Netherlands and Slovakia), user fees represented the smallest source of financing for both levels of sngs in 2013. Nonetheless, greater use would be appropriate for at least two reasons: first, they are “naturally” own revenues, as their level and use is fully decided at the corresponding level of government. In addition, as highlighted by Bird, they could promote economic efficiency by providing demand information to public sector suppliers and to ensure that what the subnational public sector supplies is valued at least at (marginal) cost by citizens.17

17

R.M. Bird, “Threading the Fiscal Labyrinth: Some Issues in Fiscal Decentralization”, National Tax Journal, 46 (1993) 207–227, at 212.

127

Rules, Practices, and Challenges

Table 6.3 Subnational governments’ revenues as a percentage of total subnational revenuesa

Country

Taxes 1995

Grants

User fees

Other

2013

1995

2013

1995

2013

1995

2013

7.6 15.6 60.3 62.6 33.2 50.5

78.2 77.8 18.2 16.8 77.4 24.4

84.4 64.6 18.6 15.2 60.2 24.9

7.4 7.7 7.1 5.9 5.2 18.1

7.5 8.2 9.3 9.6 6.3 18.7

5.5 8.3 15.6 12.3 1.3 5.8

0.6 11.6 11.8 12.6 0.3 5.9

15.9 30.9 36.5 35.5 44.5 50.5 39.0 22.4 23.8 39.3 47.0 25.5 9.1 33.6 13.0 42.4 52.2 63.0 58.6 14.1

52.6 46.9 46.3 46.2 33.4 27.8 35.7 62.4 70.0 42.4 60.5 40.8 74.4 31.0 11.0 42.4 34.8 20.8 11.0 72.8

61.9 49.7 47.3 58.2 30.3 28.5 36.1 53.7 45.0 37.5 39.3 54.9 70.0 28.1 66.6 37.9 34.3 28.7 11.2 68.6

27.9 6.6 11.1 5.6 15.7 13.4 20.9 11.2 13.0 5.5 10.2 20.1 10.8 21.2 12.6 15.7 10.2 15.9 20.5 9.3

15.2 7.2 12.7 4.7 22.5 15.7 17.3 9.2 23.7 3.8 10.9 17.8 15.1 15.6 17.6 14.9 10.6 8.3 24.0 12.8

2.3 15.2 5.7 1.3 4.5 9.3 9.0 18.5 10.3 20.4 13.0 1.6 9.9 23.7 59.8 10.7 8.2 3.2 7.3 6.8

7.0 12.1 3.5 1.7 2.7 5.4 7.6 14.7 7.5 19.3 2.7 1.7 5.9 22.7 2.9 4.7 2.9 0.1 6.2 4.4

State Austria Belgium Canada Germany Spain Switzerland

8.9 6.1 59.1 65.0 16.1 51.7 Local

Austria Belgium Canada Denmark Finland France Germany Hungary Ireland Israel Italy Luxembourg Netherlands Portugal Slovakia Slovenia Spain Sweden Switzerland United Kingdom

17.2 31.3 37.0 46.9 46.4 49.5 34.4 8.0 6.7 31.8 16.3 37.4 4.9 24.0 16.6 31.2 46.8 60.1 61.3 11.1

a The data for the Netherlands is from 2012. Source: Our calculations are based on the oecd Fiscal Decentralization Database and imf Government Finance Statistics.

128

Ambrosanio, Balduzzi and Peiti

Likewise, the Council of Europe argues that “recourse to fees and charges can make a considerable contribution to subnational authority revenues”, provided that access to essential services for all sections of the population is ensured.18 Going back to grants, their weight is of great importance because it is one of the main indicators of the so-called vertical fiscal gap. Or should we say vertical fiscal imbalance? There is no general consensus on the specific definition of the vertical imbalance. Boadway and Shah, for example, distinguish between vertical fiscal gap and vertical fiscal imbalance: the former is defined as a revenue deficiency arising from a mismatch between means and ­expenditures, typically of lower orders of governments; the latter occurs when a vertical fiscal gap is not adequately addressed by the reassignment of responsibilities or by fiscal transfers and other means.19 We refer to a fiscal gap as the gap between sngs’ own revenues and expenditures. The most frequently used measures of vertical fiscal gaps are the share of each subnational expenditure financed by sngs’ own revenues (Table 6.4) or financed by grants (namely, the transfer dependency ratio). At the state level, the most relevant change concerns Spain, where the share of expenditure financed by own revenues increased from 28.8 per cent in 1995 to 44.5 per cent in 2013. For other countries, no significant changes occurred. If we consider local governments, the average share of expenditures financed by local governments’ own revenues was quite stable in this period, showing a slight decrease from 56 per cent in 1995 to 53.2 per cent in 2013, though the trends in individual countries varied widely. For example, the vertical fiscal gap increased in Austria, Denmark, Luxembourg, Hungary, Slovakia and Spain, while it decreased in Finland, Ireland, Italy and the Netherlands. As is well known, a vertical imbalance arises when the decentralisation of spending responsibilities is not matched with an adequate decentralisation of revenue autonomy, leaving room for transfers from the central government and for borrowing. As argued by Eyraud and Lusinyan, some degree of fiscal imbalance is inevitable, however, since the “degree of spending decentralization called for by efficiency considerations tends to exceed the degree of tax 18

19

Council of Europe, “Recommendation Rec(2005)1 of the Committee of Ministers to Member States of the Financial Resources of Local and Regional Authorities”, adopted by the Committee of Ministers on 19 January 2005 at the 912th Meeting of the Ministers’ Deputies, p. 12. R. Boadway and A. Shah (eds), Intergovernmental Fiscal Transfers: Principles and Practice (Washington, dc: World Bank, 2007), https://openknowledge.worldbank.org/handle/ 10986/7171 (accessed 10 December 2016).

129

Rules, Practices, and Challenges Table 6.4 Vertical fiscal gapa

Country

1995

2005

2013

13.0 18.3 76.5 82.1 42.2 65.6

7.8 24.1 80.6 80.7 44.5 74.1

35.4 50.8 55.8 52.2 71.9 71.0 64.2 51.5 42.6 55.9 60.0 55.8 33.8 67.2 28.2 52.4 59.1 78.9 90.9 32.1

31.8 51.6 55.9 39.9 70.2 72.1 61.3 27.3 54.7 58.9 59.5 39.1 30.8 70.7 31.1 62.4 55.2 71.1 87.8 32.8

State Austria Belgium Canada Germany Spain Switzerland

13.0 8.6 79.2 80.3 28.8 74.5 Local

Austria Belgium Canada Denmark Finland France Germany Hungary Ireland Israel Italy Luxembourg Netherlands Portugal Slovakia Slovenia Spain Sweden Switzerland United Kingdom

47.0 51.3 53.9 51.7 61.7 72.2 63.5 36.9 30.4 58.3 37.9 52.4 25.5 68.4 94.6 56.1 59.8 79.4 89.2 30.0

a The data for the Netherlands is from 2012. Source: Our calculations are based on the oecd Fiscal Decentralization Database and imf Government Finance Statistics.

130

Ambrosanio, Balduzzi and Peiti

autonomy that would be consistent with optimal tax assignment” (considering the problems connected to a high degree of tax autonomy).20 Furthermore, the presence of a vertical fiscal imbalance allows the central government to control, to some extent, the level of subnational expenditure by changing the amount of grants provided (as recently happened, for instance, in Italy, where grants were cut during fiscal consolidation21), as well as the quality and composition of subnational expenditure, for example, through conditional and/ or matching grants. Nonetheless, most of the literature, also on empirical grounds, seems to make a case for reducing vertical fiscal imbalances. The first option for narrowing the fiscal gap is the further decentralisation of the taxing power, trying to minimise the drawbacks discussed above (tax base mobility, higher administrative costs and horizontal disparities). A second option consists in recentralisation on the expenditure side, a process that conflicts with the well-consolidated literature on the merits of the subsidiarity principle. This topic has been receiving a great deal of interest of late in different contexts. For instance, Dickovick argues, on the basis of the experience of several developing countries, that decentralisation should not be viewed merely as a one-way process but that room should be left for recentralisation.22 A natural third way to address the negative consequences of relevant fiscal gaps consists in designing improved architecture for subnational financing through a more efficient allocation and combination of tax revenues, grants and fees. 4

The Role of Shared Taxes: Theoretical Issues

But what does all of the above data actually tell us about the taxing power of sngs and thus about the “effective” fiscal gap? Measuring the tax autonomy of sngs is still an open question in economics literature. Traditional measures, such as those presented above, while easy to manage or build, are still rather unsatisfactory, as they tell us nothing about the real power of sngs to change the amount of resources they control (namely, their tax base or tax rates). 20

21

22

L. Eyraud and L. Lusinyan, “Decentralizing Spending More than Revenue: Does It Hurt Fiscal Performance?” imf Working Paper, 226 (2011), https://www.imf.org/external/pubs/ ft/wp/2011/wp11226.pdf (accessed 5 December 2015), at 5. M.F. Ambrosanio et al., “Economic Crisis and Fiscal Federalism in Italy”, in E. Ahmad et al. (eds), Multi-level Finance and the Euro Crisis: Causes and Effects (Cheltenham: Edward Elgar Publishing, 2016) 212–248. J.T. Dickovick, Decentralization and Recentralization in the Developing World: Comparative Studies from Africa and Latin America (University Park: The Pennsylvania State University Press, 2011).

Rules, Practices, and Challenges

131

Starting from the 1990s, some international institutions, such as the imf, oecd, World Bank and the European Union, began addressing this relevant issue both on methodological and on empirical grounds. The results of empirical studies are often conflicting, given the inherent complexity in the construction of synthetic indicators of tax autonomy.23 Such indicators require two elements: a homogeneous dataset containing detailed information about the different types of revenues of all levels of governments and a set of qualitative information aimed at identifying the revenues for which subnational authorities are free to determine either the tax rates or the tax base or both. In other words, the core issue is to correctly identify sngs’ own tax revenues, which are composed of “proper” own taxes and of shared taxes. Sharing, in turn, can be of two different types: base sharing and revenue sharing.24 “Proper” own taxes—to the extreme; independent legislation and ­administration—allow subnational governments to enact their own tax laws independently of higher levels of government; each jurisdiction chooses which taxes to levy, the determination of its tax base and tax rates, and it is responsible for tax administration and enforcement. But more frequently, the taxing power of subnational governments is limited by national legislation. For instance, tax rates may be varied within a predetermined range, or the freedom to influence the tax base consists only in having partial decision-making power about allowances and reliefs.25 Base sharing is a form of concurrent taxation of the same tax base in which shared resources are raised by one authority in addition to those raised by ­another authority on the same basis.26 Usually, they take the form of surcharges, where both the central government and sngs levy the same tax. sngs impose surcharges on the tax base specified by the central government. In this case, the taxing power of subnational jurisdictions is less than it would be if they were to collect “proper” own taxes and, instead, can be found in their choice of tax rates—which often have to be within the limits determined by the central government—in respect of their share of the total tax base. 23

J. Rodden, “Comparative Federalism and Decentralization: On Meaning and Measurement”, Comparative Politics, 36 (2004) 481–500. 24 On the significance and classification of tax sharing, see J. Kim, “The Role of ­Intergovernmental Fiscal Institutions: The Case of Tax Sharing”, in J. Kim and H. Blöchliger (eds), Institutions of Intergovernmental Fiscal Relations: Challenges Ahead (Paris: oecd Publishing, 2015) 11–26. 25 For a more detailed classification, see Ambrosanio and Bordignon, “Normative versus Positive Theories”, supra. 26 Council of Europe, “Recommendation Rec(2005)1”, supra.

132

Ambrosanio, Balduzzi and Peiti

Finally, revenue sharing may assume two different forms: sngs are entitled either to a fraction of particular tax revenues arising in their jurisdiction or to a given percentage of nation-wide tax receipts. In both cases, autonomy is limited to the possibility of bargaining sharing quotas or formulas with the national government. Hence, we can consider revenue sharing as just a particular form of grants. As stressed by Musgrave and Polinsky more than 40 years ago: There is no sharp distinction between revenue sharing and grants. Revenue sharing, after all, involves the making of grants, and grants involve the sharing of revenue.27 Revenue sharing does, however, have certain benefits, such as transparency, administrative ease and simplicity. On the other hand, base sharing is less transparent, and if sngs have important fiscal power, citizens may find it difficult to understand the role of different levels of government and this, in turn, may reduce sngs’ accountability.28 Both types of tax sharing involve disadvantages in the form of horizontal (working among governments of the same level) and vertical externalities (between different levels of government) that could lead to a distorted tax structure or to the distorted localisation of individuals and firms among jurisdictions. Vertical externalities refer to the fact that the tax decisions of a certain level of government influence the revenues of other levels of government, i.e. the tax base of each level of government potentially depends on the tax rates set by the other levels of government. Hence, we can observe both downward and upward vertical externalities (a downward externality means that a decision taken by an upper level of government affects the revenues of lower levels of government). Vertical externalities are different under base sharing and under revenue sharing. With revenue sharing, only downward externalities seem possible. Revenue-sharing arrangements are subject to unilateral and arbitrary change by the central government: when the central government varies the tax base or the tax rate of a revenue-shared tax, this decision automatically 27

28

R.A. Musgrave and A.M. Polinsky, “Revenue-Sharing: A Critical View” in Financing State and Local Governments: Proceedings of the Monetary Conference (Nantucket Island: Federal Reserve Bank of Boston, 1970) 17–52, at 24. I. Joumard and P.M. Kongsrud, “Fiscal Relations across Government Levels” oecd Economics DepartmentWorking Papers, 375 (2003), http://www.oecd-ilibrary.org /economics/fiscal-relations-across-government-levels_455513871742?crawler=true (accessed 7 December 2015).

Rules, Practices, and Challenges

133

changes, ceteris paribus, the revenues of the subnational government as well. In this case, the central government may or may not “sterilise” the increase or decrease in the subnational government’s revenue so as to leave the amount of resources assigned unchanged. In many countries—generally in federal countries—it is the constitution that provides the criteria for the allocation of revenues to different levels of government and also indicates the distribution of these revenues among jurisdictions of the same level. If the percentage of revenue sharing of national taxes is established at the constitutional level, sngs are better protected from the risk of adverse decisions taken at the central level. With base sharing, both downward and upward externalities may occur. Horizontal externalities arise when different governments of the same level tax the same base and when this base is mobile across the different levels of government. Such externalities are the result of a process of tax competition (in the traditional sense of the term), where the tax policy pursued by a subnational government affects the tax policy of its neighbouring jurisdictions, leading, according to most theoretical and empirical literature, to excessively low tax rates (a race to the bottom), under benevolent policymakers who want to maximise the welfare of their citizens.29 An increase in tax rates in a state/ jurisdiction produces revenue gains for other states/jurisdictions due to the outward movements of mobile factors and individuals; in equilibrium, tax rates will be lower than the optimal level. While the literature on vertical externalities is more recent, it is receiving growing interest and attention, also on empirical grounds. Vertical externalities, or vertical tax competition, tend to raise tax rates: each level of government neglects the adverse effect it has on the other levels by raising its tax rate and thereby causing the common tax base to contract.30 Thus, horizontal and vertical fiscal externalities interact, and vertical tax competition tends to partially offset the effects of horizontal tax competition. But which of the two is dominant? And under which conditions? As stressed 29

30

R.H. Gordon, “An Optimal Taxation Approach to Fiscal Federalism”, Quarterly Journal of Economics, 98 (1983) 567–586; G.R. Zodrow and P. Mieszkowski, “Pigou, Tiebout, Property Taxation, and the Under-provision of Local Public Goods”, Journal of Urban Economics, 19 (1986) 356–370; D.E. Wildasin, “Nash Equilibria in Models of Fiscal Competition”, Journal of Public Economics, 35 (1988) 229–240; D.E. Wildasin, “Interjurisdictional Capital Mobility: Fiscal Externality and a Corrective Subsidy”, Journal of Urban Economics, 25 (1989) 193–212. M. Keen, “Vertical Tax Externalities in the Theory of Fiscal Federalism”, imf Staff Papers, 45 (1998), 454–485; M. Keen and C. Kotsogiannis, “Does Federalism Lead to Excessively High Taxes?” American Economic Review, 92 (2002), 363–370.

134

Ambrosanio, Balduzzi and Peiti

by Brülhart and Jametti,31 the net outcome of these competing effects is theoretically unclear within benevolent32 federal government systems, and this is currently receiving growing interest, as, all around the world, fiscal policy responsibilities are becoming increasingly vertically fragmented because of both the tendency to delegate tax policy from central governments to subnational authorities and the tendency to circumscribe central governments’ independence in fiscal matters “from above”, by means of international treaties and institutions. To confirm the ambiguity of the theoretical predictions, the results of empirical studies also leave open the question of whether horizontal or vertical tax externalities are dominant in a decentralised society with a benevolent government. A review of the relevant recent empirical literature on this issue is outside the scope of this chapter; but it should be stressed that the overall effect of horizontal and vertical tax externalities depends not only on the tax mix and the elasticity of the shared tax bases to their tax rates ,33 but also on the country-specific architecture of the decentralised system—­ including the extent of sngs’ taxing power34—and on the “hierarchical” relationship between the central government and lower levels of government. Finally, it is essential for an empirical analysis of the vertical tax externality hypothesis that “the researcher has to check the robustness of the results against other competing explanations of observed fiscal interdependence among layers of government”.35 31 32

33 34

35

M. Brülhart and M. Jametti, “Vertical versus Horizontal Tax Externalities: an Empirical Test”, Journal of Public Economics, 90 (2006) 2027–2062. Keen and Kotsogiannis argue that the net outcome depends theoretically on the government’s objectives: if governments behave like revenue-maximising leviathans, the vertical externality will be dominant, and the tax rates of different hierarchical government levels will be sub-optimally high, M. Keen and C. Kotsogiannis, “Leviathan and Capital Tax Competition in Federations”, Journal of Public Economic Theory, 5 (2003) 177–199. Blöchliger and Pinero Campos, “Tax Competition”, supra. As stressed in Joumard and Kongsrud, limited tax autonomy can serve to prevent subnational governments setting personal income tax “too high”, Joumard and Kongsrud, ­“Fiscal Relations across Government Levels”, supra. They refer to the case of Denmark and Sweden: to avoid an upward drift in personal income tax rates, in 1996 the Swedish central government introduced “a tax on local government tax”, for any municipality increasing its tax rate (which was subsequently abolished in 1998); in Denmark, a “tax freeze” was instituted in 2002 as an agreement between the central government and the County Council Association regarding their tax levels. A. Esteller-More and A. Sole-Olle, “Vertical Income Tax Externalities and Fiscal Interdependence: Evidence from the us”, Regional Science and Urban Economics, 31 (2001), 247–272.

Rules, Practices, and Challenges

135

The role of shared revenue may also have an impact on economic activity. In more general terms, the relationship between fiscal decentralisation and growth is all but clear. The data shows a correlation between the level of income and the level of fiscal federalism; nonetheless, no causality has yet been demonstrated. Several reviews on this topic have highlighted a substantial ambiguity both in the theoretical research and in the empirical evidence.36 On the theoretical side, research has mainly exploited the traditional advantages of fiscal federalism: decentralisation leads to greater efficiency, more accountability and stimulates positive tax competition, all of which positively contribute to economic growth. On the other hand, in the presence of soft budget constraints or coordination problems, federations may in fact create problems for macroeconomic and stabilisation policies. Probably, the only certain result is that, if a connection between federalism and economic activity really exists, this has not yet been determined. As regards empirical research, there are two main issues. On the one hand, the question is clearly about the impact of decentralisation on growth. In oecd countries, decentralisation is positively correlated to the level of income; nonetheless, it is also negatively correlated to the gdp growth rate.37 When studying causality, relationships become much less clear, and evidence of positive, negative and null effects are all well documented in the literature.38 An additional and useful distinction is possible between expenditure decentralisation and revenue decentralisation, highlighting the positive effects especially of the latter and that a closer match (though not necessarily a perfect one) between spending and revenue decentralisation is desirable. On the other hand, and more importantly, the problem is to correctly define and measure decentralisation itself. As already noted by Rodden, traditional measures of fiscal decentralisation may fail to capture the real power of subnational authorities,39 and this is true for both expenditure decentralisation and for revenue decentralisation, which is more relevant for this chapter. It is thus clear that the presence of shared revenues without any real taxing power 36

37 38 39

T. Baskaran et al., “Fiscal Federalism, Decentralization and Economics Growth: Survey And Meta-Analysis”, CESifo Working Paper, 4985 (2014); H. Blöchliger, “Decentralisation and Economic Growth—Part 1: How Fiscal Federalism Affects Long-Term Development”, oecd Working Papers on Fiscal Federalism, 14 (2013), http://dx.doi.org/10.1787/5k4559gx1q8r-en (accessed 12 December 2015). Blöchliger, “Decentralisation and Economic Growth”, supra. N. Gemmell et al., “Fiscal Decentralization and Economic Growth: Spending versus Revenue Decentralization”, Economic Inquiry, 51 (2013) 1915–1931. Rodden, “Comparative Federalism and Decentralization”, supra.

136

Ambrosanio, Balduzzi and Peiti

should be taken into account. Despite the fact that some efforts have already been devoted to resolving this issue, ambiguity still prevails.40 In order to avoid design problems and to exploit most of the benefits of decentralisation and financial autonomy, it is thus crucial to correctly specify the rules for tax assignment. Which taxes are best suited for sngs? Following Bird and the Council of Europe, subnational taxes should have some desirable features.41 The tax base should be relatively immobile so subnational authorities can vary rates without risking the loss of too much of their tax base; the tax yield should be adequate to finance subnational expenditures, relatively stable and predictable over time to support sound subnational fiscal practices; subnational taxes should have a certain degree of flexibility so that tax revenue can be adjusted to changing budget costs; subnational taxes should minimise economic distortion, as well as demographic and social distortions; to strengthen subnational accountability, it should not be possible to export the tax burden to non-residents,­and the tax base should be visible to taxpayers, also in relation to the level of services provided; the tax should be relatively easy to administer so that the cost of administration is adequate in respect of the revenue collected. None of the taxes used to finance sngs possess all of the above-mentioned merits.42 Personal income taxes, in most cases in the form of a surtax on the national income tax base, may cause horizontal imbalance and require equalisation schemes, since the income tax base is generally not evenly distributed across jurisdictions. It may also produce horizontal tax competition in the presence of different subnational tax rates and mobility of individuals, as well as vertical fiscal externalities. Profit taxes are generally not considered a good source of revenues for subnational governments for several reasons: if applied at different rates, they may distort the allocation of economic activity (as in an international context) and produce tax exporting; they are difficult to administer and are affected by cyclical fluctuations. In addition, value-added taxes are often considered an inappropriate source of revenue for subnational governments because of their effects on trade between different jurisdictions, problems of tax fraud and tax exporting, high administrative and compliance costs, but vat 40 Baskaran et al., “Fiscal Federalism”, supra; J. Thornton, “Fiscal Decentralization and ­Economic Growth Reconsidered”, Journal of Urban Economics, 61 (2007) 64–70. 41 R.M. Bird, “Subnational Taxation in Developing Countries: a Review Of The Literature”, Journal of International Commerce, Economics and Policy, 2 (2011) 139–161; Council of ­Europe, “Recommendation Rec(2005)1”, supra, at 10. 42 Ambrosanio and Bordignon, “Normative versus Positive Theories”, supra.

Rules, Practices, and Challenges

137

is usually only applied by the central government, which determines tax bases and tax rates, and subnational governments receive a share of the national revenue.43 Specific taxes on goods and services, which have the merit of administrative ease and do not produce significant distortionary effects (when applied according to the destination principle), do not provide sufficient resources to finance a great amount of subnational public expenditure. Finally, property taxes, according to most literature, represent one of the best tax instruments for assigning a real taxing power to subnational governments, as the tax base is immobile, tax revenue is stable, and there are few problems of tax avoidance. Property taxes may, however, cause tax exporting problems if they rely on assets owned by non-residents. But what happens around the world? The tax assignment prevailing in almost all countries conflicts with the desirable features for subnational taxation as illustrated above. Table 6.5 shows very different approaches to tax assignment in federal and unitary countries. sngs are assigned income and profit taxes, as well as consumption and property taxes. If we consider federal countries, tax revenues at the state level in 2012 came, for the most part, from income and profit taxes in Austria, Canada, Germany, Spain and Switzerland. On the contrary, in Belgium, about 70 per cent of tax revenue came from property tax. Only in Canada, Germany and Spain were states assigned general consumption taxes, whereas revenues from taxes on specific goods and services varied from 1.2 per cent in Germany to 16.4 per cent in Spain. In Austria, the weight of payroll taxes and social contributions is particularly relevant, in Belgium the weight of taxes on uses of goods. The picture did not fundamentally change between 1995 and 2012 except in Spain, where there was a shift from property and specific consumption taxes to income and general consumption taxes. As for local governments, both in unitary and federal countries, the share of income and profit taxes in 2012 was above 75 per cent in Scandinavian countries and in Germany, Luxemburg, Slovenia, and Switzerland, with the share of property taxes varying from 2.7 per cent in Sweden to above 90 per cent in 43

There is, however, a growing body of literature according to which vat could become a good source of financing for local governments, C.E. McLure Jr., “Implementing Subnational Value Added Taxes on Internal Trade: The Compensating vat (cvat)”, International Tax and Public Finance, 7 (2000), 723–740; R.M. Bird and P. Gendron, “cvat, vivat, and Dual vat: Vertical ‘Sharing’ and Interstate Trade”, International Tax and Public Finance, 7 (2000) 753–761; M. Keen, “vivat, cvat, and All That: New Forms of Value-Added Tax for Federal Systems”, imf Working Paper, 83 (2000), https://www.imf.org/external/pubs/ ft/wp/2000/wp0083.pdf (accessed 7 December 2016).

138

Ambrosanio, Balduzzi and Peiti

Table 6.5 Composition of tax revenues

Country

Income and profits

Property

General consumption

Specific goods and services

Othersa

1995 2012

1995

1995

2012

1995

2012

1995

2012

2012

State Austria Belgium Canada Germany Spain Switzerland

42.3 4.7 48.7 51.9 8.5 78.3

46.7 – 45.0 53.5 41.4 78.9

2.4 70.4 6.0 6.1 55.9 14.7

2.1 – 71.4 – 3.5 19.9 5.4 35.4 7.7 9.2 13.2 –

– – 21.7 39.9 33.6 –

12.2 3.5 15.5 1.8 23.6 1.0

8.4 1.6 14.1 1.2 16.4 2.6

43.1 21.4 9.9 4.7 2.9 5.9

42.9 27.1 15.7 – 1.0 5.3

0.0 33.7 – 89.3 93.3 0.1 79.3 – – – 24.9 91.3 – 29.5 – 78.5 20.5 97.3 84.8 –

13.2 14.8 – 18.8 58.0 – 85.7 97.4 0.1 6.4 10.5 – 4.5 6.6 – 47.8 51.6 – 19.3 14.7 – 28.9 20.1 63.5 87.9 93.8 – 93.5 94.8 – 38.6 16.0 – 5.9 6.8 – 69.2 52.4 – 29.1 45.1 29.7 83.4 52.1 – 20.7 15.3 – 29.6 39.0 14.8 – 2.7 – 13.3 13.7 – 100.0 100.0 –

– – 0.2 – – – 4.7 68.3 – – 5.6 – – 11.8 – – 15.4 – – –

19.5 6.8 – 0.1 – 7.4 0.5 1.3 – – 7.3 0.8 1.6 11.6 10.3 2.8 4.1 0.3 0.7 –

3.9 6.5 0.1 0.1 – 20.7 0.8 1.1 – – 9.6 1.2 1.8 6.7 1.8 3.5 6.1 – 1.3 –

65.7 2.7 14.1 – 0.1 44.8 0.6 6.3 12.1 6.5 26.5 0.4 29.2 1.4 6.3 0.8 31.6 – 0.2 –

81.3 1.7 2.4 0.1 0.1 27.6 0.5 10.5 6.2 5.2 43.9 0.6 45.9 6.9 46.1 2.8 18.9 – 0.2 –

Local Austria Belgium Canada Denmark Finland France Germany Hungary Ireland Israel Italy Luxembourg Netherlands Portugal Slovakia Slovenia Spain Sweden Switzerland United Kingdom

1.5 71.7 – 93.5 95.4 – 79.5 – – – 27.6 92.8 – 28.2 – 75.8 19.8 99.7 85.8 –

a Other taxes include taxes on use of goods, social security contributions, payroll and other taxes. Our calculations are based on oecd revenue statistics from various years.

Rules, Practices, and Challenges

139

Canada, Ireland and Israel, and all the way up to 100 per cent in the United Kingdom. General consumption taxes played a relevant role only in Hungary. As regards the final category (“others”), this consisted of different taxes for each country. In Austria, for instance, other taxes mainly included payroll taxes; in Italy, the regional tax on productive activity; in France, the Contribution Economique Territoriale (cet) on business. The composition of local taxes significantly changed between 1995 and 2012 in Italy, where the property tax was reformed; in Belgium, with a shift from income taxes to property taxes; and in Slovakia, where the relative importance of property taxes decreased. But, most important, the data in Table 6.5 yet again tells us very little about the tax autonomy of sngs. Tax revenues, in fact, may refer to own taxes or to shared taxes in the form of base sharing or revenue sharing. 5

Shared Taxes and Tax Autonomy in Practice

Comparing the degree of tax autonomy across countries is a complex task that requires the identification of subnational autonomy and discretion over revenue arrangements. A number of studies have attempted to quantify tax autonomy, but, in several cases (cross-country analyses), they used an imperfect measure: the subnational share of total government revenue. This choice was strongly driven by the availability of data. More specifically, if data on the revenue structure of sngs is fundamental to describing the subnational government’s discretion over fiscal resources, difficulties in collecting such data in a uniform way limits the analysis. This is especially true for tax sharing, whose criteria for classification differ slightly across manuals, leading different databases to treat the same tax arrangements differently.44 The Government Finance Statistics (gfs) issued by the imf is one of the most used datasets for cross-country analyses on fiscal decentralisation. However, although the gfs has consistent definitions of the main fiscal variables across countries and over time, the need for standardisation inevitably eliminates details about the design of tax systems. As a consequence, as has been argued by Ebel and Yilmaz, although the gfs distinguishes among four groups of revenues (tax revenues, grants, other revenues and social contributions), the dataset does not include information on whether revenues are collected 44

Blöchliger and Petzold, “Finding the Dividing Line”, supra.

140

Ambrosanio, Balduzzi and Peiti

through shared taxes or locally determined taxes.45 Thus, this limits the use of the gfs dataset to measure tax autonomy. Another important dataset used by researchers is the Revenue Statistics database issued by the oecd, which contains subnational tax revenues reported by oecd member countries, detailed by types of tax revenue. Nevertheless, despite the detailed information on revenue composition and the recommendations on tax sharing, as argued by Kim, the definition of subnational tax varies widely from country to country, and this implies that using subnational tax revenues reported in the Revenue Statistics database require a certain degree of caution.46 The oecd network on fiscal federalism has made a considerable effort to resolve this issue, making more information available for a set of countries, thanks to a survey conducted by the oecd in 2001 and updated in 2005, 2008 and 2011. In this survey, subnational taxes categorised as five different types, identified by the letters (a) to (e), are ranked by the degree of discretion granted to sngs to manage revenues: (a) discretion on rates and relief with or without consulting a higher level of government; (b) discretion on rates with or without upper and/or lower limits set by a higher level of government; (c) discretion on relief, tax allowances and/or tax credits; (d) tax-sharing arrangements; (e) rates and relief set by the central government. A residual category (f) was used for revenues not allocated in previous categories. Tax-sharing arrangements are, in turn, classified according to four subcategories to capture the various rules for determining the sharing formula: a tax-sharing arrangement in which subcentral governments determine the revenue split (d.1); a tax-sharing arrangement in which the revenue split can be changed only with the consent of subcentral governments (d.2); a tax-sharing arrangement in which the revenue split is determined in legislation, and where it may be changed unilaterally by a higher-level government, but less frequently than once a year (d.3); a tax-sharing arrangement in which the revenue split is determined annually by a higher-level government (d.4). The (d)-categorisation highlights the difficulties in collecting uniform information in the various available datasets.47 45

46 47

R.D. Ebel and S. Yilmaz, “On the Measurement and Impact of Fiscal Decentralization”, in J. Martin Vazquez and J. Alm (eds), Public Finance in Developing and Transitional Countries: Essays in Honor of Richard Bird (Cheltenham: Edward Elgar, 2003) 101–126. Kim, “The Role of Intergovernmental Fiscal Institutions”, supra. Blöchliger and Petzold propose a guideline consisting of four criteria that have to be cumulatively fulfilled for a tax to be considered tax sharing, H. Blöchliger and O. Petzold, “Finding the Dividing Line”, supra. The criteria relate to the revenue risk that subcentral governments are exposed to (risk sharing), the freedom to use the revenue obtained (unconditionality), the rules and formulas that define the distribution of financial revenue

Rules, Practices, and Challenges

141

In conclusion, to obtain a proper measure of tax autonomy, a deeper analysis of tax sharing should be conducted on each single shared tax and completed with a more comprehensive approach. As argued by Spahn, while the refinement of the revenue metric proposed by the oecd comes close to expressing the degree of subnational tax autonomy, there are still problems to be resolved for assessing tax sharing.48 These problems can be overcome with a more differentiated composite index that could be calculated starting from oecd metrics—(a) to (e)—and assigning different weights to each component according to the importance of the constraint on subnational autonomy. Finally, the Council of Europe defines own resources as those of which subnational governments: “[…] can vary the level, possibly within a predetermined range. These resources may, for example, be fiscal or non-fiscal, exclusive or shared, etc.”49 In addition, it should be stressed that, for the eu, tax sharing is limited to base sharing: “An authority’s ‘shared resources’ are resources that are raised by the authority in addition to resources raised by another authority on the same basis”. On the contrary, the level of transferred resources may not be varied. On the basis of the previous arguments, we propose a criterion for which own resources are just those where an sng can make a fiscal effort, which represents the only way to use taxing power.50 We calculated three indicators of tax autonomy, for 2011, following different approaches in weighting the tax categories from (a) to (f) discussed above. The first index (Index 1) follows the oecd’s methodology, which assigns positive but decreasing weights to all categories except (e). The second index (Index 2) is based on Stegarescu, which assigns weight 1 to categories from “a” to “c” and weight 0 to all other categories.51 The third index (Index 3) is the one we propose, considering that categories from (a) to (c) are characterised by different degrees of autonomy, so we assign weight 1 to (a), 0.8 to (b), 0.7 to (c) and weight 0 to categories from (d) to (f).

48

49 50 51

(formula stability) and the institutional decision-making mechanism defining each subcentral annual share (individual proportionality). The application of these criteria leads to a slightly different classification. P.B. Spahn, “Measuring Decentralization of Public Sector Activities: Conceptual Issues and the Case of Germany”, in Kim et al. (eds), Measuring Fiscal Decentralization: Concepts and Policies (Paris: oecd Publishing, 2013) 88–99. Council of Europe, “Recommendations Rec(2005)1”, supra. Kim, “The Role of Intergovernmental Fiscal Institutions”, supra. D. Stegarescu, “Public Sector Decentralization: Measurement Concepts and Recent International Trends”, zew Discussion Paper, 04-74 (2004).

142

Ambrosanio, Balduzzi and Peiti

Table 6.6 State/Region tax autonomya

Country

Index 1

Index 2

Index 3

Switzerland Canada Spain Italy Germany Austria

100.0 91.3 79.4 56.8 49.6 40.3

100.0 88.9 60.1 47.1 3.1 38.8

100.0 88.9 59.5 37.7 2.5 31.0

a Our calculations are based on the oecd Fiscal Decentralization Database. Belgium was not taken into consideration due to problems with data inconsistency.

We calculated the tax autonomy indexes separately for both the state/­region level and for the local level, because the taxing power—namely the possibility of using fiscal effort is different for the different levels of government in a country. So a unitary (state and local together) indicator would seem to be incorrect, as it might offer a misrepresented view of the picture. Table 6.6 illustrates the indexes of taxing power at the state/regional level. For each country—except Switzerland, where local taxes all fall under category (a)—the level of taxing power decreases from Index 1 to Index 3. The main difference concerns Germany, whose index drastically decreases, since most of the Länder taxes are classified under category (d). Table 6.7 relates to taxing power indicators at the local level. It shows different rankings of tax autonomy under the three indexes (the data is presented in decreasing order following Index 1). Ireland is not considered because all local taxes are classified in category (f). But since 1 January 2015, local authorities have been able to vary local property tax rates in a range of −/+ 15 per cent of the central government’s rate. At the local level, the main difference in the relative position concerns ­Germany, Hungary and Slovenia. As for the state level, the main reason for these differences stems from the fact that the oecd assigns a positive weight to shared revenues (category (d)). It then becomes inevitable to claim that grading countries depends on two main discretional elements: on the one hand, how a country classifies its revenues from shared taxes; and, on the other hand, on the specific weight assigned to this particular category of revenues.

143

Rules, Practices, and Challenges Table 6.7 Local tax autonomya

Country

Index 1

Index 2

Index 3

Italy Belgium Israel Slovakia Switzerland United Kingdom Spain Denmark Canada Luxembourg Netherlands Sweden Finland Hungary Germany France Portugal Slovenia Austria

82.2 81.2 81.0 80.9 80.3 80.0 79.7 79.1 78.9 78.9 78.1 77.9 75.7 72.1 67.1 64.2 61.3 37.6 21.5

93.7 99.8 100.0 100.0 100.0 100.0 81.2 98.1 98.0 97.2 97.3 97.4 91.4 84.2 58.4 63.2 72.9 14.1 23.0

82.2 81.2 81.0 80.9 80.3 80.0 70.7 78.5 78.8 78.7 77.8 77.9 73.1 67.4 46.7 59.5 58.3 14.1 20.0

a Ireland is not considered because all local taxes are classified in category (f). But since 1 January 2015, local authorities have been able to vary local property tax rates in a range of −/+ 15 per cent of the central government's rate. SOURCE: Our calculations are based on the OECD Fiscal Decentralisation Database.

6

Case Studies: Taxing Power and the Vertical Fiscal Gap

An interesting issue concerns the relationship between taxing power and the vertical fiscal gap. A relevant taxing power, as we discussed in previous sections, has the merit of increasing accountability; a relevant vertical fiscal gap has the demerit of weakening accountability. But the two often coexist: this happens when the taxing power refers to a very low share of total subnational revenues. Figure 6.1 illustrates the issue for local governments. Countries are divided into four groups according to their relative position with respect to average

144

Ambrosanio, Balduzzi, and Peiti

H

A

NL FK UK L DK

Mean SLO

D P

F

E CDN FIN

B IL

I

S

CH

Figure 6.1 Taxing power and vertical Fiscal gap, local governments (2011).

v­ alues of taxing power and fiscal gap. In the top left (tl) quadrant are countries with low taxing power and high vertical gap; in the top right (tr) quadrant are countries with high taxing power and high vertical fiscal gap; in the bottom right (br) quadrant are countries with high taxing power and low vertical gap; finally, in the bottom left (bl) quadrant are countries with low taxing power and low vertical gap. Countries in the br quadrant (e.g. Italy, Switzerland and Canada) have a higher level of financing autonomy, whereas countries in the tl quadrant (Austria and Hungary) have the lowest. Countries in the tr quadrant (e.g. uk and Belgium), despite their ability to control their own tax revenues, can only raise a small amount of resources compared to the level of expenditures they face. Finally, countries in the bl quadrant (France and Germany) have a more limited taxing power (for instance, they rely significantly on shared taxes), but they are able to finance a higher share of their expenditures. In the following case studies, we present details concerning one of the countries belonging to each of the four quadrants, starting with Italy. 6.1 Italy: Bottom Right Italy has three levels of subnational government: 20 regions (15 “ordinary” regions and 5 “special” regions, with the latter enjoying a particular status, more autonomy and a different financing system than ordinary regions), 107 provinces and about 8,000 municipalities. Ordinary regions have exclusive legislative power with respect to any matters not expressly reserved to state law, but

Rules, Practices, and Challenges

145

their most important function is healthcare. They finance their expenditures mainly through own taxes and shared taxes. The most relevant proper own tax is the Regional Tax on Productive Activity (irap), introduced in 1998: the basic tax rate is 3.9 per cent (4.65 per cent for the financial sector, 5.9 per cent for the insurance sector, 1.9 per cent for the agricultural sector). Italy’s regions are allowed to vary the rate by a maximum of 0.92 per cent. Regions may even decide to set the rate at 0 per cent, which is conditioned on the fact that they do not finance this policy by increasing another revenue source assigned to them, i.e. through a share of personal income tax. This takes the form of tax-base sharing: regions are obliged to levy a surcharge at a basic rate of 1.23 per cent, with the possibility of varying it by not more than 0.5 per cent with respect to the basic rate. A share of vat, in the form of revenue sharing, is also assigned in respect of specific regional characteristics (e.g. population size). Regional tax revenues are integrated by transfers for the provision of health services and equalisation mechanisms. Likewise, regions and municipalities finance their expenditures through own taxes, shared taxes and transfers. The main own tax is the unique municipal tax—Imposta Unica Comunale (iuc)—which is composed of three different taxes: a property tax (imu), a tax on buildings to finance indivisible services (tasi) and a garbage tax (tari). Municipalities can vary the tax rates within ranges determined by the central government with respect to the imu and tasi; as regards tari, revenues should perfectly match the corresponding expenditures for garbage collection. Municipalities are also assigned a base share of the personal income tax, so they may (on a voluntary basis) apply a surcharge of up to a maximum of 0.8 per cent. Finally, municipalities are entitled to transfers that, since 2015, have been partially assigned (20 per cent) on the basis of the standard costs for the provision of essential and compulsory services. 6.2 France: Bottom Left France has three main tiers of subnational administration: communes, departments and regions. These are all districts in which administrative decisions made at the national level are carried out and also authorities with powers of their own. Excluding overseas territories, there are 22 administrative regions, 94 administrative departments and about 37,000 communes (80 per cent of which have fewer than 1,000 residents). Subnational revenues come from taxes set and raised locally or from shared taxes (about 50 per cent), from grants (about 30 per cent) and from user fees and charges (about 16 per cent). In France, there are four main local taxes: the cet from businesses, whose rates are set by the local government; a property tax on buildings, whose rate is set by departments and municipalities; a residence tax and a property tax

146

Ambrosanio, Balduzzi and Peiti

on n ­ on-built land, whose rates are set by municipalities. The cet is a local tax imposed by the departmental and regional councils on businesses, introduced in January 2010, replacing the taxe professionelle, and is made up of two components: the Cotisation Foncière des Entreprises (cfe), based on the rateable value of the property occupied by the business, and the Cotisation sur la Valeur Ajoutée des Entreprises (cvae), based on the value added each year by the business, that applies only to those businesses with a turnover of greater than 500,000 euros. Local councils have the discretion to set a minimum rate and to decide allowances for businesses operating on a seasonal basis. As for local property tax, owners of residential property in France are liable for two annual local property taxes, called the taxe d’habitation and the taxe foncière. These local taxes fund the services provided through the local commune and inter-communal bodies, while the latter also contributes to the budget of the departmental councils (France’s regions have not participated in this tax since 2011). The first is an annual residence tax imposed on the occupier of a property and on second homes capable of occupation. The second is an annual property ownership tax whether or not the property is occupied or rented out. The tax rate is determined by local councils, but the calculation and collection of the tax is carried out by the central government tax authority. Some local councils also apply a separate charge for garbage collection (taxe d’enlèvement des ordures ménagères), which is paid with the taxe foncière. 6.3 Austria: Top Left Austria is a federal state that is composed of nine autonomous Länder, further divided into counties and municipalities or statutory cities (2,357 municipalities, or Gemeinden). The federal level retains many of the most important powers, including all judicial powers, responsibility for the police and the military, and control of public accounts and the administration of public funds at all levels of government. The Austrian states are responsible for executing key functions such as administering education, social welfare, healthcare and housing. The constitution does not define tax-raising powers; this is left to federal legislation, which cannot be vetoed by the states (Fiscal Adjustment Act). Austrian regional and local governments are part of a financial equalisation mechanism that regulates the division of revenue and expenditure among the different levels of government. Around 90 per cent of the revenues of Austrian states and municipalities come from shared federal government taxes; municipalities levy more than 4 per cent and Länder less than 1 per cent of the total revenue. The first step in the financial equalisation mechanism is the distribution of the tax share for both the Länder and the municipalities (grouped

Rules, Practices, and Challenges

147

­together by the label “Land”) by the federal government to the individual Länder, according to two main criteria: tax revenue (allocation is based on a tax’s regional or local revenue) and demographics (allocation is based on the number of inhabitants of a Land or municipality). The Länder then allocate the local share of revenues to the individual municipalities in accordance with various criteria such as financial requirements, regular demographic criteria and a special demographic criterion (a scaled population multiplier, which is a correction mechanism that favours large municipalities, as they usually have larger financial requirements) and other allocation criteria decided by the distributing Land. Shared taxes include tax on personal income as well as corporate tax, sales tax and financial transaction tax. There are also other joint taxes (gambling, advertising, real estate transfers and land value) of which municipalities receive a higher share. The second step of the financial equalisation mechanism consists of intergovernmental transfers to equalise the average revenues of Länder and municipalities (paid by the federal government) or for special needs or purposes, such as housing development, environmental purposes and infrastructure. Finally, states and municipalities also levy their own taxes. For example, municipalities levy a general payroll tax of 3 per cent on total salaries and wages paid each month by permanent establishments based in Austria and an annual real estate tax of up to 2 per cent of a property’s assessed value. 6.4 United Kingdom: Top Right Each part of the United Kingdom has a distinct system of local government. Local governments (councils) have different responsibilities across uk. In Scotland and Wales, there are only single-tier or unitary councils, whereas England has a mix of single- and two-tier principal authorities. The level of fiscal decentralisation in the uk is rather limited. Local governments are mostly financed by block grants from the central government that can be spent freely by the subnational authorities. Councils are entitled to raise and set a domestic property tax (“council tax”), which provides about 25 per cent of local government revenues. The council tax is the main source of locally raised revenues for many local authorities. It was introduced in 1993 by the Local Government Finance Act adopted the previous year and replaced the so-called “community charge”. It determines the way households contribute to services provided by local authorities. The tax rate is set by local councils annually and charged according to valuation bands. In 2011–2012, the central government introduced a scheme to pay a grant (council tax freeze grant) to local authorities that froze or reduced their

148

Ambrosanio, Balduzzi and Peiti

council tax that year. The freeze grant has continued to be available every year since. In England, for example, 57 per cent of local authorities received the Council Tax Freeze Grant for 2015–2016. Local authorities’ remaining revenues come from charges for the provision of services. 7

Conclusions

The role of own revenues is crucial to defining the level of accountability and decentralisation within a country. This chapter has presented and thoroughly discussed the links between local governments’ accountability and the different sources of sngs’ financing, on both quantitative and qualitative grounds. It has emerged that practices of revenue assignment generally differ greatly from theoretical rules and that traditional measures of decentralisation fail to offer a complete and satisfactory picture of reality. In addition, it has emerged that taxing power indicators are not a sufficient measure of a country’s ability to finance its own expenditures. Hence, taxing power alone cannot be considered a correct proxy for sngs’ accountability. In order to determine the level of a country’s financing power more accurately, it is important to consider the joint role of both taxing power and of fiscal gaps. To illustrate this issue, we divided a sample of countries into four groups according to their relative position with respect to average values of taxing power and fiscal gap. Some countries, such as Italy, Switzerland and Canada, have a high level of financing autonomy and a low fiscal gap, whereas Austria has a low taxing power and a high vertical fiscal gap. The uk and Belgium, despite their ability to control their own tax revenues, can only raise a small amount of resources compared to the level of expenditures they face; thus they have both high taxing power and high vertical gap. Finally, countries such as France and Germany have a more limited taxing power (for instance, they rely significantly on shared taxes), but they are able to finance a higher share of their expenditures. Hence, they have both a low taxing power and a low vertical fiscal gap.

chapter 7

Taxing Powers of Subnational Entities: Between Domestic and Supranational Constraints Gisela Färber 1

Introduction

Discussions about reforms of multilevel financial constitutions all over the world always deal with the tax autonomy of subnational jurisdictions. Not only does the central level of government not like autonomous taxation competences on the part of lower levels of government, many states, provinces or regions, as well as municipalities and counties, refuse to have substantial tax autonomy. However, the power of taxation is essential for democracy, on the one hand (Boston Tea Party: “No taxation without representation!”), and for the efficient and accountable allocation of public goods and services provided by public expenditures, on the other hand, at all levels of government. All national constitutions contain rules about who should tax, where, and what,1 as it was formulated by Richard A. Musgrave more than 30 years ago. Tax assignments exist not only in federal countries, but also in regionalised countries where subnational tiers of government do not possess full statehood, but have restricted powers of local governments that, from a functional perspective, are equivalent to federal decentralisation. The power to tax includes not only the right to decide on tax bases and tax rates, but also the right to earn tax revenues as proper receipts even if the right to pass legislation is assigned at a higher level of government. The third taxation power is the right of tax administration, which becomes important if the administrating tier of government is not the same as the level that holds the right or legislation or in the case of shared taxes. This chapter starts in Section 2 with a review of the theory of tax assignment in multilevel government systems. The third section discusses the relationship between tax competition in a globalised world and decentralised taxation competences and resulting problems particularly for subnational governments. Section 4 introduces supranational institutions into what has been a “domestic” analysis up to this point and explains constraints on decentralised­ 1 R.A. Musgrave, “Who Should Tax, Where and What?” in C.E. McLure (ed.), Tax Assignments in Federal Countries (Canberra: Australian National University Press, 1983), at 2ff. © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_009

150

Färber

taxation powers that have been established by supranational treaties. The section focuses on the regulations contained in the Treaty on European Union (teu), which, with regard to constraints on national taxation powers, is the most advanced supranational legislation in the world. This chapter focuses on the “intermediate” level of government, i.e. Länder, regions, cantons, provinces and states, but it mentions lower levels of local government if needed for the analysis. 2

Tax Assignments in Theory and Practice: The Domestic Perspective

2.1 Theoretical Aspects of Tax Assignment The question of which level of government should decide on which taxes to levy, the specific tax bases and tax rates applied and which level of government should use them to fund public expenditures is a key element of intergovernmental financial relations. The functions of allocation (efficiency, accountability), distribution and stabilisation require not only the power to decide on public expenditures but also on the tax revenues necessary to cover the costs of the respective production of public goods.2 More recently, younger public finance experts have concretised the basic rules of the assignment of taxation in federal systems. In 1996, Olson specified three fundamental requirements:3 1.

2. 3.

Decentralised levels of government should have the power to tax those mobile tax sources that are based on a regional or local principle of equivalence. This is the case, on the one hand, for user fees and charges that correspond with specific public services and, on the other hand, for those taxes that have a strict relationship to defined user groups of regional or local public goods and services (in Germany, that relationship is called a group equivalence4). Higher levels of government should tax those tax bases that do not correspond with certain public goods and services and taxes that are used for distributive or redistributive policies. Local governments should not be assigned mobile tax sources.

2 See, R.A. Musgrave, The Theory of Public Finance, A Study in Public Economy (New York: McGraw-Hill, 1959). 3 See, W.E. Oates, “Taxation in a Federal System: The Tax-Assignment Problem”, Public Economic Review, 1 (1996), 36. 4 See, Wissenschaftlicher Beirat beim Bundesministerium der Finanzen, “Gutachten zur Reform der Gemeindesteuern”, bmf—Schriftenreihe, 31 (Bonn, 1982).

Taxing Powers of Subnational Entities

151

In 1997, Teresa Ter-Minassian described three features of taxes for decentralised jurisdictions in a more detailed manner:5 1. 2. 3.

They should be constructed on relatively immobile tax bases. The tax bases should be spread rather “evenly” among the jurisdictions. The tax sources assigned to decentralised levels of government should not vary across the economic cycle, and tax revenues should be rather stable.

From an economic point of view, the assignment of taxes across levels of government is connected with a bundle of goals: the assignment typically should support the efficiency and effectiveness of the provision of decentralised public goods by means of intergovernmental competition. It should, in addition, promote the equivalence of living conditions across the whole country as far as this corresponds with the preferences of regional and local voters. The assignment of taxes should also avoid externalities resulting from the tax sources assigned if jurisdictions de facto tax economic values that are economically “located” in another jurisdiction (e.g. the case of so-called tax exports). Intergovernmental competition is based on the principle of mobility of capital and enterprises and labour and citizens migrating to locations where they can find the optimal relationship between public goods, on the one hand, and taxes levied to cover the costs of the provision of public goods, on the other hand. Politicians who want to be re-elected decide on an efficient combination of public goods and corresponding tax rates according to voters’ preferences. Voters and enterprises influence the efficient bundle of—decentralised— public goods and tax rates by voting or by migrating to a jurisdiction that better fits with their preferences of public goods.6 “Vote or exit” is the channel of higher accountability within the public sector and adds to the simple model of democratic elections in revealing voters’ preferences. This model, however, requires the fulfilment of certain conditions, the most important of which is the congruence of voters, taxpayers and beneficiaries of the respective regional and local public goods. If the three groups are not congruent, the superior level of government, which, at the highest level, is the central government, has to internalise regional spillovers through regulation or through fiscal equalisation in order to guarantee the virtues of intergovernmental competition. 5 Compare, T. Ter-Minassian, “Intergovernmental Fiscal Relations in a Macroeconomic Perspective”, in T. Ter-Minassian (ed.), Fiscal Federalism in Theory and Practice (Washington: International Monetary Fund, 1997). 6 C. Tiebout, Charles, “A Pure Theory of Local Expenditures” The Journal of Political Economy, 64 (1956), 416–424.

152

Färber

Technical progress, increasing incomes, the decline of transportation costs and—from an international perspective—shrinking costs of cross-border transactions7 have led to a sharp increase in externalities related to public goods.8 These externalities create systematic opportunities for unfair tax competition because the jurisdictions that provide public goods and services and have to cover the costs are no longer capable of taxing those economic subjects that consume the benefits of public infrastructure. Economists advise a change of funding instruments from taxes to user fees; however, user fees are not applicable for those many public goods for which the exclusion principle does not work. That might also change the traditional recipes for the subnational assignment of taxes. 2.2 Typical Pattern of Tax Assignment in Practice The reality of the assignment of tax competences partly follows the abovementioned theory and partly depends on national historical paths that not only concern the tax system as a whole but also political compromises. Since taxes and tax exemptions not only serve to collect revenues, but are also aimed at influencing economic processes and income distribution, there are inbuilt trade-offs between both of the tax system’s main goals. It is, in particular, the “nature” of the available taxes in a modern economy that influences the assignment of taxation rights to the various levels of government. Since income taxes—both corporate and personal—and the general turnover tax dominate the tax systems in all modern economies, accounting for 60–80 per cent of total tax revenues, access to these taxes is crucial for proper fiscal capacity at all levels of government. Besides these two main tax sources, most countries impose excise taxes on mineral oil, tobacco, energy/electricity and alcohol, as well as property taxes and other taxes related to land transactions, taxes related to motor vehicles and taxes on wealth and on inheritances and gifts. However, many tax assignments in decentralised (federal and non-federal) countries show a broad variety of intergovernmental constructions particularly with regard to decentralised autonomy and with centralisation of tax bases and tax rates.

7 A statement that can be regarded as a very short summary of globalisation. 8 G. Färber, “Steuer- und Standortwettbewerb unter globalem Wettbewerbsdruck: Was muss der Finanzausgleich hierbei leisten?” in E.M. Belser, and B. Waldmann (ed.), Mehr oder weniger Staat? Festschrift für Peter Hänni zum 65. Geburtstag (Bern: Schulthess, 2015), at 302.

Taxing Powers of Subnational Entities

153

In many countries—about half of the oecd countries—states, provinces and regions tax the residential income of their citizens and decide at least on the tax rates and, in many cases, also on the tax base.9 They often adapt the tax base for the income tax levied by the central government and include some specific changes (e.g. in Canada) or they use very simple tax bases. They apply flat rates or even progressive tax rates (e.g. in Switzerland). More restricted are the competences in Spain. The autonomous communities can apply a specific tax rate on the national income tax base, while tax administration is centralised. In some countries, including in Scandinavia, Switzerland and in some us states, local governments also levy a personal income tax, in most cases piggybacking on the subnational tax base or tax payments, but deciding on the local tax rate. Only Germany concedes a pure shared revenue competence for the nationally uniform personal income tax to the Länder, which are responsible for tax administration, but do not have any autonomy over the tax base or tax rates. Corporate income taxes are more centralised than personal income taxes because of the higher mobility of enterprise capital. However, some us states apply this type of tax in addition to the federal corporate income tax. In Italy, the irap (imposta regionale sulle attività produttive) is a regional tax levied upon the net value added of enterprises where each region can vary the tax rate (currently 3.9 per cent) up and down by a maximum of 0.92 percentage points. In Germany and Switzerland, the corporate income tax is uniform throughout the country, and the revenues are shared between the federal government and the Länder and cantons, respectively. As far as the general taxation of consumption is concerned, the technical options for substantial decentralisation are limited. Though a simple sales tax on retail sales burdening regional final consumption is suitable as a tax for subnational governments and is applied, for example, in many us states, the type of value-added tax (vat) with deduction of input tax—applied in all eu member states—or of a related type of a goods and services tax (gst)— applied in Canada and in Australia is not adequate for decentralisation. If a province applies a higher gst rate, the revenues of the other provinces are affected by a reduction in the input taxes. Canada has tried to resolve this problem by means of a so-called harmonised gst with a uniform tax rate that is administered by the central government and distributed according a formula. 9 See R. Bahl and C. Musharraf, “Tax Assignment: Does the Practice Match the Theory?” International Studies Program, Working Paper 10–04, Andrew Young School of Policy Studies, (2010), at 6.

154

Färber

But only some provinces have agreed with this new concept. While Ontario is still searching for a solution, three other provinces maintain their “traditional” provincial retail sales tax. The Canadian province of Quebec levies a proper vat. In Australia, the gst, introduced in 2000 with a nationally uniform tax rate and administered by the Commonwealth (i.e. the federal government), is used entirely as source for fiscal equalisation. The European vat rate differs only from country to country without any subnational autonomy. Switzerland has applied a vat since 1995, the revenues from which are fed into the federal budget. Tax rates were increased in 1999 in order to stabilise the central government pension system. Excise taxes are often centralised as a result of their collection in the few places of production in a country against the need to simplify their tax administration. In the usa and in Canada, however, both the central and the state or provincial governments levy excise taxes on alcohol, gasoline, cigarettes and vehicle air conditioning. Vancouver and the Capital District in British Columbia and Montreal in Quebec also levy a local fuel tax. Property taxes are typically local taxes, for which the municipalities have the autonomous right to decide the tax rate. The tax base is, in most cases, the value of the real estate in question, although some taxes only burden land or buildings. In many countries where local governments consist of several layers (municipalities, counties, school districts, regions etc.), all local tiers have access to the same tax base. In France, not only are real state owners burdened by the property tax, but tenants also pay an additional taxe d’habitation. In most countries, land transfers are burdened by a specific tax that is often assigned to states, provinces or regions. In Germany, for example, the counties in most Länder receive a certain share of the state land acquisition tax. Many countries levy taxes on above-average wealth and on gifts and inheritances above a certain value. In all countries where these taxes are assigned to subnational governments, such as Australia, tax competition made them disappear. The only exception is Switzerland, where the cantons— with the exception of the canton of Schwyz—levy an autonomous gift and/ or inheritance tax, which is sometimes shared with the municipalities. But the revenues are of only marginal importance and therefore obviously irrelevant for tax competition. In all other cases, the central government has the legislative competences for these taxes in favour of an at least nationally uniform tax, while the revenues, as in Germany, can be assigned to the subnational level. Austria has abolished the wealth tax, to favour the country’s attractiveness for wealthy taxpayers, on the one hand, and probably as a wise concession to the high administrative costs of that type of tax, on the other hand.

Taxing Powers of Subnational Entities

3

155

International Tax Competition and Decentralised Powers of Taxation

The theories of fiscal federalism and intergovernmental competition outlined above have been put under pressure. Before globalisation, national borders not only protected the private provision of goods and services but also the mobility of enterprises and voters by restricting the mobility of capital and labour or at least by imposing high costs on all actors undertaking economic activities across national borders. The establishment of the European Single Market in 1993 and of the Economic and Monetary Union in 2002 are the most extensive examples in history of the establishment of transnational common markets without the costly barriers of national borders in order to guarantee the free movement of goods and services and of capital and labour. Hence, the volume of trade and inter-industry trade increased more and more, the prices of many goods and services decreased, and many jobs were transferred to lowwage, emerging market economies.10 The high-wage industrialised economies underwent a long-lasting and severe structural change in favour of knowledgeand innovation-based industrial production and service-sector jobs, which coincided with the digitisation of production and consumption. Europeanisation and globalisation not only changed the conditions of private production but also the conditions of the production of public goods and services. Long-lasting structural economic changes increased the number of unemployed and the amount of social transfer payments, while the resulting growing social insurance contributions became a greater and greater threat to the international competiveness of national jobs. The increased mobility of capital and labour and the openness of capital and labour markets established a type of Tiebout competition among countries’ subnational entities, creating a “competition of systems” within entire national public sectors.11 This new “competition of systems” also changed the rules and conditions of international tax competition in general and the modes of operation of subnational taxation and tax autonomy in particular. Tax Competition: A Race to the Bottom or a Shift in the Mobile Shares of Income to Tax Havens? The legally guaranteed mobility of capital and labour among European member states and the sharply increased mobility thereof worldwide provided a

3.1

10 11

Deutscher Bundestag, “Schlussbericht der Enquete-Kommission Globalisierung der Wirtschaft”, BT-Drs. 14/9299, Berlin, 12.6.2002, at 50. See e.g. H.-W. Sinn, The New Systems Competition (Oxford: Wiley-Blackwell, 2003).

156

Färber

new pattern of tax competition that was called “a race to the bottom”. In order to attract enterprises and jobs to their countries, governments reduced tax rates, particularly the corporate income tax and the withholding tax on capital income, reducing those tax rates closer and closer to zero. As a result of this ruinous tax (rate) competition, all governments would suffer from a lack of revenues, which again would lead to reductions in expenditures, thereby shrinking public sectors in total. The result of tax competition may well be a tendency toward less than efficient levels of output of local services. In an attempt to keep taxes low to attract business investment, local officials may hold spending below those levels for which marginal benefits equal marginal costs, particularly for those programs that do not offer direct benefits to local business.12 Reality, however, shows that the bottom has a considerable value above zero. Neither wages, nor labour and social standards13 nor social transfers declined as expected to a worldwide minimum level. Several reasons can be identified: • labour mobility is restricted as long as people find sufficiently paid jobs in their local, regional or national context, but it varies with their level of education; • most enterprises depend on local “natural resources” and infrastructure, as well as on the availability of a specifically qualified labour force and the physical distance to their markets; • politicians needed to introduce social transfer programmes in response to increasing job losses resulting from globalisation. However, social insurance contributions, which are a direct part of labour costs, and direct taxes on capital and on mobile labour forces have come under pressure during the last 20 years. Social insurance contribution rates in competitive countries remained rather stable, after years of increases as a result of increasing unemployment from globalisation; maximum personal income tax rates were lowered; and special, low tax rates for highly skilled expatriates were 12 13

W.E. Oates, Fiscal Federalism (New York: Harcourt Brace, 1972), 143. See, N. Protafke, “Globalisierung und Arbeitsmarktinstitutionen”, in ifo-Schnelldienst, 4 (2013), at 32.

157

Taxing Powers of Subnational Entities 55.00% 50.00% 45.00% 40.00% 35.00% 30.00% 25.00% 20.00%

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 deu

fra

gbr

lux

nld

usa

eu average

G20 average

Figure 7.1 Corporate tax rates 1999–2015. Source: J. Vella, Nominal vs. Effective Corporate Tax Rates Applied by mne and an Overview of Aggressive Tax Planning Tools, Instruments and Methods (Brussels: European Parliament—Directorate General for Internal Policies, 2015), at 12.

introduced.14 Most countries lowered the rates of the corporate income tax and the tax on dividends and other capital gains (see figure 7.1). For the majority of countries, the “race to the bottom” obviously does not end up at zero. The reason for this is that national governments often combine reductions in tax rates with expansions of the tax base by, for example, reducing depreciation allowances or taxation of minimum profits.15 In addition, labour—even highly skilled labour—which is complementary to capital, is not completely mobile due to personal and family preferences, as well as to transaction costs of labour mobility, such as language barriers, diverging prices in housing markets and living costs, divergent school systems for children etc. Capital and enterprises also meet mobility costs, which result from diverging legal, institutional and bureaucratic systems; therefore, the factual mobility of changing production places or establishing new ones abroad differs with the size of enterprises—favouring larger ones—and with branches and sectors. 14

15

See, E. Reimer, “Internationales Finanzrecht”, in J. Isensee and P. Kirchhof (eds.), Handbuch des Staatsrechts der Bundesrepublik Deutschland. Band xi: Internationale Bezüge (Heidelberg: C.F. Müller, 2013), § 250, at 981. See Sachverständigenrat zur Begutachtung der gesamtwirtschaftlichen Entwicklung, „Erfolge im Ausland—Herausforderungen im Inland“ Jahresgutachten 2004/05 (2004), at 755, https://www.sachverstaendigenrat-wirtschaft.de/fileadmin/dateiablage/download/ gutachten/04_gesa.pdf (accessed 28 July 2017).

158

Färber

The proximity to markets and clients, the availability of qualified labour force and the size and quality of public infrastructure count much more than tax rates.16 The costs for the latter are financed by tax revenues, particularly from the budgets of decentralised levels of government, which are usually responsible for education, public transport, water provision and sewerage, economic permits etc. As there is a clearly beneficial relationship between regional and local infrastructure services with only limited externalities for the majority of local public goods, local enterprises and the mobile labour force they employ accept tax rates that are substantially above zero. Larger enterprises, however, which typically produce goods in many countries and deliver to international markets, have developed strategies for avoiding taxes by applying low tax rates in specific locations and so-called “tax rulings”­for certain shares of their taxable profits. Techniques for “aggressive tax planning” were first published by the Organisation for Economic Co-operation­ and Development (oecd)17 and have become very popular for many international enterprises. They typically shift parts of their profits to subsidiaries in countries with low tax rates. These subsidiaries bill their parent companies for the respective “services” and the latter then deduct these costs from the tax base in the high-tax country where production takes place and where national, regional and local infrastructure is used for production at internationally competitive prices. The most common tax avoidance models are known as the “Double Irish Dutch Sandwich” and “ip Holding Structure Using an ip Box Regime”.18 Using these strategies, companies not only minimise their own tax payments, but they also reduce the tax revenues in the countries where they have their production facilities even more, because the latter lose a considerable share of their local or regional tax bases. The oecd therefore emphasises the problems as “Base Erosion and Profit Shifting” (beps).19 Reforms that reduce the damages of beps activities are difficult to implement, as a majority of them require international cooperation, including in those countries that actually profit from profit shifting in a double sense: they receive tax revenues 16

See M. Berlemann and J. Tilgner, “Determinanten der Standortwahl von Unternehmen”, ifo Dresden berichtet, 6 (2006), at 14 with more detailed references. 17 See oecd, Engaging with High Net Worth Individuals on Tax Compliance (Paris: oecd, 2009). 18 See C. Fuest et al., “Profit Shifting and “Aggressive” Tax Planning by Multinational Firms: Issues and Options for Reform”, zew Discussion Paper No. 13–078, (Mannheim, 2013), at 4. 19 See oecd, “beps Final Reports”, (Paris, 2016) http://www.oecd.org/tax/beps-2015-final -reports.htm (accessed 28 March 2017).

Taxing Powers of Subnational Entities

159

without economic value added within their territory,20 and they need not cover the costs of infrastructure for effective places of production. Effects of International Tax Competition on Subnational Taxation Powers Subnational taxation powers are severely affected by these reduced tax rates and beps activities on the part of multinational firms. The effects can be categorised as those related to the power of tax legislation and those related purely to revenue competences. Subnational governments typically have limited taxing powers, but they often own and share bases of enterprise profit taxes and/or can set tax rates for shared tax bases. In Germany, for example, revenues from the corporate income tax are shared equally between the federation and the Länder. In Italy, the regions can increase or decrease the nationally uniform irap tax rate. Several countries levy a local business tax on enterprise profits (Germany, Luxembourg, Hungary), while local governments can decide on the tax rates or on local multipliers. The revenues from taxes that are owned or at least shared by subnational entities on enterprise profits, particularly the corporate income tax and the trade tax, suffer from substantial losses. In Germany, the corporate income tax rate was reduced to 25 per cent in 2001 and to 15 per cent in 2009. Although both years were marked by profound economic recessions, the decrease in the tax rate reduced tax revenues to far below gdp growth rates and enterprise profits. In the following economic recovery, tax revenues were considerably weaker than economic production and profits resulting from lower tax rates and continued tax evasion on the part of firms active across national borders (see figure 7.2). Shrinking tax bases cannot be compensated by increases in tax rates or local multipliers because they are subject to tax competition. German municipalities increased the local multipliers on the trade tax between 2005 and 2015 by 10 percentage points from 389% to 399%. In the same period, the local multiplier on property tax increased by 63 percentage points from 392% up to 455%. Particularly near the borders of Länder, e.g. from Northrhine-Westfalia to Rhineland-Palatinate, enterprises shifted their official locations to neighbouring communities with lower trade tax multipliers when the municipality where they had been located for many years increased its trade tax multiplier. 3.2

20

With the exception of remuneration for lawyers, leases for letter-box companies and fees for the registration of subsidiaries.

Färber 1.4%

6%

1.2%

5%

1.0%

4%

0.8%

3%

0.6%

2%

0.4%

1%

0.2% y = –0.0001x + 0.0087

19 9 19 1 92 19 9 19 3 94 19 9 19 5 96 19 9 19 7 98 19 9 20 9 0 20 0 20 01 0 20 2 0 20 3 04 20 0 20 5 06 20 0 20 7 08 20 0 20 9 1 20 0 20 11 1 20 2 1 20 3 14 20 15

0.0% Corporate Income Tax/gdp

0%

Corporate Income Tax/Profits and Capital Gains

Corporate Income Tax/gdp

160

Corporate Income Tax/Profits and Capital Gains

Figure 7.2 Corporate tax revenues in relation to gdp and to profits and capital gains in Germany 1991–2015. Source: Statistisches Bundesamt; own calculations.

Other countries with a century-long tradition of trade taxes have restricted the taxable base for local governments to those parts of the added value that were not as mobile as the pure financial capital (e.g. the payroll tax in Austria and the taxe professionelle in France on the rental value of assets available to the respective enterprise). The Austrian payroll tax also avoids domestic tax competition by means of a nationally uniform tax rate of 3 per cent on wages. Switzerland, where the cantons possess far-reaching tax autonomy concerning, among other things, personal and corporate income tax and the wealth tax, and where the distances from canton to canton are not very far, has seen another development. For many years, some cantons had been engaged in intensive tax competition with their neighbours. Low corporate income tax rates and even regressive personal income tax rates attracted enterprises and inhabitants. In 2007, the Swiss Constitutional Court judged the regressive tax rates of Obwalden to be unconstitutional as they were violating the principle of taxation according to the ability to pay.21 Tax competition continued, however, via

21

The judgement against Obwalden is published under http://www.servat.unibe.ch/dfr/ bger/070601_2P_43-2006.html (accessed 27 March 2017).

Taxing Powers of Subnational Entities

161

effectively regressive tax rates,22 which contributed to continuing growth of “letter-box enterprises” in low-tax-rate cantons and an increase in population and housing prices, as well as rents.23 Growing externalities created a convergence of tax bases because more people with greater mobility came into contact with several cantonal taxes.24 All these interventions mean de jure or de facto restrictions on subnational taxation competences. Subnational governments therefore suffer not only from a loss or lack of revenues, but—much more importantly—from restrictions on the efficiency of the democratic decision-making process, which includes a certain degree of revenue powers. They have, on the one hand, become more dependent on vertical transfer payments from higher levels of government or they have, on the other hand, substituted lost revenues with other revenues and tax sources, which could both contribute to more inefficiency. At the least, higher levels of government, especially national governments, should have taken into account necessary adjustments in the assignment of tax competencies as well as the volume and structure of fiscal equalisation grants resulting from these changes. 4

Constraints from Supranational Treaties: The Example of the European Union

While the North American Free Trade Agreement created a pure trade union, the European Union created, from the beginning, a far-reaching system of rules in order to secure the free movement of goods and services and of capital and labour. Although many detailed regulations were implemented during a longer period, agreements on the mobility of labour or the banning of public aids existed as part of the founding treaties of the European Economic Community. The treaties transferred national competencies to the European level. The European Court of Justice passed important judgments regarding many of the details of practices by national and subnational governments that are now 22

23

24

M. Roller and K. Schmidheiny, Effective Tax Rates and Effective Progressivity in a Fiscally Decentralized Country (London, 2016) http://cep.lse.ac.uk/seminarpapers/26-02-16-KS .pdf (accessed 27 March 2017). A. Gigon, “Kantone ziehen im Steuerwettbewerb die Handbremse”, swi swiss.info, 04.02.2014 http://www.swissinfo.ch/ger/steuersatz_kantone-ziehen-im-steuerwettbe werb-die-handbremse/34888572 (accessed 27 March 2017). L.P. Feld, Braucht die Schweiz eine materielle Steuerharmonisierung? (Zürich: economiesuisse, 2009), at 54ff.

162

Färber

banned because of their damaging impact in terms of the European Union’s four basic freedoms. European regulations concern national and subnational tax autonomy in a three ways: • The competences for indirect taxes (vat and excise taxes) are transferred to the European level. The eu regulates tax bases and minimum tax rates. Decisions concerning taxes according to Articles 110–13 of the teu must be unanimous. vat—one of the most important taxes with regard to revenues in every member state—is, in contrast to sales taxes in us states or Canada’s gst, not open to decentralisation, as the principle of the deduction of a prepaid tax distorts regional and local revenues in favour of places of production and at the burden of places of consumption. These distortions cannot be compensated by divergent tax rates without distorting the competitive ability of enterprises and locations. For the same reasons, vat does not follow the principle of destination in case of inter-EU trade; instead, it still applies the principle of origin.25 Therefore, divergent subnational vat tax rates are de facto ruled out in the European Union. In principle, excise taxes are open to the decentralisation of tax bases, tax rates and tax revenues. Good examples are the independent excise taxes that the Canadian provinces levy on alcohol, tobacco, fuel and natural resources, many of which have parallel federal taxes.26 The German Länder own the revenues from the uniform national beer tax without their own competences regarding the tax base or the tax rate, while all other excise taxes are federal taxes because of their extremely uneven regional revenues as a result of their collection in the few places of production and not on the “regional consumption” of the taxed products.27 However, eu regulations have restricted the number of excise taxes to those on fuel, tobacco, 25

26 27

An agreement was discussed for many years. Even financial compensation for the revenue losses of member states where imports exceed exports from the gains of countries with export surpluses could not find the necessary unanimous support among member states (see Ursprungslandkommission, “Umsatzbesteuerung in Europa nach dem Ursprungslandprinzip ab 1997”, Bundesministerium der Finanzen, Schriftenreihe no. 52 (1994), at 67. See R. Boadway and R.L. Watts, “Fiscal Federalism in Canada, the usa, and Germany”, Institute of Intergovernmental Relations, Queen University, Working Paper No. 6 (2004), at 7. Germany is the only wine-producing country that does not levy a wine tax. This stems from the late 1920s, when winegrowers violently protested against the wine tax until it was abolished. See H. Becker, Handlungsspielräume der Agrarpolitik in der Weimarer Republik zwischen 1923 und 1929 (Stuttgart: Franz Steiner Verlag, 1990), at 248.

Taxing Powers of Subnational Entities

163

alcoholic beverages and coffee, which are not open for decentralised tax administration. Therefore, in Europe, excise taxes are national taxes, with the exception of Germany’s beer tax. • Further restrictions result from the European guarantee of its “four fundamental freedoms”. If a national tax hinders these freedoms or particularly discriminates against foreign eu competitors concerning cross-border transactions, Brussels can intervene against the respective national tax laws. This “European law” has mainly been developed by the European Court of Justice since the 1980s28 through judgments that have forbidden all tax regulation that restricts the free movement of goods and services, labour and capital, or that make these things less attractive. For example, the taxation of capital appreciation upon taking up residence abroad has been banned by European case law. Other cases have concerned the compensation of losses from abroad with domestic gains, the integration of personal and corporate income tax, cases of undercapitalisation and the difference between unrestricted and restricted liability for taxation.29 eu case law concerns subnational tax competences in two ways. On the one hand, regional and local tax administrations have lost their ability to prevent enterprises from moving abroad because they cannot tax their hidden reserves, which in fact are untaxed profits earned using the infrastructure in the respective enterprise’s former location. More fiscally important risks arise from revenue losses resulting from the deduction of foreign enterprise losses from domestic profits and high tax repayments. While national governments’ tax revenues are rather high compared with regional or even local ones, the risks of tax revenue losses weigh much higher at lower, subnational and local levels of government. Although European case law concerning enterprise taxation is implied with the rules of the barrierfree European Single Market, experts classify these risks as hindering fair tax competition,30 particularly because the restrictions are not only valid for the taxation of trans-European enterprises, but also for international enterprises from non-EU countries. European regions and municipalities therefore suffer from a loss of taxation competences and from a much more harmful loss of tax revenue capacities.

28 29 30

See J. Kokott, and H. Ost, “Europäische Grundfreiheiten und nationales Steuerrecht”, Europäische Zeitung für Wirtschaftsrecht, 22 (2011), at 498. See C. Fuest, “EuGH-Rechtsprechung zur Unternehmensbesteuerung”, Wirtschaftsdienst, 1 (2005), at 21. See, ibid.

164

Färber

• Oversight over state aids in accordance with Article 107 teu has been valid since the beginning of the European Economic Union. Its application, however, was developed later and has included tax exemptions in case “any aid granted by a Member State or through State resources in any form whatsoever which distorts or threatens to distort competition by favouring certain undertakings or the production of certain goods shall, in so far as it affects trade between Member States, be incompatible with the common market”.31 Exceptions are only permitted for the targets mentioned in Article 107(2) of the teu and need a specific allowance. “The criteria for identifying potentially harmful measures include: • an effective level of taxation which is significantly lower than the general level of taxation in the country concerned; • tax benefits reserved for non-residents; • tax incentives for activities which are isolated from the domestic economy and therefore have no impact on the national tax base; • granting of tax advantages even in the absence of any real economic activity; • the basis of profit determination for companies in a multinational group departs from internationally accepted rules, in particular those approved by the oecd; • lack of transparency”.32 The most recent interventions in national tax competences have been the decisions taken against so-called tax rulings. These are specific decisions on the definition of a corporate tax base taken by national governments in advance that, in the end, reduce the individual enterprise tax burden. This is how the Irish taxation administration approved a specific construction project on the part of two Amazon subsidiaries that did not correspond with national economic activities and whose only goal was to avoid tax payments in other European countries. As a result, Amazon was able to reduce the effective tax rate on its total European profits from 1 per cent in 2003 to 0.005 per cent in 2014, conditions that other enterprises located in Ireland can never achieve. The European Commission decided that Amazon had to repay this illegal state aid to the Irish government.33 31 32

33

Art. 107 (1) teu. European Commission, “Taxation and Customs Union: Harmful Tax Competition” http:// ec.europa.eu/taxation_customs/business/company-tax/harmful-tax-competition_en (accessed 06 June 2017). See European Commission, Report on the Competition Policy 2016; com 2017 (285) final, 31 May 2017, http://ec.europa.eu/competition/publications/annual_report/2016/part1 _en.pdf (accessed 06 June 2017).

Taxing Powers of Subnational Entities

165

Similar decisions have been taken in the cases of Fiat and McDonald’s in Luxembourg and Starbucks in the Netherlands. The Commission is reviewing further tax rulings in these countries and in Belgium.34 In January 2016, the European Commission passed an Anti Tax Avoidance Package aimed at fairer, simpler and more effective corporate taxation in the eu. The package contains measures against aggressive tax planning by establishing transborder tax transparency for enterprise profits and new transborder-taxation principles limiting the advantages of a pure transfer of parts of profits to countries with extremely low corporate income tax rates.35 In the long run, the Commission favours a common corporate income tax base for all member states,36 which, however, needs the unanimous vote of all member states, including those that actually profit from their low and unfair tax rates. Subnational governments and their taxation competences are affected by European policies in a limited way because of their limited competences in the field of corporate income tax. In all cases where states, regions and provinces have access to corporate income tax by deciding on the tax rate or by receiving a share of tax revenues, they are in principle subject to the same restrictions as national governments with regard to the definition of the tax base and share the consequences of European regulations concerning their tax revenues. These are typically negative in case of national tax regulations that discriminate against foreign companies or protect domestic ones, and they will be positive in case the regulations contained in the new Anti Tax Avoidance Package are transferred into the national law of member states. These procedures will, however, take a number years to be introduced, although some measures, like the limited deduction of interest payments or the change from tax-free regimes to a deduction of taxes paid abroad, work directly in favour of securing national tax bases and will therefore be adopted by the member states where profits are produced.

34 35

36

See W. Franz, “Steuerwettbewerb und Beihilfenverbot—die Fälle Fiat, Starbucks und McDonald’s”, Deutsche Steuerzeitung, 5 (2016), at 142. Proposal for a Council Directive Laying down Rules against Tax Avoidance Practices that Directly Affect the Functioning of the Internal Market, com(2016) 26 final, 2016/0011 (cns), (28 January 2016), http://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1454056979779& uri=COM:2016:26:FIN (accessed 06 June 2017). See European Commission, “Corporate Income Tax Reform—Pro Business / Anti Avoidance”, (Brussels, 25 October 2016, https://ec.europa.eu/taxation_customs/sites/taxation/ files/ctr_factsheet2016.pdf (accessed 05 June 2017).

166 5

Färber

Decentralised Taxation Powers and Intergovernmental Financial Relations

Globalisation and Europeanisation have increased international tax competition and de facto—and sometimes de jure—reduced not only national tax autonomy, but also the subnational competencies to decide on the tax bases and tax rates necessary to cover the costs of public services and transfer payments. The unfair competition of so-called tax havens threatens the subnational and local levels of government more than the national tiers, which have typically been assigned a broader package of taxes and “owns” the general consumption tax, which, in comparison, is less affected by tax competition. For municipalities, modern-style tax competition has a dual nature, since internationally mobile enterprises no longer move their production facilities to locations with lower tax rates, they only “transfer” important shares of the tax base abroad, while the local government in the place of production continues to cover the costs of local services for these enterprises. The changing quality of international tax competition raises the question of whether a reduction in taxes that makes sense with regard to tax competition should not be compensated by fiscal equalisation. Fiscal equalisation is, in most cases, conducted via vertical transfer payments or, on an exceptional basis, through deductions of the subnational taxes paid from the federal tax base. Until now, it has been aimed—among other allocative goals and the distributive target of reducing horizontally diverging fiscal capacities—at compensating for deficiencies in the vertical tax assignment, which mainly results from restrictions on subnational tax competences in favour of an efficient national tax policy and from specific constructions of tax bases and tax exemptions particularly in the cases of vertically shared tax sources. Intensified and unfair international tax competition has created a new “deficiency” in subnational tax autonomy, which subnational entities cannot fight against with their own constitutional powers because the legislative competences for “mobile” taxes is restricted or centralised at the central level. As long as national governments cannot find a reliable agreement at least on how to avoid unfair tax competition, subnational tax competences will be severely reduced. Solutions can be found, for example, through specific grants for compensation of tax revenue losses if the new eu regulation on increased transparency makes it possible to calculate such grants in detail. With regard to the very few typical benefit taxes—property tax, residential income tax and, with restrictions, local business tax on net value added, which often are traditional local taxes—an increase in vertically shared income taxes with tax rate autonomy or a piggybacking element might help to avoid at least a harmful national tax

Taxing Powers of Subnational Entities

167

competition. The worldwide trend in favour of general consumption taxes, particularly VAT- and GST-type taxes, which are only restrictively open for decentralisation or would otherwise damage domestic and regional trade in goods and services, finally needs decentralisation of a larger share of the respective revenues. Tax-sharing formulas like in Canada or fiscal equalisation approaches like in Australia or in Germany could build the basis for reliable financial equipment at the subnational and local level of governments. However, the advantages of efficiency-increasing tax competition are preserved or even strengthened, if these levels of government are assigned substantial tax autonomy on benefit taxes and on taxes burdening regional transactions. 6

Summary and Conclusion

This chapter shows that subnational and local tax assignment in principle follows the theoretical requirements and proposals of the theory of tax assignment in fiscal federalism, although a tremendous variety of specific constructions is available. In many countries, however, subnational competences concerning their autonomous decision-making at least on tax rates, but also on tax bases, are restricted and often conflict with the demands of a “rational” tax policy that supports economic growth in a globalised world with free trade across all types of borders. Globalisation has not only increased free trade, but it has also caused an increase in unfair tax competition. While (subnational) tax competition usually guarantees a higher level of efficiency in terms of the production of regional and local public goods and services, modern, unfair international tax competition has damaged subnational taxation powers, as companies shift important parts of their tax bases to locations with low tax rates, typically to so-called “letter boxes” without any further production of economic value. A restriction of national and subnational tax autonomy also results from European law banning all tax structures that discriminate against foreign enterprises and all tax exemptions that “[distort] or [threaten] to distort competition by favouring certain undertakings or the production of certain goods shall, in so far as [they] [affect] trade between Member States” (Article 107 of the teu). Finally, the growing importance of “immobile” general consumption taxes of the gst and vat type, which do not interfere with free trade, has provided centralisation of tax competences, e.g. explicitly in Canada and implicitly in all countries where central governments possess a growing share gst or vat of all tax revenues. A discussion of whether these developments can cause changes in the functions of fiscal equalisation schemes and which constructions in particular

168

Färber

could help maintain or create the substantial subnational tax autonomy necessary for public-sector efficiency has not yet taken place. New challenges might result from the United States and the tax plans of President Trump. He announced that the United States will no longer follow the traditional principles of international corporate income taxation, but, instead, will establish a specific regime that favours profits resulting from production within the United States and taxes imported products with higher rates which economically means a higher taxation of the consumption of goods and services produced abroad. This type of corporate income tax is called a “destination-based cash flow tax”, which, according to prominent economists, would provide less deterioration of international investment decisions and reduce incentives for the “construction” of enterprises purely based on tax saving reasons in case that all counties would apply that type of cash flow tax.37 In case not all countries plan to follow this new tax regime, the question arises as to whether there will be a new and global form of unfair tax competition that damages free trade and also how this new regime will fit with subnational tax competences and affect interregional tax capacities.

37

See A.J. Auerbach et al., “Destination-Based Cash Flow Taxation”, Oxford University Center for Business Taxation, Working Paper 17/01 (Oxford, 2017), at 11.

chapter 8

Can Lessons from Equalisation Transfers in Industrial Countries be Applied to Reforms in Emerging-Market Countries? Ehtisham Ahmad and Giorgio Brosio 1

Introduction

Equalisation transfers occupy a large, if not excessive, place in the literature on intergovernmental relations. A few landmark contributions explore the equity and efficient properties of the various distinct equalisation systems, and their place in the broader framework of regional, macroeconomic and personal redistribution policies.1 The analytical contributions are flanked by a myriad of policy-oriented works, suggesting the application to single countries of the various normative models elaborated by the theory. There is also relatively new empirical literature mainly started by the Organisation for Economic Cooperation and Development (oecd)2 that focuses, inter alia, on the properties of the various equalisation models starting with their distributional impact and their cost-effectiveness. Political economy approaches are also a major component of the literature on intergovernmental grants. The allocation of grants is obviously subject to political pressure and gaming among different levels of government.3 International financial organisations and donors have exerted a huge effort aimed at adapting the equalisation systems of industrialised countries 1 For a recent review, see R. Boadway, “Intergovernmental Transfers: Rationale and Policy” in E. Ahmad and G. Brosio (eds.), Handbook of Multilevel Finance (Cheltenham: Edward Elgar, 2015) 410–436. 2 oecd Intergovernmental Network, since 2004. 3 See, for example, P. Grossman, “A Political Theory of Intergovernmental Grants”, Public Choice, 78 (1994) 295–303; A. Dixit and J. Londregan, “Fiscal Federalism and Redistributive Politics”, Journal of Public Economics 68 (1998) 153–180. See, for India, S. Khemani, “Partisan Politics and Intergovernmental Transfers in India”, Development Research Group—the World Bank, Working Paper 3016 (2003); and for Mexico, A. Diaz Cayeros, Federalism, Fiscal Authority, and Centralization in Latin America (Cambridge: Cambridge University Press, 2006).

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_010

170

Ahmad and Brosio

for introduction in developing countries. These have been largely unsuccessful with two main exceptions: China in 19944 and Indonesia in 2000. In both cases, the equalisation systems facilitated an overhaul of revenue assignments, and were part of a political solution sought by the concerned central government in consultation with subnational entities. In the Chinese case, there was a centralisation of tax administration, combined with the introduction of a value-added tax (vat), and shared revenues on an origin basis. This was complemented by the gradual introduction of an equalisation framework in a very interesting manner that brought about the tradeoffs between structural change and redistribution, as we shall see later in this chapter. In the Indonesian case, there was a major reaction to the centralisation of the Suharto years, and the sharing of petroleum revenues with producing regions was critical in maintaining national unity at a time of considerable centrifugal tendencies, including the secession of East Timor. This sharing arrangement had to be balanced by an explicit equalisation framework that, like the Chinese case, attempted to emulate the best-practice Australian and Scandinavian model based on spending needs and revenue capacities. The purpose of this chapter is to contribute to this debate by focusing on the origin and evolution of national equalisation systems. This historical perspective, which also makes use of political economy factors, helps us understand why equalisation systems have not taken hold in most developing countries, with a few exceptions, such as the two mentioned above. It can also provide suggestions crucial to shaping them so that they exhibit their equalisation impact within existing constraints deriving from the scarce availability of resources that characterises these countries. The paper is structured in three large sections. The first provides the analytical framework that equalisation is expected to operate within. We illustrate, in section two, the utilisation of intergovernmental equalisation transfers in a group of oecd countries, both unitary and federal, including Australia, Canada and Italy. The third section focuses on the introduction of equalisation systems in emerging market economies—the few examples include China and Indonesia. The concluding section provides some policy reform suggestions.

4 See J. Lou, “Constraints in Reforming the Transfer System in China”, in E. Ahmad (ed.), Financing Decentralized Expenditures: An International Comparison of Grants (Cheltenham: Edward Elgar, 1997) 349–360.

Can Lessons from Equalisation Transfers

2

171

Some Analytics of Equalisation Transfers and Advanced Country Examples

Equal Ability to Provide Similar Levels of Services at Similar Levels of Revenue Generation—Legal Constraints and Standards There is a national interest in every country to avoid significant divergences in the ability of subnational governments to provide public services, independent of the country’s level of decentralisation, i.e. given the assignment of policy responsibilities between the various levels of government. Uniformity does not necessarily imply that each local government has to provide exactly what is provided by others at the same level. Adaptation to local circumstances is needed. With this consideration in mind, one could also say that uniformity of access is almost synonymous with compliance with the standards or rules established by the central government. Since these rules tend to be equal for all, the concept of uniformity applies. There are a number of reasons for sticking to the rules, as they are the essence of what constitutes a nation, or common economic space. These reasons are connected to the positive or negative externalities deriving from the local provision of local services, and from the use of local tax instruments. Some of these externalities impact directly on individuals (or firms). Other externalities impact on the policies pursued by the central government. Most of them impact on both. This would be the case where a local government provides very poor primary education services, impacting first on the human capital of pupils and, then, on the provision of secondary education by the central (or any other level of) government. Uniformity is not an absolute concept and can vary in intensity. A convenient way of representing it for expenditure lies in the following expression: 2.1

∑E t

1

c ,d ,e , f … j .



(1)

where the expenditure, E, for each service, t, by any local government is determined by a set of characteristics/constraints, c,d,e,  f…j, that determine the quality and quantity of the service that local governments are legally obliged to comply with. For total revenue, R, for each revenue source, R the equivalent expression is the following:

∑

r 1

Rw , y…n . 

(2)

meaning that the proceeds from each tax or fee, R are determined by the characteristics w, y...n. Examples of characteristics would be tax rates, exemptions

172

Ahmad and Brosio

from public transportation fees or from payment for health services for the elderly poor. Full uniformity of service provision requires the inclusion of very detailed constraints determining every relevant characteristic of quality and quantity. If effectively implemented, the constraints would make the operation of a ­decentralised system of government analogous to that of a centralised system, but then there would be no rationale for the existence of a decentralised system of government. And as pointed out in Ahmad and Thomas,5 the use of full equalisation turns the process into one of “gap-filling” that generates adverse incentives in the intergovernmental framework. This permits local entities to reduce their revenue efforts and inflate their spending in order to influence their grant allocations. And with typically imperfect monitoring of budget execution at the subnational level in many cases (including in eu countries), there are enhanced incentives for intergovernmental “game-play”.6 Full homogeneity is neither feasible nor desirable, of course, and the essence of decentralisation is to be found in local responses to voters’ preferences. Good governance is then achieved through either an interjurisdictional competition generated through the electoral process, as in Western-style political systems, or detailed performance measurement in administrative promotion systems. Equalisation systems based on combinations of standardised needs and revenue capacity permit the application of local preferences, with effective “lump-sum” compensation for exogenously determined cost factors or revenue bases. This allows local jurisdictions to provide standardised levels of service at similar levels of tax effort, should they so desire. Thus, variations in service delivery would not be rewarded nor would they adversely affect the equalisation/ vertical transfers. This generates positive incentives as far as local governments are concerned, as they are unable to influence the equalisation/vertical grant through their own actions, as the grant becomes effectively a “lump-sum”. It is common that, in some areas, the central or federal government requires absolute equality—especially if there are interjurisdictional externalities. As this overrides local preferences, it is then important to use “special-purpose transfers” for such activities if they happen to fall under the competence of more junior levels of administration. The effectiveness of such “directed” spending is often questionable, as the central government typically relies on junior administrations to implement these responsibilities. With imperfect 5 E. Ahmad and R. Thomas, “Types of Transfers: A General Formulation”, in Ahmad, Financing Decentralized Expenditures, supra, 361–381. 6 E. Ahmad, M. Bordignon and G. Brosio (eds.), Multi-level Finance and the Euro Crisis: Causes and Effects (Cheltenham: Edward Elgar, 2016).

Can Lessons from Equalisation Transfers

173

information about how well or what such directed “special-purpose” transfers are spent on, it is often problematic to ensure compliance, even when uniformity is implied in purely legalistic terms. This has been the case, for example, in unitary states, such as China, or an even more striking example was the Soviet Union. Establishing uniform standards using special-purpose grants, e.g. for education, has proven to be quite difficult in federal states, including Mexico and the United States (e.g. the mixed results of the No Child Left Behind initiative started by former President George W. Bush and continued under President Barack Obama, which has been seen as overriding state responsibilities). It is important to be clear about the functions that are subject to local ­preferences—the crux of a decentralisation system—and for which an equalisation framework is deemed to be necessary for efficiency, and both spatial and interpersonal equity issues. We focus in this paper on “equalisation systems”, and not on special-purpose transfers that also dominate the fiscal landscape in most countries, particularly in emerging markets. Typical public services also interact with dynamic issues that have not been very well captured in the analytical literature, but which are of considerable policy importance. Of particular significance is the issue of financing infrastructure for new urban hubs for sustainable growth, something that is not typically covered under equalisation schemes. The issue at hand is the interaction between current local public services, which are typically supported by equalisation, as well as the allocation of larger investments that generate cross-jurisdictional externalities. The success of sustainable development strategies in both emerging and developed countries will lie in a better alignment of current spending and investment allocations among different levels of government. 2.2 Uniformity and Interjurisdictional Equity The principle of interjurisdictional equity provides the rationale for equalisation transfers. A general formulation of the principle says that people in comparable circumstances should have access to comparable public services in all localities.7 In other words, in the intergovernmental framework equity implies that residence should not create differences among citizens in their access to public services and to the cost of access. There are, however, different interpretations of this principle.8 7 Boadway, “Intergovernmental Transfers”, supra. 8 G. Brosio, “Intergovernmental Equalization Transfers with National Standards”, in R.F. Geißler et al (eds.), Das Teilen beherrschen: Analysen zur Reform des Finanzausgleichs 2019 (Baden-Baden: Nomos, 2015) 313–332.

174

Ahmad and Brosio

The application of equalisation in terms of actual revenues and spending would mean that citizens in similar conditions, being members of the same group m, should have access, wherever they reside, to exactly the same quantity/quality mix of services and pay the same amount of taxes. Using the same terminology as before, the principle is represented by the following expression:

∑E ∑ t

1

c ,d ,e , f … j , n

R i w, y … n

m

= k for each local jurisdiction n 

(3)

Where, in addition to the previous notations, k is an equity parameter based on the distance between the targeted level of service provision and the revenue raised for its financing. Interjurisdictional equity is ensured by the equality of parameters k—one for each group of individuals—across all jurisdictions. This will lead to the result that individuals in comparable conditions, e.g. elderly people living alone, will be subject to the same proportional difference between what they receive in terms of healthcare and what they pay for it. Full equalisation implies that the transfer to each local government, Tn, is equal to the difference between: Tn = ∑ 1 E c ,d ,e , f ... j − ∑ i Rw, y … n  t

n

(4)

The average national value of k across all groups of individuals and all subnational governments also measures the vertical fiscal imbalance. We define the vertical fiscal imbalance as the share of local expenditure financed by local revenues. If actual spending and tax levels determine transfers for full equalisation, it is easy to see from equation (4) that these degenerate into filling local deficits, with all the disincentives associated with such a “gap-filling” approach. In practice, the disincentive effects are avoided if standards rather than actuals are used. The general principle then becomes to ensure that transfers to subnational governments provide “lump-sum” financing to ensure that people in comparable circumstances have access to comparable public services in all localities after paying comparable levels of taxes and fees. 2.3 Canada: Revenue Equalisation and Provincial Disparities Some sporadic ad hoc financial assistance has been provided since the start of the Canadian federation over a century ago to provinces that did not have enough tax resources to fund their expenditures. Those difficulties became more acute during the Great Depression, inducing the federal government to

Can Lessons from Equalisation Transfers

175

establish a Royal Commission on Dominion-Provincial Relations (the RowellSiros Commission). In its 1940 report, the Commission concluded that the provinces did not have sufficient taxing power to meet their constitutional responsibilities. The report considered a wide range of alternatives, rejecting any option that that would have reduced the policy responsibilities of the provinces or, alternatively, that would have given the provinces greater fiscal capacity. Instead, the Commission recommended the centralisation of taxation powers with the federal government with the guarantee of an annual income to the provinces and territories to be provided by the federal government. Quebec, nonetheless, instituted its own personal income tax (pit) system in the mid-1950s. Since the provinces had the constitutional right to levy direct taxes, the federal government feared that the move would trigger the creation of series of separate provincial income taxes.9 Therefore, in 1957, the federal government decided to transfer a share of each of the three so-called standard taxes to the provinces: 10 per cent of the pit, 9 per cent of the corporate income tax (cit) and 50 per cent of succession duties. These transfers were paid according to the derivation principle, meaning larger per capita transfers for the richer provinces. To correct this revenue inequality, the federal government launched the Equalisation Program, which, in its initial version, guaranteed that all provincial revenue from these shared taxes would be brought up to the per capita ­level of the average of the two richest provinces. The formula used for the allocation is shown in equation (5), where the two richest provinces set the standard. While Canada currently applies a revenue-based approach to equalisation, full equalisation as defined in equation (4) is written explicitly in Subsection 36(2) of the Constitution Act of 1982: Parliament and the government of Canada are committed to the principle of making equalization payments to ensure that provincial governments have sufficient revenues to provide reasonably comparable levels of public services at reasonably comparable levels of taxation. The Canadian system is as follows: Tn = t si × ( Bsi / P – Bni / Pn ) × Pn 

(5)

where: 9 T. Courchene, “A Short History of Equalization”, Policy Options Politiques, (2007) 22–27, http:// policyoptions.irpp.org/magazines/equalization-and-the-federal-spending-power/a-short -history-of-equalization/ (accessed 17 January 2016).

176

Ahmad and Brosio

TT is the total grant; Tn is the grant to province n; t is the tax rate; Bi is the tax base of each of the i revenue sources subject to equalisation; P is the population; Si is the standard for equalisation, e.g. the national average of total provincial revenues from each revenue source subject to equalisation, or the average of a group of provinces (as in Canada); and n represents beneficiary provinces, i.e. those for which the difference in the parentheses is positive. In turn: TT =

∑T

n



(6)

If the standard provinces get richer, e.g. following a huge increase in the price of the natural resources they exploit, the difference between them and the other provinces will increase, forcing the central government to expand the total amount paid for equalisation. A number of additional lessons can be drawn from the Canadian experience. It had been argued that, as there were no significant differences in the cost of service delivery levels across most Canadian provinces, it was sufficient to equalise across revenue capacity only (Representative Tax System, rts10). However, more recently, it has been recognised that service delivery costs are higher in the Northwest Territories than in the rest of Canada, and a special grant was constituted to offset the higher costs—implicitly recognising the relevance of need factors in an equalisation framework. Moreover, specialpurpose grants, especially for health care and education—essentially meeting central government objectives—are based on needs criteria, and their magnitude far exceeds the resources devoted to equalisation. Changing tax assignments and additional own-revenue powers for the Canadian provinces, which have constitutionally guaranteed access to all major tax bases with the obvious exception of foreign trade, call into question the scope of any vertical imbalance, as the provinces should in principle be able to compensate for any loss of central revenues by raising their own taxes. Furthermore, fixing the amount of the equalisation payments for a period of 10 years effectively replaces the rts. This is because changes in revenue ­bases are effectively ignored for a decade. Also, the treatment of the centre’s 10

The Representative Tax System combines the tax bases of the tax instruments used by the provinces to compile a composite index of provincial tax capacity. It obviates the problems derived from the use of different tax bases and tax rates by the provinces.

Can Lessons from Equalisation Transfers

177

significant “individual-based” programmes in an equalisation framework should also be evaluated, as these affect the sense of Subsection 36(2) of the Constitution referred to above. In general, equalisation over revenue capacities only might not be an appropriate option for large or disparate countries with significant differences in the cost of providing public services across their various jurisdictions. 2.4 Australia: Equalisation Based on Needs and Revenue Capabilities Australia applies a full equalisation framework based on revenue ­capacity and  spending needs. Unlike in Canada, the legal basis is less well defined, and depends to a large extent on agreements between the federation and the states. The interjurisdictional equity principle is not referred to in the Constitution. During the 1920s, the less populous states (Western Australia, Southern Australia and Tasmania) became less satisfied with the revenuesharing arrangements, and Western Australia voted to secede in 1933.11 The Commonwealth Parliament established the Commonwealth Grants Commission Act in May 1933 to create a permanent and independent body to assess Western Australia’s claims for secession. It is described as the “glue that holds the country together”, and the history of its evolution is described in Rye and Searle.12 The current Commonwealth Grants Commission’s (cgc) definition of the goal of equalisation transfers is as follows: State governments should receive funding from the pool of goods and services tax revenue such that, after allowing for material factors affecting revenues and expenditures, each would have the fiscal capacity to provide services and the associated infrastructure at the same standard, if each made the same effort to raise revenue from its own sources and operated at the same level of efficiency.13 The standardised expenditure for each function is determined by applying to the existing average per capita expenditure of the states for the various ­functions a number of parameters (“relativities”) that impact on the expenditure needed to provide the services at a level that is considered to be adequate. 11 12 13

T. Musgrave, “The Western Australian Secessionist Movement”, Macquarie Law Journal, 3 (2003) 95–129, 209. R. Rye and B. Searle, “The Fiscal Transfer System in Australia” in Ahmad, Financing Decentralization Expenditures, supra, 43–69. Commonwealth Grants Commission, Report on gst Revenue Sharing Relativities—2010 Review (vol.1, Canberra: Australian Government, 2010), 34.

178

Ahmad and Brosio

Using the symbols from equation (1), the Australian system can be described as follows:

∑ SE t

1

SR j

j

= k for each state 

(7)

where: SE is standardised expenditure, i.e., the amount of money that is needed to provide the same quality and quantity mix for each service assuming a national average rate of efficiency; SR is standardised revenue, i.e., the revenue that can be collected by applying the average national tax rate to the potential (not the assessed) tax base. Interjurisdictional equity and efficiency require that all subnational expenditures and all the revenue sources assigned to the subnational government be considered in the determination of the equalisation grant. In the past, the amounts available for equalisation transfers in Australia were determined annually on the basis of macroeconomic considerations and available resources. This, however, introduced an element of uncertainty for subnational governments. The most significant change in recent years was the facilitation of tax reforms by the use of the equalisation framework. The political agreement that led to the replacement of state-level sales taxes by a centrally administered vat (known as the Goods and Services Tax, gst, in Australia) was only made possible by an arrangement to transfer the entire amount of the vat collected to the states through the equalisation system. The cgc continued, however, as an independent body operating with the representation of both the states and the federal government, and determining horizontal allocations. The pool of funds based on vat collected by the central government also provided a more predictable basis for equalisation transfers, with greater certainty for the recipient subnational governments. This was the second such political economy linkage between significant tax reforms involving the vat and an equalisation system—the first was in China in 1993–1994. An issue that might militate against the further application of the Australian model is the complexity of the estimation of relative costs. Complexity reduces the transparency and attractiveness of the logic of the approach. A major simplification of the expenditure needs formulae and approach was undertaken in Australia in 2008 and significantly addressed the problem of opacity.14 14

In its Report on State Revenue Sharing Relativities, 2004 Review, Australia’s cgc recommended that work be undertaken to simplify the determination of its equalisation-based grants to state governments.

Can Lessons from Equalisation Transfers

179

The full expenditure needs and revenue capacity approach is also utilised in other advanced unitary states, including the United Kingdom, Denmark, Korea and Japan. In the Italian case (see below), equalisation transfers are derived from the transformation of the revenue-sharing mechanism. 2.5 Robin Hood Models: Germany Compared with Vertical Models There are two types of equalisation mechanisms: • the vertical equalisation model, such as the Australian and the Canadian systems, whereby grants are paid by the central government to the subnational governments; and • the horizontal equalisation model, such as in Germany, whereby grants are paid from relatively richer jurisdictions (Länder) to relatively poorer jurisdictions, without central government funding (Länderfinanzausgleich).15 This is a version of so-called “Robin Hood” redistribution, whereby richer jurisdictions provide transfers directly to poorer jurisdictions. The Chilean Fondo Común Municipal represents another example of a horizontal system,16 but it generates strong disincentives for subnational entities to raise their own revenues or manage spending efficiently. The German system does not have the same degree of disincentives, since grants are derived from shared taxes that are legislated by the federal government even if they are administered by the Länder. Leakages in revenue due to multiple tax subnational tax administrations remained a source of concern in the inconclusive discussions to reform the ­German intergovernmental system that took place from 2004 through 2006. In the vertical model, the skewness of the distribution of the revenues to be equalised influences the total amount of the grant. More precisely, in open-ended systems, such as in Canada, where there is no upper limit to the total amount 15

16

See, P.B. Spahn and O. Franz, “Consensus Democracy and Interjurisdictional Fiscal S­ olidarity in Germany” (2000) mimeo. See also P.B. Spahn, “Multi-level Finance and the Euro Crisis: The German Experience”, in Ahmad , Bordignon and Brosio, Multi-level Finance and the Euro Crisis, supra, 83–102. The abovementioned German equalization system will expire at the end of 2019. Starting in 2020, a new mechanism will come into force—a result of the agreement reached within the Bund-Länder intergovernmental conference on 14 October 2016. Among the changes, it is worth stressing the new role the distribution of vat revenue will play in ensuring the correction of horizontal imbalances. Furthermore, as part of the agreement, the federal Parliament has approved a constitutional revision on 13 July 2017 (amending articles 90, 91c, 104b, 104c, 107, 108, 109a, 114, 125c, 143d, 143e, 143f, 143g). E. Ahmad, L. Leterier and H. Ormeño, “Design of Transfers in Chile: Achieving Effective Service Delivery and Convergence of Opportunities”, Paper presented at the iv IberoAmerican Conference on Local Finance (Santiago de Chile, 2015).

180

Ahmad and Brosio

disbursed by the federal government, whenever the standard tax base—­ reflecting the potential revenues—increases, the total amount of the grant is bound to increase also, ceteris paribus. Thus, central government ­finances may be subjected to such a severe strain that the formula has to be changed. Vertical closed-end equalisation systems, such as in Australia after the introduction of the gst in 2001, minimise the fiscal strain on federal finances because the total pool of resources to be distributed is exogenous. However, when revenues from natural resources are included in the equalisation scheme and their distribution across areas is highly skewed, as in most cases, including, notably, Indonesia, Peru, Bolivia and other emerging economies, it may become more difficult to maintain effective equalisation mechanisms when commodity prices are high. And when prices drop precipitously, as during 2015 and 2016, there is a severe stress on local finances, and basic public service delivery may be jeopardised. This may, in turn, put pressure on national/federal finances if there is an implicit “political” floor on local public service delivery. Horizontal models better insulate the federal/national government finances. The degree of equalisation is based on the formula, and is not imperilled by sudden changes in the total amount of natural resource revenue and/or in the skewness of the distribution. A typical horizontal model formula based on the equalisation of tax capacity, which amounts to standardisation of revenues, would be: TT J = b J [t s (TB J – TBs )] 

(8)

TTI = bI [t s (TBs – TB I )] 

(9)

and

where, in addition to the previously mentioned symbols, bJI are the standards of equalisation applied to the paying and receiving jurisdictions, J are the paying jurisdictions, and I are the beneficiary jurisdictions. Thus, TTJ is the total grant paid by the contributing jurisdictions, and TTI is the total grant received by the beneficiary jurisdictions. A significant difficulty with the Robin Hood mechanism lies in the incentives it creates for contributing jurisdictions to reduce revenue collection so as to share less, and for recipient jurisdictions also to reduce their revenue effort and

Can Lessons from Equalisation Transfers

181

increase spending in order to improve their chances of getting more transfers. Moreover, the degree of equalisation depends on the standard that is adopted. If pressures from contributing regions lead to low standards, the degree of  equalisation reached with horizontal equalisation may also be low, leading to a widening of disparities in regional incomes and employment opportunities. Horizontal mechanisms are also extremely hard to introduce politically, except in extraordinary historical and political circumstances. This was the case in West Germany after World War ii, in China in conjunction with the major tax reforms in 1993/4, and in Indonesia after the “big bang decentralisation” of 1999/2000. 3

Keeping Countries Together and Generating Spatially Inclusive Growth

There are two significant trade-offs with equalisation systems. The first is between the need to keep countries together and the intensity of equalisation. This applies especially where there are significant differences in resource endowments, particularly, although not exclusively, in relation to natural resources, or generally in relation to a higher level of economic activity. These differences can exert a huge stress on national finances, as shown by the ­Canadian experience. The opposite response, as described and reflected in the experience of Indonesia, for example, or, for reasons not related to natural resources, in Italy, is for an asymmetric sharing of resources with or without a full equalisation system. The second major issue is related to equalisation and “spatially inclusive growth”. Here, the issues are quite complex. At a very basic level, the overall budget constraint for general government implies a trade-off between resources allocated for equalisation transfers and what might be assigned for public investments in physical and human capital. Connectivity constraints may make it essential for public investments to be allocated to a few “sustainable” hubs. The most striking example of this has been in China since 1993, with the development of the coastal belt accompanied by the migration of 150 million people from the interior. Connectivity, plus the availability of skills and supplies, predisposes the private sector to investing in “hubs”, and this has led to the creation of mega-cities and metropolises. This also underlies the growing importance of London, Milan and Santiago de Chile. Of course, the result in each case is the generation of increasing spatial inequalities. This is evident

182

Ahmad and Brosio

in China and Chile, between southern England and the rest of the United Kingdom and between northern and southern Italy. It is also clear that the existence of equalisation systems to ensure similar levels of public services at similar levels of revenue effort (at least in terms of setting sub-national tax rates, if not administration) are insufficient to address spatial inequality issues. Thus, despite the “equalisation systems” in place in China and the uk, regional disparities have widened. Conversely, the provision of public investment by the centre, e.g. in highspeed trains and motorways, may also be insufficient to stop the generation of spatial imbalances, and may indeed increase them, as workers find it easier to commute to existing “hubs”. The rebalancing of public investment to ­generate new “hubs” in depressed regions may not succeed without corresponding improvements in local service delivery to enable the private sector to take ­advantage of the levelling of the “playing field”. This involves a judicious combination of equalisation transfers, as well as centrally or locally-financed public investments. The dynamic interactions between equalisation transfers and the allocation of funds for investment are critical for ensuring the success of both distinct policy tools. 3.1 From Forced Uniformity to Asymmetric Arrangements in Italy The trade-offs between central transfers, differentials in service delivery and differential costs of living are clearly seen in Italy, where the historical context is of considerable interest, as many of the conditions and issues seen there mirror those currently being witnessed in emerging markets. Italy17 was unified in 1861, as the Kingdom of Savoy annexed distinct states on the rest of the Italian Peninsula. Each state had its own systems of taxation and local government. At the time of the establishment of the country, the central and northern states had much higher levels of local (and national) taxation than those of the south. The gap in local taxation was explained mainly by the differences in the expenditure assignments existing before the annexation. The obvious consequence of the local taxation gap was that the level of municipal expenditure (as a percentage of gdp) in 1860 was about three times higher in the north than in the south. Local government financing problems were made more acute by the ­decision—one of the first taken by the new state—to assign primary education to municipalities. This was a typical national government policy of the 17

The illustration of the Italian case draws on P. Giarda, Le dinamiche dei rapporti finanziari tra Stato e Autonomie locali in Italia: una prospettiva storica (2014), www.cattolicanews.it/ Giarda_prolusioneaa2013_2014.pdf (accessed 14 January 2016).

183

Can Lessons from Equalisation Transfers

era (like in England, for example), where local governments were considered mostly as branches, if not pure appendices, of the central government, which wanted to divest itself of the burden of certain expensive functions in order to concentrate its resources on strategic sectors, such as, primarily, wars. The focus of Italy’s policy was on ensuring that every local government ­implemented national mandates, hence making service provision uniform. In the Italian narrative, the country “still had to be unified”, and locally provided services, being a main component of government policies, had to make their contribution. The idea of assisting the local authorities of the south with transfers was never considered at that time. Instead, the burden of national and local taxes was quickly and abruptly increased, giving rise to violent protests and local revolts that were labelled as brigandage.18 In just 10 years, from 1860 to 1870, the gap in per capita local tax revenue between the south and the centre-north was reduced by almost 50 per cent (see Table 8.1). More precisely, while, in 1860, per capita tax revenue in the south was only 32 per cent of that of the north, this percentage had come close to 58 per cent by 1870. The reduction in the gap between the two areas continued until 1889. This heavy increase in local taxation was hard to sustain as far as the poorer southern taxpayers were concerned. Two landmark acts provide a very accurate illustration of the national government’s policy towards municipalities. The first is the Law on Roads of 1868, which called on municipalities to extend and improve their road networks. The law also mandated that the huge financial cost necessitated by its implementation had to be covered with surcharges on local taxes, tolls on new roads, Table 8.1

Per capita tax revenue of Italian local governments 1860–1936a

Centre-North South Ratio of South to Centre-North Total

1860

1870

1876

1889

1912

1925

1936

5.75 1.88 0.33

7.99 4.61 0.58

9.08 5.87 0.62

12.21 8.96 0.73

17.36 9.24 0.53

94.70 112.81 45.30 56.73 0.48 0.50

4.03

6.56

8.03

10.93

14.22

76.36 n.a.

a In Italian Lira Source: Giarda, Le dinamiche dei rapport finanziari, supra.

18

See E. Dal Lago, “Italian National Unification and the Mezzogiorno: Colonialism in One Country?” in R. Healy and E. Dal Lago (eds.), The Shadow of Colonialism on Europe’s Modern Past (London: Palgrave, 2014) 57–72.

184

Ahmad and Brosio

contributions from the landowners of affected estates, loans and, finally, with central government support. The latter turned out to be very modest. The second law, called the Health Reform Act, was passed in 1888 and referred specifically to water, sanitation and local (family) doctors. Again, the financial burden imposed on municipalities was high, and the central government’s contribution was kept to a minimum. Also very revealing was the law’s mandate (initially introduced in 1859 in the Kingdom of Savoy and then extended to the Italian state), which established free universal primary education and assigned it to municipalities. Municipalities had to “provide services according to the needs of their inhabitants and in proportion to their financial means”, which more or less was like dictating that municipal governments had to square the circle. This resulted in extreme disparities in the level of service provision among municipalities, as shown by large disparities in the incidence of illiteracy.19 To reduce disparities and alleviate the burden on municipalities, the central government intervened, increasingly supplementing the payment of teachers’ salaries. First, in 1876 with the Legge Coppino, it contributed to the direct payment of statutory salary increases. Then, in 1886, it intervened with the payment of a specific grant to finance the cost of a new statutory salary increase. As we can see, the intent of these interventions was not redistribution, but rather the contrary. Richer municipalities were able to hire more teachers and were getting greater help from the national government to pay their salaries. Finally, in 1904, the national government took it upon itself to pay the entire salary bill for all teachers. To assuage the discontent of taxpayers and municipalities in the south, the national government introduced, in the last decades of the 19th century, a limited amount of financial support by relieving municipalities of a share of the service of debt raised for the financing of public works. The programme started with the Sicilian municipalities, continuing with those of Campania, including then the entire south and, finally, extending its reach to the whole of the country. The consequences of these policies on spatial inequality are easy to see. At the time of the creation of the country, the economic gap between the centrenorth and the south was small. According to Felice and Vecchi, it did not exceed 15 per cent in terms of per capita gdp.20 In the last decade of the 19th 19

20

E. Felice and G. Vecchi, “Italy’s Modern Economic Growth, 1861–2011”, Quaderni del Dipartimento di Economia Politica e Statistica—Università di Siena, 663 (2012), www.econ-pol .unisi.it/quaderni/663.pdf (accessed 19 January 2016). Ibid.

Can Lessons from Equalisation Transfers

185

century, however, following the start of the industrialisation process in Italy, this gap started to grow very quickly. Its impact on the finances of the municipalities was immediate. The ratio of the local tax revenue of the south to that of the north declined by 20 percentage points between 1889 and 1912 and never recovered thereafter. The draft of a revised law on local government presented by the government to Parliament in 1930 mentioned the need to introduce equalisation transfers to municipalities on the basis of approved deficits. The government rejected the idea of allocating transfers based on indicators of need and revenue. The reason was that the introduction of “complicated automatic allocation criteria based on population, and tax capacity” would have impacted negatively on “the possibility of controlling the dynamics of these funds”.21 In the aftermath of World War ii, attempts to reduce disparities were resumed. A programme based on the allocation of capital grants to all municipalities was introduced in 1949. In 1952, a general transfer fund fed by a share of the proceeds from the turnover tax was created. The key for the distribution was population. The origin of the current system of equalisation transfers is, however, somewhat different. It is to be found in the creation during World War ii of a fund for paying contributions aimed at making up for the deficits facing the municipalities most affected by the war. Individual transfers—consisting of permission to raise loans that would eventually have to be repaid by the central government—were negotiated between the central government and the concerned municipalities. This obviously led to interference on the part of the central government in the local decision-making process. The programme was renewed and extended until the tax reform of 1970, which led to the elimination of practically all local taxes, and their substitution with a system of revenue-sharing for equalisation purposes. The asymmetric arrangements that were introduced after World War ii were designed to keep the country together and to prevent peripheral regions, which were assigned a special autonomy status, from seceding.22 Uniformity was sacrificed to keep the country together. General transfers for equalisation purposes replaced revenue-sharing for intermediate/regional and municipal governments. Structural funds were used to supplement the capital transfers described above for infrastructure and connectivity, but failed to offset the 21 22

Commissione Parlamentare per la riforma della Finanza Locale, Relazione e Schema di proposte (Roma: Tipografia del Senato del dott.G. Bardi, 1931) Translated by the authors. G. Brosio, “Moving toward a Quasi-Federal System: Intergovernmental Relations in Italy”, Journal des Économistes et des Études Humaines, 13 (2003) 1–26.

186

Ahmad and Brosio

­diverging tendencies. As pointed out by Felice, continuing differences in spatial gdp are due to employment opportunities rather than the p ­ roductivity patterns of many regions.23 The divergence between the centre-north and the south instead follows social and institutional differences—in the effective working of political and economic institutions and in the levels of social ­capital—that originated in preunification states and did not die (but in part even increased) in post-unification Italy. Equalisation to Supplement Tax Reforms for Structural Change in China China represents an excellent case study, where a “package” of fiscal reforms was undertaken in 1993–1994 to consolidate the structural changes that had taken place in the preceding decade, and which provided the springboard for the country’s spectacular growth performance in the two decades since then, and the most significant reduction in poverty in recent history (700 million people out of a global total of 750 million people taken out of poverty in the past two decades). As described in Ahmad et al., China’s 1993–1994 tax reform was critical for consolidating the structural change that had begun a decade earlier by Deng Xiao Ping, which allowed enterprises to retain profits.24 This had led to a precipitous drop in the tax/gdp ratio from around 30 per cent in the early 1980s to around 10 per cent in 1992, as well as a drop in the share going to the central government (from around a half to a quarter of total collections in the same period).25 It is important to note that all the administration was at the provincial level—and China, like many other medieval Asian empires, practised a version of upward revenue-sharing. This broke down as the total take declined, and provincial and local governments faced increasing revenue constraints and reduced incentives to share funds with higher levels. The crux of the reform was to establish a central tax administration and manage a new vat, as well as other major taxes. But this required a careful

3.2

23

24 25

E. Felice, “Regional Income Inequality in Italy in the Long Run (1871–2001): Patterns and Determinants”, Unitat d’Historia Econòmica, Working Paper 2013_08, Universitat Autònoma de Barcelona (2013). E. Ahmad, J. Rydge and N. Stern, “Structural Change Drives Tax Reforms Drives Structural Change”, lse Paper, presented at the China Development Forum, Beijing (2013). E. Ahmad, “Taxation Reforms and Sequencoing of Intergovernmental Reforms in China: Preconditions for a Xiaokang Society”, in J. Lou and S. Wang (eds.), Public Finance in China: Reform and Growth for a Harmonious Society (Washington, d.c.: The World Bank, 2008) 95–128.

Can Lessons from Equalisation Transfers

187

balancing of winners and losers in order to obtain agreement from the provinces for a loss of tax administration powers. Critical in this agreement were: • a lump-sum transfer equivalent to revenues in 1993—a stop-loss provision to prevent any province from losing in absolute terms; • a share of the new tax (the vat) on a derivation basis—clearly benefitting the more advanced provinces; • offsetting the above was a new equalisation transfer—this was a simplification of the Australian model with simple factors for which information was readily available. This was not implemented fully in 1994, however, but was phased in over time instead; and • a substantial “revenue-returned” component. This was critical in enabling the coastal provinces to establish “hubs” with appropriate connectivity that permitted the operation of a rapid growth scenario for the whole country. As pointed out by Felice, in the case of Italy, the rapid growth scenario has been critically dependent on the provision of jobs by the private sector.26 The Chinese coastal hubs did just that, with 170 million people migrating to the coast in the past 20 years. This enabled the maintenance of full employment and the poverty reduction referred to earlier. Indeed, had the equalisation system worked fully from 1994 onwards, there would have been much less investment in the more advanced coastal areas—the result would have been much slower growth overall and significantly less employment generation and poverty reduction. The downside of this “success story” has been the creation of mega-metropolitan areas along the coast (Guangzhou-Shenzhen in the Pearl River Delta is now home to 50 million people, making it the largest conurbation in the world), leading to congestion and pollution. Also, the existence of an equalisation system has not prevented increasing spatial and interpersonal inequalities—reflecting the patterns in Chile, for instance—as private producers find it easier to operate from established hubs with a ready supply of skilled labour. It is also clear that, going forward, a “rebalancing” in China towards new hubs is now feasible given the investment in connectivity that has taken place with high-speed train networks and motorways and the establishment of modern airports in the interior. However, as in the Italian case, the physical investment is not sufficient—it has to be accompanied by more effective local governance, and this depends to a large extent on appropriate own-source revenues, access to credit, plus a more effective within-province equalisation system (note that 26

Felice, “Regional Income Inequality”, supra.

188

Ahmad and Brosio

Chinese provinces are on average bigger than most European countries). These issues are illustrated for the province of Guangdong in Ahmad et al.27 3.3 Indonesia Indonesia achieved fairly substantial poverty reduction and enhancements in its service delivery under the “dirigiste” Suharto regime. However, the centralised imposition of national standards led to considerable political and local reactions, and Indonesia’s unity was jeopardised as a result (East Timor actually seceded in the late 1990s). Local demands for greater sharing of Indonesia’s abundant natural resources (oil and gas, as well as minerals and forestry) formed part of the post-Suharto political compact needed to keep the country together. In order to prevent a macroeconomic problem, there was also a significant devolution of functions to local governments. In order to complete the financing and prevent regional imbalances (natural resources are, also in Indonesia, very unevenly distributed), however, the country had to introduce an equalisation system.28 Again, this was modelled on the Australian system, even if the system of own-source revenues was very weak in Indonesia— shared revenues had to be “equalised”. In due course, the uniform revenue-sharing arrangement was not acceptable to two regions threatening to secede: Aceh and East Kalimantan. As in the Italian case, an asymmetric arrangement was finally agreed, which gave these regions a very much higher portion of shared revenues than other regions. In order to prevent the emergence of widely different standards of living, the equalisation formulae were adjusted from the standardised format to take into account the actual shared revenues received. This, however, completely changed the incentive structures facing the local governments, and there has been relatively little success with improvements in service delivery since then. Again, the parallels with the political economy of revenue-sharing and asymmetric arrangements in Italy is instructive. However, both countries now face the dilemma of introducing better local governance and balancing equalisation tendencies against the need for “hubs” to generate growth and employment.

27 28

E. Ahmad, J. Rydge and N. Stern (eds.), Fiscal Underpinnings of Sustainable Development in China: Rebalancing in Guangdong (Singapore: Springer, 2017). E. Ahmad and A. Mansoor, “Indonesia: Managing Decentralization”, in E. Ahmad and V. Tanzi (eds.), Managing Fiscal Decentralization (London: Routledge, 2002) 306–320.

Can Lessons from Equalisation Transfers

4

189

Concluding Comments

Equalisation transfers make it possible to achieve uniformity of subnational service provision without impinging on interjurisdictional equity, i.e. they allow people in comparable circumstances to have access to comparable public services in all localities. Equity implies that residence should not create ­differences among citizens in their access to public services or to the cost of access. Evidence shows that equalisation and transfer design have resulted, particularly in industrialised countries, in improvements in access to service delivery. Equalisation transfers have had, and still have, an important role to play in keeping disparate countries together and in addressing centrifugal tendencies in both federal and unitary states alike. The presence of an equalisation tool has also been useful in facilitating agreement on major tax reforms and reassignments of responsibilities. We have illustrated these points with comparative examples from advanced and emerging markets alike. At the same time, full equalisation can be costly, especially when the economic and social gaps between the various areas of countries are very wide. And if financed by taxing richer regions, there can be an incentive to secede, as in Catalonia. The equalisation task can become impossible when spatial gaps are exacerbated by the assignment of revenue from natural resources to subnational governments. Financial constraints explain, among other factors, why countries have taken a gradual approach to the introduction of equalisation transfers, as the experience of Italy shows. Full equalisation implies consideration of both revenue disparities and expenditure needs, as most industrial and emerging countries are attempting to do. When the costs of provision do not largely differ between the various areas, equalisation of fiscal capacity may be a reasonable solution, although it only partially satisfies the principle of interjurisdictional equity. Another solution is to introduce asymmetric assignment of revenue for “problem” areas, as is done for peripheral regions in Italy and for oil-producing areas in Indonesia, and to focus on full equalisation in the rest of the country. However, the issue of interpersonal equality and lagging regions remains, whether one focuses on Italy, the United Kingdom, Chile, Australia, Indonesia or China. Here, the trade-offs between resources for equalisation and investment in “sustainable growth hubs” remains an issue of important policy and research significance.

chapter 9

Fiscal Decentralisation and Decentralising Tax Administration: Different Questions, Different Answers Richard M. Bird 1

Introduction

Three critical aspects of any government fiscal system are (1) revenue policy— what taxes (and other charges) are imposed; (2) revenue administration—how taxes are administered and collected; and (3) revenue management—how the revenue obtained is spent and who gets what. Most analytical attention has perhaps been focused on issues of tax policy, while most public attention has probably been paid to the endlessly controversial questions of who pays what and who gets what. In some key respects, however, the critical link between the decision about how to raise public sector revenue and how to spend it is tax administration—the organisation and management of the process of assessing, collecting and enforcing taxation (and other current revenue).1 One important aspect of tax administration that has received some attention recently is the extent to which tax administration should be decentralised. Should there be one central revenue administration or should every government in a country have its own revenue administration?2 Different countries * Key words: fiscal decentralisation; tax administration; Canada; China; Germany; Spain jel Codes: H7, H83, P35, P43, O57. Some of the material in Section 4 is drawn from a forthcoming joint work with Roy Bahl, although I am solely responsible for its use here. 1 The related but distinct questions of financing capital spending or, for that matter, deficits are not considered here, and non-tax current revenues such as fees and charges are also left aside. Slemrod and Gillitzer provide an analytical treatment of tax systems that pays close attention to tax administration, but they do not consider the decentralisation question that is the focus of this paper, J. Slemrod and C. Gillitzer, Tax Systems (Cambridge, ma: The mit Press, 2014). 2 For excellent previous reviews, see C. Vehorn and E. Ahmad, “Tax Administration”, in T. Ter-Minassian (ed.), Fiscal Federalism in Theory and Practice (Washington: International Monetary Fund, 1997) 108–134; J. Mikesell, “Developing Options for the Administration of Local Taxes: An International Review”, Public Budgeting & Finance, 27 (2007) 41–68;

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_011

Fiscal Decentralisation & Decentralising Tax Administration

191

have reached very different answers to this question. In Denmark, for example, central government and local government tax administrations were merged in 2005 and the unified administration was further restructured in 2013 when the regional structure was completely abolished.3 However, Denmark is both a unitary country and an unusually unified one with relatively little evidence of significant regional or local differences in preferences.4 In contrast, in some federal countries, notably the United States and Switzerland, most governments at every level are responsible for administering most or all of their own taxes. In other developed federations, however, although state, regional and local governments usually have their own tax administrations, some state taxes are often actually administered at the federal level and some local taxes largely determined at the state level.5 In Australia, for example, all the revenues of the Goods and Services Tax (gst), which is administered by the federal government, go to the state governments. However, because the gst is imposed by federal law at a uniform rate throughout the country, from most perspectives it is not really a state tax but rather a form of “tax sharing” equivalent to an intergovernmental transfer distributed on the basis of estimated collections in each region.6 On the J. Martinez-Vazquez and A. Timofeev, “Choosing between Centralized and Decentralized Models of Tax Administration”, International Journal of Public Administration, 33 (2010) 601– 619. A related issue, not discussed further here is the extent to which a central tax administration itself should operate in a decentralised fashion. A number of countries in recent years have undertaken large-scale rationalisation of their territorial office networks, recognising the greatly increased role of information technology and especially the internet in reaching and communicating with taxpayers and reducing substantially the number of local tax offices. 3 oecd, Comparative Tax Administration, 2015 (Paris 2015) 81. 4 Of course, there remain some continuing tensions between central and local governments even in Denmark, J. Løtz et al., “The Changing Role of Local Income Taxation in Denmark”, in J. Kim et al. (eds.), Interaction between Local Expenditure Responsibilities and Local Tax Policy (Korea Institute of Public Finance and Danish Ministry for Economic Affairs and the Interior, 2015) 231–251. 5 Each country has its own governance structure and often its own terminology. In the us and Australia, the intermediate-level governments are states and lower-level (general) governments are local; in Canada, they are provinces and municipalities; in Switzerland, cantons and communes; and so on. For simplicity, however, we often refer here to intermediate (regional) governments as states or regions and to lower-level local governments (of which there are often several varieties and even levels) as local governments. 6 R.M. Bird and M. Smart, “Assigning State Taxes in a Federal Country: The Case of Australia”, in Melbourne Institute, Australia’s Future Tax and Transfer Policy Conference (Melbourne: Melbourne Institute of Applied Economic and Social Research, 2010) 72–94.

192

Bird

other hand, in Canada, where the main regional revenues are income and sales taxes, the federal government currently administers nine (out of 10) provincial personal income taxes, as well as eight provincial corporate income taxes, and five provincial sales taxes. Nonetheless, since in all cases the provinces are responsible for setting the rates and, within some limits, can also alter the base of the tax, all these federally-administered taxes are essentially provincial taxes.7 More unusually, one province in Canada (Quebec) administers not only its own sales tax but also the federal sales tax. Germany takes this much further, with the states being responsible for most aspects of administering federal taxes, as discussed further in Section 3 below. In Latin America, some local taxes are administered at the national level in Argentina, Brazil, Chile, Dominican Republic, Guatemala, Nicaragua, Panama and Peru.8 In Mexico, although the states and the Federal District (Mexico City) have ceded most tax authority to the central government, the agreements providing for tax coordination and administrative cooperation with the federal government empower the states to collect and audit federal taxes. In practice, however, the highly centralised State Tax Administration (sat, Servicio de Administración Tributaria) seems to be in full charge.9 Even when states do collect taxes as with respect to a simplified tax on small business they appear to do a poor job, with widespread evasion.10 To a considerable extent, however, 7

8

9

10

R.M. Bird, “The gst/hst: Creating an Integrated Sales Tax in a Federal Country”, in J. Mintz and S. Richardson (eds.), After Twenty Years: The Future of the Goods and Services Tax (Toronto: Canadian Tax Foundation, 2014) 1–50. Inter-American Development Bank, “State of the Tax Administration in Latin America: 2006–2010” (2012), http://www.ciat.org/index.php/en/products-and-services/publicat ions/books/2539-estado-de-la-administracion-tributaria-en-america-latina-2006-2010 .html (accessed 20 November 2015), Table 9.5. J.L. Velasco, “Servicio de Administración Tributaria de México”, Institutions and Development Conference Papers, n. 08–05k, Center for Migration and Development, Princeton University (2008), www.princeton.edu/cmd/working-papers/idlac08/wp0805k.pdf (accessed 18 November 2015). Most of the transitional countries of central and eastern Europe also have a centrally-controlled tax administration, with regional governments sometimes having no tax administration role at all although local governments often collect some revenues, especially property taxes. Hőgye provides an early overview, M. Högye (ed.), Local and Regional Tax Administration in Transition Countries (Budapest: Local Government Public Service Reform Initiative, 2000). For a recent examination of the situation in Poland, P. Swianiewicz and J. Łukomska, “Local Tax Policies in the Limited Autonomy of the Revenue Collection System in Poland”, in Kim et al., Interaction between, supra, 255–280. V. Fretes Cibils and T. Ter-Minassian (eds.), Decentralizing Revenue in Latin America: Why and How (Washington: Inter-American Development Bank, 2015); E. Ahmad, “Multilevel

Fiscal Decentralisation & Decentralising Tax Administration

193

this failure partly reflects the perverse incentives arising from the high level of dependence on poorly-designed federal transfers.11 In Argentina, where much the same happens, Piffano argues that the present system of tax sharing should be replaced by a combination of more coordinated and centralised administration of most important sub-national taxes (for example, establishing a uniform national cadastre and automobile register) and giving provinces more autonomy in setting tax rates (as in Canada).12 As a final example, local taxes are sometimes administered at least in part by the next higher level of government in many countries. In both Australia and Canada, for example, the main local tax is the property tax, but the base of that tax, the assessed value of real property, is often set by the state government. In the United States, local sales taxes are administered by 38 of the 45 states which have sales taxes of their own.13 In contrast, in Sweden and other unitary Nordic countries, although the main local tax is the income tax and localities usually have some power to vary the tax rate, the tax is administered solely by the central revenue administration. The extent to which taxes for which one level of government is formally politically responsible are administered by other governments does not seem related in any fixed way either to the formal constitutional structure of a country or its level of development. What countries do with respect to decentralising tax administration—though no doubt often reflecting the inertia of past choices made in response to changing political institutions, technology, and economic conditions—appears to be largely up to them. Before discussing

11

12 13

Fiscal Institutions and Mechanisms for Reducing Tax Cheating: The Case of Mexico”, in J. Kim and H. Blöchinger (eds.), Institutions of Intergovernmental Fiscal Relations: Challenges Ahead (Paris: oecd, 2015) 239–254. For a more positive view of a somewhat similar system in Ethiopia, which is argued to increase both the fairness of taxation and state legitimacy, see A. Joshi et al., “Taxing the Informal Sector: The Current State of Knowledge and Agendas for Future Research”, Journal of Development Studies, 50 (2014) 1325–1347. As Weingast argues, local political autonomy in terms of elections and making spending decisions may, in the presence of fiscal dependence, all too easily become a means of political control rather than citizen choice. Requiring (and enabling) sub-national governments to be responsible for raising their own revenues is thus one important way to make political autonomy real, B. Weingast, “Second-Generation Fiscal Federalism”, Journal of Urban Economics, 65 (2009) 279–293. However, transfers must also be properly designed, see; R.M. Bird and M. Smart, “Intergovernmental Fiscal Transfers: International Lessons for Developing Countries”, World Development, 30 (2002) 899–912. H. Piffano, “Fiscal Decentralization, Tax Competition, and Federal Tax Administration: A Note from the Argentine Experience”, Revista de Economía y Estadística, 43 (2005) 7–28. S. Drenkard, “State and Local Sales Tax Rates in 2014”, The Tax Foundation, http://taxfoun dation.org/article/state-and-local-sales-tax-rates-2014 (accessed 20 November 2015).

194

Bird

further the question of how decentralised taxes are or should be administered, however, Section 2 first touches briefly on the surprisingly complex question of how one may characterise a tax as being decentralised in the first place. Section 3 then provides a more detailed account of how decentralised taxes are administered in practice in several very different countries. Section 4 discusses the main key arguments for and against decentralising the administration of decentralised taxes. Section 5 concludes. 2

What is a “Decentralised” Tax?

Is a decentralised tax: (1) One where regional or local governments have the power to decide whether to impose it or not? (2) One where they determine the tax base? (3) One where they set the tax rate? (4) One where they determine the liability of particular taxpayers? (5) One where they collect and enforce the tax? (6) One where they receive the revenue? (7) Or one where certain combinations of the above conditions exist? As Blöchliger and King discuss in detail, there are many possible ways to “mix and match” these characteristics.14 They distinguish 13 different possibilities in the 30 oecd countries, taking into account only the degree of discretion over tax policy decisions and not such questions as whether and how more than one level of government controls some aspects of administration. Whether a government controls its own revenue sources in a meaningful way is critical no matter how one measures control. As Martinez-Vazquez, McLure and Vaillancourt note, if fiscal decentralization is to be a reality, subnational governments must control their own sources of revenue. Subnational governments that lack independent sources of revenue can never truly enjoy fiscal autonomy, because they may be—and probably are—under the financial thumb of the central government.15 Simply having a constitutionally guaranteed share of central revenues may seem to guarantee autonomy in this sense. But spending autonomy financed from on high is much less likely to be spent as local citizens would wish than 14 15

H. Blöchliger and D. King, “Less than You Thought: The Fiscal Autonomy of Sub-Central Governments”, oecd Economic Studies, 43 (2006) 155–188. J. Martinez Vazquez et al., “Revenues and Expenditures in an Intergovernmental Framework”, in R.M. Bird and F. Vaillancourt (eds.), Perspectives on Fiscal Federalism (Washington, dc: World Bank, 2006) 15–34.

Fiscal Decentralisation & Decentralising Tax Administration

195

if they had to find the funds in their own pockets. History and theory suggest strongly that for local governments to be responsibly autonomous they must be responsible to local citizens for how they raise as well as spend revenues. Democratic local elections and transparent—and preferably comparative— reporting to citizens about local spending patterns are two obvious ways to move towards this goal.16 But even the most democratic local government is likely to spend more responsibly if it also bears some responsibility for raising revenue by imposing taxes on residents in a visible and accountable way.17 The single most important factor ensuring that sub-national governments are accountable to their citizens is probably to make them clearly and visibly responsible for determining tax rates. The tax rate is for most people the most visible and understandable characteristic of any tax.18 The more power regional and local governments have in terms of collecting revenue—choosing which taxes to impose, how the tax base is defined, and actually assessing and collecting the tax—the greater their fiscal autonomy. But without the ability to establish and alter tax rates, even if only within some limits, the transparency and accountability of the local revenue system is likely to fall short of what is needed to support the economic case for fiscal decentralisation.19 Provided regional and local governments can meaningfully establish tax rates, it does not seem necessary for them to administer taxes themselves in order to operate effectively and efficiently. There may be good reasons why some sub-national taxes should be administered locally because it is efficient to do so and because doing so clearly adds to local autonomy which may be considered desirable in itself. But fiscal and administrative decentralisation, while related, are distinct: one may have much of one with little or none of the other, as shown by the case studies next discussed. 3

Tales from Four Countries—All Decentralised and All Different

Four brief case studies of fiscal and administrative decentralisation are set out in this section: Canada, China, Germany, and Spain. A few basic parameters 16

17 18 19

R.M. Bird, “Fiscal Decentralization and Competitive Governments”, in G. Galeotti et al. (eds.), Competition and Structure: The Political Economy of Collective Decisions (Cambridge: Cambridge University Press, 2000) 129–149. This is one of the main tenets of the so-called “second generation” theories of fiscal federalism, Weingast, “Second-Generation”, supra. C. McLure, “Tax Assignment and Subnational Fiscal Autonomy”, Bulletin for International Fiscal Documentation (2000) 626–635. W.E. Oates, Fiscal Federalism (New York: Harcourt Brace Jovanovich, 1972).

196

Bird

Table 9.1

Setting the stage: The structure of tax administrationa

Number of Employees Number at hq (per cent of total) Number of hq offices Number of regional offices Number of local offices Citizens/fte tax staff Administrative cost/net revenue collections

Spain

United States

756,000 110,494

26,231

86,977

8,897 (23.3) 1

800 (0.1) 1

1,236 (1.1) 17

3,416 (13.0) 1

4,073 (4.7) 1

5

71

12

56

139

39

6,667

546

227

119

921

n.a.

743

2,081

3,635

1.16

n.a.

1.35

0.67

0.47

Canada

China

38,172

Germany

a Canada—number of employees excludes 1,359 dealing with non-tax functions; China— regional offices include provincial offices and those in municipalities directly under central government, while local offices include those at municipal level and county and district offices directed by provincial offices; Germany—hq include 16 Länder administrations. There are also 34 call centres, 28 of which are located in one state (Hesse); cost ratio excludes social security contributions and excises; Spain—includes 17 regional offices, 39 provincial offices, 192 local tax offices and 31 customs and excise local offices; costs include customs but exclude social security contributions; United States—no major federal indirect tax; the regional-local breakdown is approximate; cost ratio differs from official irs figures, which are based on gross rather than net collection. Source: oecd, Comparative Tax Administration 2015 (Paris 2015), Tables 2.3; 2.4; 5.4; 5.6

with respect to the tax administration structures in these countries (plus the United States, for comparative purposes) are set out in Table 9.1. The main conclusion suggested by this table is simply that each country is very different in some respects from the others both in how centralised its tax administration is and in terms of costs of collection and the citizen/staff ratio.20 20

While numbers are provided for these two ratios where available, as oecd explains, extreme care is needed when comparing these numbers owing to the very different situations found in different countries, oecd Comparative Tax Administration, supra.

Fiscal Decentralisation & Decentralising Tax Administration

197

To provide a base case that may be familiar to some, data for the United States are also shown in Table 9.1. Although the federal tax administration in the u.s. appears in relative terms to be relatively efficient and decentralised— only 4.7 per cent of employees at headquarters (hq), and costs (0.47 per cent) are relatively low and the citizen/staff ratio (3,635) relatively high—these factors of course reflect in part the fact that there is, uniquely, no major federal indirect tax in the u.s.. The u.s. as a whole, with 536,312 public employees at the federal, state and local levels concerned with financial administration in 2013,21 seems comparable in scale only to that of China. However, unlike China (or any of the other countries discussed here), there is effectively no central direction or coordination of the very different tax systems and tax administrations found at the state and local level in the United States. Those who want to see what a truly decentralised tax administrative system looks like need look no further. 3.1 Germany In contrast, Germany provides an example of what may perhaps be called a highly centralised decentralised system. Many taxes in Germany are constitutionally assigned to one level of government—federal, state (Länder), or local (communities). However, most important taxes, accounting for about threequarters of total revenue, are “common” and are shared between two or all three levels of government in specific percentage shares. Such tax-sharing arrangements are not uncommon in federal countries. What is considerably less common, however, is the way in which taxes are administered in Germany. The federal tax administration administers only customs duties and some federally regulated excises (including the beer duty although all the yield of this tax goes to the states).22 All other taxes are administered by the states, although they often delegate the administration of such specifically local taxes as the real property tax to local governments.23 State tax administration offices are thus 21

22 23

United States Census Bureau, “Categories of Employees at Federal, State and Local Levels: March 2013”, 2013 Annual Survey of Public Employment & Payroll, http://factfinder.census .gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk (accessed 4 November 2015). The federal administration is also responsible for administering any levies imposed on behalf of the European Union. Interestingly, however, the state administrations determine the assessed value of real property, which serves as the base for both the local property tax and the federal inheritance tax. There is considerable dissatisfaction at all levels of government with the present state of property valuation in Germany but no consensus on how to reform the system, see G. Färber et al., “Property Tax Reform in Germany: Eternally Unfinished?” in Kim et al., Interaction between, supra, 167–201. Assigning the assessment task to a higher-level government is common in many countries in which property taxes are

198

Bird

both federal and state offices, with the federal government assuming the costs related to administering federal taxes. Not only do German states administer almost all taxes, they can also decide how to do so. The main rule governing state administration is simply the basic constitutional rule that taxes must be administered uniformly (apart from some very small differences in local taxes in different communities). In most cases, there are regional tax offices that house both federal and state tax administrations and have a common president who is partly a federal and partly a state official. However, in Bavaria the state tax department is completely separated from the federal administration, and in other states there are sometimes specialised tax offices and no or more than one regional office. States can finance tax administration how and to the extent they wish and may also choose the extent to which and how they adopt new administrative technology such as specialised software.24 On the other hand, all tax rates are uniformly established at the national level. Moreover, the training of tax officers is also specified by national law and states are required to establish training institutions—although it is left to them to decide how to do this. Finally, as is true for all public servants in Germany, the basic wages of tax officers (though not necessarily their hours or bonuses) are set at the national level and are uniform throughout the country. The German system is thus unique. On one hand, the states have no autonomy in setting tax rates and only limited control over tax officials. On the other, they have complete autonomy in deciding how to organise and administer both state and most important shared taxes (e.g., with respect to auditing efforts).25 Furthermore, there is no formal central authority to coordinate state tax administrations, although a complex network of boards, committees and work groups (e.g., to assess the performance of local tax offices) exists, with ­different states participating to different extents in coordinating different

24

25

e­ ssentially local, see R.M. Bird and E. Slack, International Handbook of Land and Property Taxation (Cheltenham: Edward Elgar, 2004). Ulbricht describes the uneven and non-uniform way in which information technology was introduced at the state level. A. Ulbricht, “The Decentralization of Tax Administration in Germany: Consequences”, in N. Bosch and J.M. Durán (eds.), Fiscal Federalism and Political Decentralization (Cheltenham: Edward Elgar, 2008) 193–208. Ulbricht, “The Decentralization of Tax Administrations”, supra, refers to the differing audit structures in different states as producing problems similar to those arising in the eu as whole owing to the lack of any central vat audit system as described in S. Cnossen, “Commentary”, in J. Mirrlees et al. (eds.), Dimensions of Tax Design: the Mirrlees Review (Oxford uk: Oxford University Press, 2010) 370–386.

Fiscal Decentralisation & Decentralising Tax Administration

199

activities in various respects. As one would expect, studies suggest that the results of this awkward mix of responsibilities for different aspects of taxation between the national and state levels are far from uniform across states.26 Increased tax evasion and increased compliance costs have been blamed on this decentralised approach to tax administration. In addition, the evidence suggests that state tax administrations have responded to the incentives created by the equalisation system by adjusting administrative efforts in such a way as to align the effective tax rates imposed on their residents to match the share of the marginal tax revenue that accrues to the states. The factors determining marginal tax back rates (mtbrs)—the share of the revenue collected that accrues to the state after equalisation—are exogenous to the state. Hence, although tax rates are uniform across states, variations in effective tax rates may arise because of variation in enforcement effort (e.g., the number and effectiveness of audits). States like North Rhine-Westphalia that are consistently net contributors to equalisation have high mbtrs and hence lower incentives to enforce the tax law than states like Saarland that are consistent net recipients with low mbtrs.27 Despite such problems, the unique German tax administration system has proven resistant to change for several reasons. Not only is it well-entrenched and popular with the states but it is considered to be more democratically responsive and accountable than a more centralised administration because states have to collect their own revenue rather than simply depend on tax shares transferred from a centralised collection system. However, as just mentioned, the strength of this last argument is clearly weakened by the offsetting effects of the particular equalisation system in place in Germany which explicitly discourages tax efforts in richer states.28 3.2 China Perhaps surprisingly, the country in which tax administration comes closest to the German model is probably China, in which an astonishingly small central tax office sets both policy and administrative guidelines but all taxes for 26 27 28

C. Baretti et al., “A Tax on Tax Revenue: The Incentive Effects of Equalizing Transfers: Evidence from Germany”, International Tax and Public Finance, 9 (2002) 631–649. T. Bönke et al., “Fiscal Federalism and Tax Administration: Evidence from Germany”, Discussion Paper 1307, diw Berlin (2013). Any equalisation system affects taxation in recipient states; the German system explicitly affects tax efforts in donor states also, M. Smart, “Taxation and Deadweight Loss in a System with Intergovernmental Transfers”, Canadian Journal of Economics, 31 (1998) 189–206.

200

Bird

all levels of government are actually collected by a vast network of regional (provincial) and local tax offices. Unlike Germany, Canada, Spain, or the United States, China is an explicitly unitary country, with four formal levels of government—central, provincial (as well five large “special municipalities” which are for fiscal purposes treated much like provinces—like Hamburg and Bremen in Germany), county, and township.29 Until 1994 there was essentially no central tax administration in China. All taxes were collected by local tax offices, which collected both central taxes and taxes shared with the provinces as well as local taxes. However, with China’s economic liberalisation beginning in the 1980s the central government essentially lost control over the tax system, and both total tax revenues and the share of those revenues going to the central government declined sharply by the early 1990s.30 The 1994 reform thus dramatically centralised taxation, including establishing a national tax administration for the first time and removing customs duties, value-added tax (vat), and excises as well as the income tax on central enterprises from the control of Local Tax Bureaus (ltbs) and giving them to new State Tax Bureaus (stbs) which were organised in a hierarchy of provincial, prefectural/municipal and county tax offices under the central State Administration of Taxes (sat). All organisational and human resource decisions at each level are under control of the next higher level, with sat on top of the whole system. sat also has some influence over the Local Tax Bureaus although they are predominately under the control of the provincial governments. Despite this apparently centrally-controlled structure, many aspects of tax administration remain under regional control to varying extents. For example, provincial governments have some discretion over such items as vat thresholds and various income tax reliefs as well as over local business and land taxes. More generally, as Cui notes—and both Spanish and German experience also suggest

29

30

In addition, there are several hundred thousand rural villages, which—uniquely in China—actually have elected officials but also—uniquely among Chinese government units—have no formal taxing power. For a brief description of some aspects of village finance, see R.M. Bird et al., “Fiscal Reform and Rural Public Finance in China”, in J. Man and Y. Hong (eds.), China’s Local Public Finance in Transition (Cambridge, ma: Lincoln Institute for Land Policy, 2011) 227–243. C. Wong and R.M. Bird, “China’s Fiscal System: A Work in Progress”, in L. Brandt and T.G. Rawski (eds.), China’s Great Transformation: Origins, Mechanisms, and Consequences of the Post-Reform Economic Boom (Cambridge: Cambridge University Press, 2008) 429–466.

Fiscal Decentralisation & Decentralising Tax Administration

201

[…] whoever controls tax collection—a centrally- or locally-responsible tax office—partially determines the extent to which local governments control the actual implementation of tax policy […] even in the absence of tax legislative power.31 Even though the relative recentralisation of the tax system after the 1994 reform succeeded in increasing both tax revenue and the central share of tax revenue, the system as it actually operates continues to depend heavily on the exercise of administrative discretion at the local level. One perhaps surprising result appears to be that the central government has so little reliable information about taxpayer operations that few audits are carried out, with much greater reliance being placed on such arbitrary rules as limiting vat credits and generally focusing enforcement efforts on relatively compliant and easy to monitor sectors.32 Moreover, although stbs, part of the central government administration, are responsible for administering central taxes, even the lowest level of the stb hierarchy—the county—encompasses such a large area that in practice for many taxpayers the real face of the tax administration is an “outpost” office covering a much smaller territory, staffed by lower-level (and less well-paid) personnel, who are sometimes partly funded by local governments.33 To a large extent the real work of tax administration in China is thus carried out by officials at the very bottom of the enormous state bureaucratic machine who may not be well qualified and are often responsive to local as well as central influences. It is not surprising that the combination of this administrative structure and the considerable degree of local discretion in administering both central and local taxes arguably explains the relative underdevelopment of tax litigation and the tax legal profession in China: as Cui says, “[…] the law ceases to be relevant beyond the boundaries of specific tax administrator’s knowledge about the law”.34 Many years ago, Casanegra noted that in developing countries tax administration often was tax policy in the sense of determining whether and how policy intentions were realised.35 The decentralised way in which Chinese taxes 31 32 33 34 35

W. Cui, “Fiscal Federalism in Chinese Taxation”, World Tax Journal, (2011) 455–480, at 472. W. Cui, “Administrative Decentralization and Tax Compliance: A Transactional Cost Perspective”, University of Toronto Law Journal, 65 (2015) 186–238. Ibid. Ibid., at 37. M. Casanegra de Jantscher, “Administering a vat”, in M. Gillis et al. (eds.), Value Added Taxation in Developing Countries (Washington dc: World Bank, 1990).

202

Bird

are administered perhaps pushes this point so far that tax administration may become not just policy but de facto law. If what low-level officials at the taxpayer interface (whether in an stb outpost or an ltb office) decide should be done differs from what the law or sat may seem to indicate should be done in the circumstances of the case, the lowest decision almost always prevails. It is hard to have a more decentralised tax administration than this. Whether the outcomes of such a system are likely to be efficient, equitable, accountable, or desirable is of course another question. 3.3 Spain Spain, like China, is not a federation. However, it is also decidedly not a unitary state. In fiscal terms, it is a uniquely asymmetrical decentralised country which is still working out precisely how and to what extent the apparently conflicting desire for increased state autonomy and national unity can best be accommodated. Perhaps the closest parallels are the case of Quebec in Canada (discussed below) and Belgium and perhaps in the near future the increasingly less United Kingdom, although the fiscal parameters of the last two cases are not further discussed here.36 Any discussion of tax administration in Spain must begin by noting the unique characteristics of the fiscal regime applying in the two foral states (technically, “autonomous communities”), Navarre and the Basque Country. In these regions, essentially for historical reasons, all tax revenues are and always have been collected at the state level—or, more precisely in the case of the Basque Country, by the three provinces that make up that state.37 These regions have almost complete fiscal autonomy over personal and corporate income taxes as well as administrative control over vat and excise duties: in effect, 36

37

On Belgium, see B. Bayenet and P. de Bruycker, “Belgium: A Unique Evolving Federalism”, in R.M. Bird and R.D. Ebel (eds.), Fiscal Fragmentation in Decentralized Countries: Subsidiarity, Solidarity and Asymmetry (Cheltenham: Edward Elgar 2007) 169–207. Belgium has recently given regions considerable autonomy over some taxes, including the right to administer them, and some regions have chosen to do so. oecd, Comparative Tax Administration, supra. A useful recent overview of the situation in Scotland may be found in D. Bell, et al. (eds.), The Economic Consequences of Scottish Independence (Hamburg: Helmut Schmidt Universität, 2014), www.scotecon.org/book.html (accessed 20 November 2015). The state tax agency in the Basque Country is principally concerned with coordinating the work of the three provincial administrations, A. Esteller Moré, “Current Situation and Proposals for Reform of Spain’s Tax Administration”, in Bosch and Durán, Fiscal Federalism, supra, 209–247.

Fiscal Decentralisation & Decentralising Tax Administration

203

they administer all central taxes other than import duties and payroll taxes. However, the central tax agency (aeat, Agencia Estatal de Administración Tributaria) is in charge of managing corporate and vat returns for companies that by law must file national returns although they are fiscally domiciled in the Basque Country.38 The foral regions benefit from but do not explicitly contribute to the equalisation transfer system.39 They can choose to exert the same tax effort as other regions and end up with higher spending or lower their tax effort and end up with the same level of spending. For the most part, they appear to have chosen to reduce taxes in order to attract investment—at no cost to their taxpayers.40 Apart from the long history of the foral system, however, tax decentralisation in Spain is a relatively new phenomenon. The Spanish tax system has undergone substantial changes since it was launched in its present form in 1977, and is unlikely as yet to have reached its final development.41 Although new taxes on income and wealth were introduced in 1977, the administration remained essentially centralised until the mid-1980s when the vat was introduced and tax administration was substantially decentralised.42 New regional offices of the central administration were created in each state, with the old provincial (district) offices being subordinated to them, and over 200 new local tax offices were created and charged with supervision and auditing of small business taxpayers. The new state administrations were put in charge of taxes on wealth. Despite substantial early efforts to improve the information 38 39

40

41

42

Ibid. The Basque Country (and Navarre) pay a negotiated amount to the central government (varying from year but usually close to 10% of the revenue they collect) to cover a few other central outlays such as defence and foreign affairs, C. Gray, “A Fiscal Path to Sovereignty? The Basque Economic Agreement and Nationalist Politics”, Nationalism and Ethnic Politics, 21 (2015) 63–82. J. Lopez-Laborda and C. Monsterio Escudero, “Regional Governments: Vertical Imbalances and Revenue Assignments”, in J. Martinez-Vazquez and J.F. Sanz-Sanz (eds.), Fiscal Reform in Spain: Accomplishments and Challenges (Cheltenham: Edward Elgar 2007) 422–452. For a critical view of the fiscal asymmetry resulting from the foral system, see T. Garcia-Milà and T.J. McGuire, “Fiscal Decentralization in Spain: An Asymmetric Transition to Democracy”, in Bird and Ebel, Fiscal Fragmentation, supra, 208–223. For a review of the first 30 years, see Martinez-Vazquez and Sanz-Sanz, Fiscal Reform in Spain, supra. A more recent appraisal of fiscal federalism in Spain may be found in V. Ruiz Almendral, “Sharing Taxes and Sharing the Deficit in Spanish Fiscal Federalism”, eJournal of Tax Research, 10 (2012) 88–125. J. Onrubia, “The Reform of the Tax Administration in Spain”, in Martinez-Vazquez and Sanz-Sanz, Fiscal Reform in Spain, supra, 484–531.

204

Bird

technology system underlying tax administration,43 problems in coordinating this fragmented administrative structure appear to be one factor resulting in regional differences in the effectiveness of enforcement.44 This question became more important in the mid-1990s when the states had not only established their own tax administration departments but were also given additional taxing powers (e.g., property transfer tax and vehicle registration tax) as well as significant shares in income tax and vat, although the latter continue to be administered by the aeat, which was established in its present form in 1997.45 State tax agencies had some representatives on the board of directors of aeat and a number of formal intergovernmental committees were established to improve interagency coordination. On the whole, however, the central agency dominated tax administration although how effective this structure was in the absence of strong incentives for regions to cooperate was not clear.46 Since there is obviously considerable overlap between, e.g., the net wealth tax administered by the states and the income tax administered by the centre as well as potential for regional competition in the form of differential enforcement efforts, some degree of cooperation at least in the form of information exchange would seem to be called for. However, although it may emerge over time, as yet it is not clear that the necessary degree of reciprocity (in terms of the extent to which the misreported revenues resulting from the 43

44

45

46

R. Moya and J.D. Santiago, “Information Technology: Strategy of the Spanish Tax Administration”, in R.M. Bird and M. Casanegra de Jantscher (eds.), Improving Tax Administration in Developing Countries (Washington dc: International Monetary Fund, 1992) 211–235. Wealth tax returns are designed by the central tax agency and filed with the income tax return, although the process of controlling and monitoring returns and verifying data is carried out by the states. The central agency can also audit if it so chooses, Esteller Moré, “Current Situation”, supra. Although state administration of the newly transferred taxes was formally subject to annual audit by the central Ministry of Finance, Onrubia notes that this is “… a formal requirement with hardly any implications for…the regional tax administrations”, Onrubia, “The Reform”, supra. Esteller Moré, “Current Situation”, supra, outlines how revenues from shared taxes are allocated and transferred to the different states and notes that timing and adjustment issues have occasionally led to disputes—as in Canada, see R.M. Bird and P. Gendron, “Sales Taxes in Canada: The gst-hst-qst-rst ‘System’”, Tax Law Review, 63 (2010) 517–582. Onrubia, “The Reform”, supra. For evidence of such competition in Spain, see J.M. DuránCabre et al., “Regional Competition on Tax Administration” (2012), https://ideas.repec .org/p/wiw/wiwrsa/ersa12p184.html (accessed 20 November 2015). As they note, one benefit of giving states more formal legal authority over tax rates is to make such competition more transparent than when it is opaquely hidden in differential enforcement efforts (as in the German case).

Fiscal Decentralisation & Decentralising Tax Administration

205

lack of such exchange exceed the cost for most regions) exists to provide sufficient incentive to make such exchanges work.47 Although some autonomous communities (states), in particular Catalonia, have continued to press for more fiscal autonomy,48 for the most part their main concern has been simply to obtain increased financial resources. First, they obtained some power over personal income tax and some other “ceded” taxes; next, they sought more control over taxes such as vat and corporation tax as well as a more important role in tax administration. However, as in Germany, Spain’s intergovernmental fiscal transfer system is so designed that states have little or no incentive to enforce effectively even the taxes they are supposed to regulate.49 In this and other respects, the “work in progress” that is fiscal federalism in Spain does not as yet appear to have reached an equilibrium that will prove sustainable over time. In all likelihood, as in the quite different case of China, there will be yet more changes to the level and nature of tax administration in the future.50 .

3.4 Canada Canada is, by most standards, one of the most fiscally decentralised countries in the world. However, its tax administration is substantially less decentralised than that in Germany, Spain, China, or the United States. Unlike many federations (e.g., India), there is no effective separation of revenue sources between federal and provincial governments.51 Moreover, there are no formal 47

48

49 50

51

J.M. Durán-Cabre et al., “Empirical Evidence of Tax Cooperation between Sub-central Regions”, Document de treball de l’IEB, 7 (2015), https://www.academia.edu/1567090/Empirical _evidence_on_horizontal_competition_in_tax_enforcement (accessed 20 November 2015). The cited study notes that some regions were particularly “sluggish” in responding to information requests. Although the issue has not been explored, perhaps “sluggishness” may reflect opportunistic challenges to federal influence, J. Bednar, The Robust Federation: Principles of Design (Cambridge: Cambridge University Press, 2009). For an early example, see A. Castells, “The Role of Intergovernmental Finance in Achieving Diversity and Cohesion: The Case of Spain”, Environment and Policy C: Government and Policy, 19 (2001) 189–206; the author later became Catalonia’s Minister of Finance. Ruiz Almendral, “Sharing Taxes”, supra. Esteller Moré discusses several possible directions for reform and suggests that perhaps the best solution might be a formal consortium arrangement, with each level of government being equally represented on the governing body of aeat, although he expresses some concern about the extent to which such arrangements (originally proposed by Catalonia) may result in an undesirable degree of tax diversity. Esteller Moré, “Current Situation”, supra. The constitution appears to establish such a separation, with provinces being limited only to “direct” taxes. However, over the years judicial interpretation enabled provinces

206

Bird

“tax sharing” arrangements—like those in Germany, for example—that divide the revenues of particular taxes between levels of government. Instead, over a long history and especially over the last 50 years, an interestingly asymmetric administrative structure (somewhat like that in Spain) has been developed. The allocation of profits for corporate tax purposes has often given rise to controversy, and some specific rules exist for certain industries such as banks and pipeline companies, but all provinces, including those with independent income taxes, have adopted similar rules—a process that took place over decades of discussion and negotiation52—and various federal and provincial working groups exist to deal with differential interpretations that might result in double taxation or alter the distribution of revenues in unacceptable ways.53 One province, Quebec, administers its own personal and corporate income taxes as well as its sales tax; in addition, (as the German states do for almost all central taxes) it administers the federal value-added tax (gst) within its borders. The province is reimbursed by the federal government for the costs of administering the federal tax. Another province (Alberta) administers its own corporate income tax, and three provinces (British Columbia, Manitoba, and Saskatchewan) administer their own retail sales taxes. On the other hand, the federal tax administration, the Canada Revenue Agency (cra), administers provincial personal income taxes in nine provinces, provincial corporate taxes in eight provinces, and provincial general sales taxes in five provinces.54 These federally-administered provincial taxes are, in accordance with explicit agreements made with each province, essentially similar in terms of their base to the corresponding federal taxes (as are the similar taxes administered independently by Quebec). All provinces can and do vary the base in certain

52 53 54

to levy important general sales taxes as well as almost any other tax they care to impose, B. Alarie and R.M. Bird, “Tax Aspects of Canadian Fiscal Federalism”, in G. Bizioli and C. Sacchetto (eds.), Tax Aspects of Fiscal Federalism: A Comparative Analysis (Amsterdam: ibfd, 2011) 77–136. On the other hand, provinces maintain strict and strong control over municipal finances, with the result that the only significant local tax is the real property tax and the province usually in effect controls that tax also, see R.M. Bird et al., A Tale of Two Taxes: Reforming the Property Tax in Ontario (Cambridge, ma: Lincoln Institute of Land Policy, 2012). E.H. Smith, Federal-Provincial Tax Sharing and Centralized Tax Collection in Canada (Toronto: Canadian Tax Foundation, 1998). P. Berg-Dick et al., “Tax Coordination under the Canadian Tax System”, in Bosch and Durán, Fiscal Federalism, supra, 169–192. Alberta, like the three northern territories (Northwest Territories, Nunavut, and Yukon) has no sales tax.

Fiscal Decentralisation & Decentralising Tax Administration

207

respects (in the case of income taxes only by credits, however) and can also determine tax rates. cra is clearly an agency of the central government, reporting to a federal cabinet minister. However, 11 of the 15 members of cra’s Board of Management, which is appointed by the federal government, are nominated by the provinces (and territories), although all come from the private sector, with the exception of the federally-appointed Commissioner who heads the agency and is accountable both to the federal Minister of Revenue for administration of the legislation and to the Board for other administrative purposes (e.g., staffing and organisational issues). The cra is subject to value-for-money audits by the federal Auditor General. That the provinces find this system acceptable is indicated both by the fact that, in 2009, Ontario turned over the administration of its corporate income tax to the cra and that six provinces have in the last decade chosen to replace their general retail sales taxes by a federallyadministered provincial value-added tax, which is administered jointly with the federal gst under the name of the harmonised sales tax (hst).55 The cra administers these provincial taxes free of cost to the province provided its tax base is essentially identical to that of the corresponding federal tax. Each province has a separate tax administration agreement with the cra which leaves them free to deviate to a limited extent from the federal tax structure, e.g., by offering credits under the income tax and zero-rating specific activities under the sales tax. Provinces may also impose whatever rates they choose for income tax purposes: until 2015, for example, Alberta imposed only a flat-rate personal income tax, while Ontario imposed an additional surtax on higher incomes (and some corporations) in 2014.56

55

56

At present, only five provinces have such a tax, because British Columbia withdrew from the agreement in 2013 following a referendum in which a majority favoured returning to the retail sales tax. The full story of Canada’s gradual (and still incomplete) move towards an integrated federal-provincial value-added tax is told in Bird, “The gst/hst”, supra. Under the initial tax collection agreements participating provinces were able only to impose a single surtax rate on the federal income tax. This “tax-on-tax” system was replaced in 2000 by a “tax-on-income” approach allowing provinces to impose whatever tax rates they wanted without having to accept the progressivity set by the federal schedule (Canada, 2000). See, Department of Finance Canada and Canada Customs and Revenue Agency, “Federal Administration of Provincial Taxes: New Directions” (2000), http://www .fin.gc.ca/fapt-aipf/fapte.pdf (accessed 29 February 2016);. R.M. Bird and F. Vaillancourt, “Changing With the Times: Success, Failure and Inertia in Canadian Federal Arrangements, 1945–2002”, in J. Wallack and T.N. Srinivasan (eds.), Federalism and Economic Reform: International Perspectives (Cambridge: Cambridge University Press, 2006) 189–248.

208

Bird

In the case of the sales tax, the federal government will administer provincial-specific provisions for free provided that no more than five per cent of the base is affected in total; it may also agree to administer additional specific deviations on a cost-reimbursable basis.57 Two provinces, Ontario and Quebec (the latter of which is not in the hst system although its tax base is essentially the same), have even been permitted to refrain from crediting some taxes on capital expenditures when made by large companies, although this deviation is permitted for only a limited period of time. Despite these differences in tax base between the federal and provincial taxes, as well as between the different provincial taxes, there have been no reported difficulties in administration and no serious problems with evasion with respect to cross-border sales, largely because the gst/hst system is operated in effect like an integrated national tax, with the allocation of revenues to the participating provinces being made in accordance with the estimated distribution of taxable base as determined by a federal-provincial working group.58 Canada’s longest-serving Prime Minister (W.L.M. King) once dealt with one of the country’s longest-lived political debates—between Francophones, who mostly did not support military conscription, and Anglophones, who mostly did—by saying at one point during the Second World War that he supported “conscription if necessary, but not necessarily conscription”. Along somewhat similar lines, Canada’s tax administration may be said to be centralised if necessary, but not necessarily centralised: most provincial taxes are now administered centrally but only with the support and agreement of the provinces; moreover, the taxes administered are clearly provincial and differ to 57

58

cra administers well over a hundred separate programs for the provinces and territories. It also administers a number of income and sales taxes for various First Nations (aboriginal reserves), although this complex area is not further explored here, P. Nault, “Tax Administration in a Decentralized Tax System: A Canadian Perspective” (2006), www.ief.es/contadorDocumentos.aspx?URLDocumento=/documentos/investigacion/ seminarios/internacional/sem_inter_tax_philipnault_taxad.pdf (accessed 10 November 2015). In the 2013–14 fiscal year, cra collected $305 billion in taxes (excluding $42 billion in pension contributions). About 31% of the taxes collected were provincial taxes: $55 billion in personal income tax, $13 billion in corporate income tax, and $27 billion as the provincial share of the hst, Canada Revenue Agency, “Annual Report to Parliament 2013–2014” (2014), http://www.cra-arc.gc.ca/gncy/nnnl/2013-2014/fn-dmnstrd-tvts-eng .html (accessed 20 November 2015). A small amount of provincial sales (and excise) taxes are also collected at the border by the Canada Border Security Agency (cbsa), which also has explicit agreements with most provinces. For a detailed discussion of how this system works, see Bird and Gendron, “Sales Taxes in Canada”, supra.

Fiscal Decentralisation & Decentralising Tax Administration

209

varying extents from the corresponding federal taxes and from those in other provinces. These arrangements, though seldom the subject of public discussion, are almost constantly under review at both levels of government and are implemented in accordance with contractually agreed arrangements— arrangements­drafted largely by federal officials but with significant provincial input and with disagreements in interpretation being resolved by joint discussions. The setup as a whole is not that easy either to explain or understand. To date, however, much like the country itself,59 it has not only managed to survive but also to accommodate substantially changing circumstances over time and across regions. 4

The Case for and against Decentralising Tax Administration

Traditionally, an important criterion in assigning taxes in the first place is whether the government to which they are assigned can feasibly administer them. Indeed, this seems to be the main reason why property taxes are so often assigned to local governments in most countries. This consideration seems likely to be especially important in the many developing countries in which tax administration has long been weak. Unfortunately, one argument cannot decide the issue. The property tax, for instance, is in some ways surprisingly difficult to administer well.60 Concern with administrative feasibility does not outweigh other important considerations in choosing good regional and local taxes such as efficiency, equity, acceptability, and accountability. The introduction of modern tax technology has in many ways made centralisation less costly, as evidenced by the almost world-wide tendency of central administrations to reduce the number of local tax offices.61 On the other hand, the continued attention being paid in many countries to the case for decentralising many public sector activities,62 like the “second-generation” theoretical arguments suggesting that decentralised governments need to be fiscally autonomous to some extent in order to deliver services effectively (see Section 2 above) point in the other direction when it comes to the administration of 59 60

61 62

Bird and Vaillancourt, “Changing with the Times”, supra. R.M. Bird and E. Slack, “Local Taxes and Local Expenditures in Developing Countries: Strengthening the Wicksellian Connection”, Public Administration and Development 34 (2014) 359–369. oecd, Comparative Tax Administration, supra. J.-P. Faguet and C. Pöschl, Is Decentralization Good for Development? Perspectives from Academics and Policy Makers (Oxford: Oxford University Press, 2015).

210

Bird

regional and local taxes. There is no one-size-fits-all solution to this question.63 Each country must weigh conflicting arguments, some implying that administration should be centralised and others that it should be decentralised. Often, whether the balance is struck one way or another depends on institutional and empirical factors that can be dealt only in the specific context where the issue must be decided. As Section 3 shows, how sub-national taxes should be administered remains an open question even in long-established developed countries. One author suggested some years ago that the choice was between incompetence (because local administrations are likely to be less capable) and indifference (because central administrations are not likely to pay much attention to local concerns).64 This view is too simple. It is true that when administration has been turned over to local governments, the results have not always been good.65 Although high costs may sometimes reflect the badly-designed taxes that local governments are supposed to administer, they may equally be the result of poor administrative capacity, incompetence, and corruption at the local level. But even when tax policies are well-designed and administrations at both levels of government are well-run, as in Canada, the evidence that a single administration is more cost-effective than multiple administrations is strong.66 One important reason for centralising tax administration is to reduce the costs of taxation. Decentralising administration increases collection and compliance costs because fixed costs are associated with collecting any tax, so that the more jurisdictions that collect the greater the total cost incurred. Evasion and avoidance may also increase with decentralisation for taxes where the 63

64 65

66

F. Ambrosanio and M. Bordignon, “Normative versus Positive Theories of Revenue Assignments in Federations”, in E. Ahmad and G. Brosio (eds.), Handbook of Multilateral Finance (Cheltenham: Edward Elgar, 2014) 231–263. W. Dillinger, Urban Property Tax Reform (Washington dc: World Bank, 1991). In Indonesia, for example, local administrative costs eat up over 50% of the revenue collected, B. Lewis, “Local Government Taxation: An Analysis of Administrative Cost Inefficiency”, Bulletin of Indonesian Studies, 42 (2006) 213–233. In Argentina, administrative costs at the provincial level vary from 1.1%—less than the 2% at the national level—to over 10%, Fretes Cibils and Ter-Minassian, Decentralizing Revenue, supra. Single administration of federal and provincial corporate income taxes was estimated to reduce compliance costs by 1.3% of the amount collected. When Ontario shifted administration of its cit and later its sales taxes to the cra, the provincial government noted that substantial reductions in administrative costs were an important reason for doing so, R. Plamondon and D. Zussman, “The Compliance Costs of Canada’s Major Tax Systems and the Impact of Single Administration”, Canadian Tax Journal, 46 (1998) 761–785.

Fiscal Decentralisation & Decentralising Tax Administration

211

tax base is mobile or straddles more than one jurisdiction. The administrative, compliance, and efficiency (and inequity) costs associated with taxes may all be reduced by limiting the number of agencies engaged in tax administration. The economies of scale and scope associated with the information systems on which modern tax systems are increasingly dependent,67 may be achieved through more effective and coordinated use of specialised staff in a more centralised administrative structure. Compliance costs are reduced when taxpayers submit fewer returns, interact with fewer officials and offices, and deal with a more uniform and harmonised administrative structure. Administrative costs are reduced when it is simpler to communicate laws and regulations to officials and establish exchanges of information and coordination of action without complex interagency or interjurisdictional negotiations. A more uniform and better coordinated administrative system should also be able to provide more equal procedural treatment to all taxpayers and to cope more effectively with the complications arising from cross-border transactions as well as with evasion in general. It may also be less susceptible to political interference and corruption. On the other hand, as experience in numerous countries around the world has shown, sub-national governments may be correct to worry that the central government will be less enthusiastic about collecting their taxes than its own. Central governments are unlikely to have much incentive to do a good job. After all, it’s not their money, and national political and other concerns seldom turn on the needs and concerns of lower levels of government. In the terms mentioned earlier, it may perhaps prove easier to remedy “incompetence” than “indifference”. Moreover, bringing administration closer to the people it is supposed to serve may result not only in improved accountability gains but also efficiency gains—and perhaps even increased revenues—because people can more easily identify how fairly taxes are being administered and what the money is being spent on. Better local information and knowledge, more flexibility in organising and staffing to fit local conditions, greater scope for innovation when there are several tax agencies rather than one monopoly, and above all the greater incentive for local administrators to collect and spend local revenues effectively may all make increased local control over tax administration a vital component of the kind of fiscal autonomy that seems most likely to produce the theoretical benefits of fiscal decentralisation. Some of the problems commonly associated with decentralising tax administration can be overcome. For example, a standard lament is that local 67

R.M. Bird and E. Zolt, “Technology and Taxation in Developing Countries: From Hand to Mouse”, National Tax Journal, 61 (2008) 791–821.

212

Bird

administrative capacity is inadequate to do the job properly. While this may sometimes be not only true but inescapable—for instance, small local governments are unlikely ever to be able to operate a standard credit-invoice vat efficiently68—there are many ways around this problem. As Sir Arthur Lewis noted long ago, too often those at the top seem to underestimate the potential of local administrators to do a good job.69 If better performance is valued and rewarded, and if pathways to improving capability and capacity are made available, there is no a priori reason to assume that people will not respond positively. As with fiscal decentralisation in general, there is undoubtedly a learning curve: it may take time, perhaps quite a lot of time, for regional and local governments and their citizens to learn how to run things effectively, let alone efficiently, especially in countries in which subnational officials have little experience with such matters. But they can and do learn, and since better local tax administration will build additional capacity in financial administration the result may also be improved local financial management and better expenditure outcomes. Additional costs associated with decentralised administration may thus to some extent be offset by benefits in terms of improved efficiency, equity, acceptability, and accountability. For example, although residential property taxes are not only unpopular but relatively costly to administer well, the higher costs may be fully justified both by the “benefit tax” aspect of such taxes and by how they increase government accountability.70 On the other hand, taxes—whether property taxes, income taxes, or excise taxes—imposed on local businesses that export most output (and hence most taxes unless they are, improbably, operating in a completely competitive market) to other jurisdictions are usually popular and may be imposed at lower costs per dollar of net revenue collected. Such taxes reduce not only accountability but also economic efficiency. In some instances, decentralisation may, instead of increasing competition between governments for a fixed amount of “tax room”, even increase the acceptable level of taxation. To the extent local administration may improve accountability by making it clearer to taxpayers what their taxes are buying, 68

69 70

Though, as Canada shows, regional governments can do so and, as a number of countries (Japan, Italy, France) have demonstrated even local governments can use a different type of vat effectively. R.M. Bird, “Below the Salt: Decentralizing Value-Added Taxes”, in Ahmad and Brosio, Handbook of Multilateral Finance, supra. A. Lewis, “Planning Public Expenditure”, in M. Millikan (ed.), National Economic Planning (New York: Columbia University Press, 1967) 210–227. Bird and Slack, “Local Taxes”, supra.

Fiscal Decentralisation & Decentralising Tax Administration

213

their willingness to pay may be increased.71 Surveys in Colombia for example, suggest that citizens in all economic groups felt they were getting more out of paying taxes to local than to national governments.72 If local administrators do a better job of identifying and assessing the tax base, overall revenue mobilisation may increase. In Armenia, for example, the delegation of property tax collection responsibility to the local government level in 2003–2005 reportedly led to a 38 per cent increase in collected tax revenue. On the other hand, the centralisation of sales tax administration in Kyrgyzstan in 2009 resulted in a decrease in the amount of collected tax.73 In Peru, those municipalities that created autonomous local tax offices on average raised their own-source revenues by 81 per cent from 1997 to 2008, compared to an increase of only 61 per cent in those that did not. The locally-run offices were found to have improved in terms of less political interference, better client focus and more trust, less corruption, more innovation and, interestingly, also in better cooperation with other tax administrations.74 Decentralised administration may also permit some new forms of taxation to be implemented. In many countries regional and local governments have broadened the tax net through a variety of special tax instruments and administrative measures such as levies on the sales of assets, licenses to operate, betterment charges and various forms of property and land taxation.75 Bahl and Linn suggest that dividing tax bases according to comparative advantages in assessment and collection may broaden the tax base that can be effectively reached,76 especially in the so-called “hard to tax” sector.77 In developing 71 72

73

74 75 76 77

Mikesell, “Developing Options”, supra. O.L. Acosta and R.M. Bird, “The Dilemma of Decentralization in Colombia”, in R.M. Bird et al. (eds.), Fiscal Reform in Colombia: Problems and Prospects (Cambridge, ma: mit Press, 2005) 247–286. N. Golanova and G. Kuryandskaya, “Local Public Finances in Eurasia”, in United Cities and Local Governments, Local Government Finance: The Challenges of the 21st Century. Second Global Report on Decentralization and Local Democracy (Cheltenham: Edward Elgar, 2011) 117–152. Fretes Cibils and Ter-Minassian, Decentralizing Revenue, supra. R. Bahl and R.M. Bird “Subnational Taxes in Developing Countries: The Way Forward” Public Budgeting & Finance, 28 (2008) 1–25. R. Bahl and J. Linn, Urban Public Finance in Developing Countries (New York: Oxford University Press, 1992). J. Alm et al. (eds.), Taxing the Hard to Tax: Lessons from Theory and Practice (Amsterdam: Elsevier, 2004). Segmentation of tax bases in terms of both structure and administration is also suggested in a recent review of how to improve local government taxation in Africa, O.H. Fjeldstad et al., “Local Government Taxation in Sub-Saharan Africa: A Review

214

Bird

countries state and local governments usually oversee a variety of licensing and regulatory activities and may be better able than the central government to track property ownership and land-based transactions as well as to identify local businesses and gain some knowledge about their assets and scale of operation. Because the potential revenue gains are much more important for local governments, they have more incentive to attempt to capture some who do not fully comply with national taxes or evade taxes altogether such as small businesses and the self-employed.78 However, even when there is a strong administrative case for a tax to be locally administered because of the comparative advantage of local governments in identifying the tax base owing to their familiarity with local conditions, such tasks as valuing properties can be complex and require coordinated action between a number of different local, state, and national agencies and departments. The central (or state) government should often play a substantial role in such tasks as setting valuation standards, training valuers, and monitoring the quality of local assessments especially when, as is common, intergovernmental transfers are based to some extent on estimates of the potential local tax base.79 All too often, however, central valuation agencies have shown little willingness in practice to respond to local needs for support in improving and maintaining local taxing capacity. The increasing “informatisation” of the world and the greatly expanded reliance on information technology (it) to deal with routine administrative processes has been a two-edged sword when it comes to tax decentralisation. On one hand, central tax administration can capture economies of scale through, e.g., centralised edp services and record-keeping, uniform approaches to assessment and audit, the development of centralised training programs, and so on.80 On the other, it can provide means to achieve more uniform service levels more efficiently and fairly in a more decentralised fashion without requiring every locality to have highly specialised skills. it permits a country to multiply its available skills by making them available at a distance when required as well as to monitor outcomes more effectively.81 Such a system

78

79 80 81

and an Agenda for Research”, cmi Working Paper, 2 (2014). For the contrary view, see Ahmad, “Multilevel Fiscal Institutions”, supra. As noted in Section 3, however, this incentive may be offset by the disincentive created by inappropriately designed intergovernmental transfer systems. In this and other ways, central governments arguably get the local governments their own policies shape. See, for example, the discussions of valuation in Bird and Slack, International Handbook, supra and of transfers in Bird and Smart, “Intergovernmental Fiscal Transfers”, supra. Vehorn and Ahmad, “Tax Administration”, supra. Bird and Zolt, “Technology and Taxation”, supra.

Fiscal Decentralisation & Decentralising Tax Administration

215

may, for instance, be one way to check the common concern about excessive corruption at the local government level, where the other side of being close enough to taxpayers to know them is being close enough to them to be susceptible to improper influences. Technical and political factors, however, are sometimes binding constraints at the local level in developing countries.82 There is no costless way to address all the constraints binding local tax administrative capacity, but some tradeoffs are always possible. A common approach is to restrict state and local governments only to taxes that are considered to be most easily administered. For example, Mexico and Australia allow states to impose payroll taxes; Argentina and Canada let them impose certain types of sales taxes; many countries allow state and/or local governments to impose taxes on the ownership or use of motor vehicle licenses; and many permit state and local governments to impose some form of business licenses as well as property taxes. However, some of these “easy-to-tax” taxes are all too often in practice poorly designed (most presumptive levies) or economically inefficient (property transfer taxes) or, like many property taxes, simply badly administered with low coverage rates, arbitrary assessment, and large delinquent lists. Central governments are usually keen to control corporate taxes for revenue, stabilisation, and regulatory purposes. Such taxes are also difficult for small governments to administer effectively, let alone efficiently. Taxes on international trade, like those on international investment flows, are also both difficult for regional and local governments to implement properly and highly distorting in efficiency terms. On the other hand, as North American experience shows, regional and even local sales taxes are possible, though since even national jurisdictions have so far been unable to tax some cross-border business activities fairly, efficiently and effectively they are even less likely to be able to do so.83 Although it is easier for sub-national governments to tax personal incomes (as in Switzerland, the u.s., and the Nordic countries) or to impose payroll taxes (as in Australia and Mexico) than to tax corporations, central governments may again be reluctant to permit them to do so, for example, because maintaining a degree of visible progressivity in direct taxation may be considered necessary for political stability.84 The regulatory role of 82

83 84

For a good (or bad) example, see J.A. Enahoro and J. Olabisi, “Tax Administration and Revenue Generation of Lagos State Government, Nigeria”, Research Journal of Finance and Accounting, 3 (2012) 132–139. Bird, “Below the Salt”, supra. R.M. Bird and E. Zolt, “Fiscal Contracting in Latin America”, World Development, 67 (March 2015) 323–335.

216

Bird

taxation may also play an important role in shaping a continued central role in excises on such products as alcohol, tobacco and petroleum, even though such taxes have sometimes been suggested as more suitable for subnational governments85 and are extensively used by such governments in some countries. Nonetheless, a possible comparative advantage for subnational government taxation relates to small taxpayers, who usually are less easily reached by central tax systems. Most tax revenue in most developing countries comes from a relatively small number of taxpayers. vat, excises, corporate income tax and even most personal income tax are mainly collected from larger firms in the formal sector of the economy.86 Indeed, Keen and Mintz argue that the vat threshold should be set at a relatively high level in terms of gross sales in order to exclude small taxpayers that it is not cost effective to reach.87 Similarly, the exemption level for the personal income tax in many developing countries is well above the average income level,88 and corporate income taxes often cover only large firms. However, local governments often impose various types of taxes and fees on small businesses excluded from the ambit of central taxes. The amounts of revenue raised through such levies are usually not large. Nonetheless, particularly in larger cities, these revenues are often both important and elastic and more use can and should be made of better-designed local business taxes.89 The policy constraints imposed on sub-national taxation may sometimes lead local governments to engage in a desperate search for revenues by piling on all sorts of specific local levies —entertainment taxes, advertising taxes, business taxes, and so on—which are usually costly to collect and often arbitrary in their administration. At one point, for example, China eliminated a major local tax and then implicitly allowed local governments to create an array of ad hoc, piecemeal and sometimes clearly illicit ways of filling the revenue hole that had been created.90 The outcome of pushing, e.g., growing cities 85

86 87 88 89 90

C. McLure, “Topics in the Theory of Revenue Assignment: Gaps, Traps, and Nuances”, in M. Blejer and T. Ter-Minassian (eds.), Macroeconomic Dimensions of Public Finance (London: Routledge, 1997) 94–109. Of course, the incidence of many of these taxes may be spread much more evenly across the population. M. Keen and J. Mintz, “The Optimal Threshold for a Value-added Tax”, Journal of Public Economics, 88 (2004) 559–576. R.M. Bird and E. Zolt, “Redistribution via Taxation: The Limited Role of the Personal Income Tax in Developing Countries”, ucla Law Review, 52 (2005) 1627–1695. R.M. Bird, “Local and Regional Revenues: Realities and Prospects”, in Bird and Vaillancourt, Perspectives on Fiscal Federalism, supra. Wong and Bird, “China’s Fiscal System”, supra; Bird et al., “Fiscal Reform”, supra.

Fiscal Decentralisation & Decentralising Tax Administration

217

to rely on such a hodgepodge of revenue sources is unlikely to be equitable or efficient, let alone to encourage responsible accountability. One reaction is to impose still stricter constraints, as when Russia reduced the list of local taxes from 22 to two in 2005.91 It is easy to go too far in this direction. Making it difficult for local governments to tax local businesses may reduce administrative costs, compliance costs, and also distortion costs. But if the result is not only to weaken the link between local government and local businesses but also the incentive for local governments to favour investment and growth, the game may not be worth the candle. A different approach to reducing the costs of collecting local revenue is by altering the structure of local taxes. One example is using area-based assessments for property taxation instead of more sophisticated valuation approaches based on comparative sales values.92 Another is a business license based on the estimated volume of sales rather than a sales tax based on actual sales records.93 However, such shortcuts may make a tax easier to administer at the expense of making it a less effective tax in other ways—for example, in the property tax case just mentioned by moving it from being a tax on property value and hence reducing its potential role as a surrogate form of benefit taxation.94 A better approach in many instances may thus be for regional and local governments to “piggyback” on the tax base of the higher level governments, as Canadian provinces do with respect to most of their taxes, thus in principle allowing them to be fully politically accountable without having to take on the task of tax administration. Like every approach this one has some drawbacks. Accountability may be adversely affected because taxpayers no longer view the tax as local because it is centrally administered and collected so that if the central government decides to alter the tax base—for example, to favour a particular industry or sector, local taxes are similarly affected.95 Although this effect may be offset—in Canada, for example, the central government is required to adjust intergovernmental transfers if its changes have a marked effect on provincial revenues and provinces may, if they wish, offset federal 91 92 93

94 95

Golanova and Kuryandskaya, “Local Public Finance”, supra. E. Slack, “Alternative Approaches to Taxing Land and Property”, in Bird and Vaillancourt, Perspectives on Fiscal Federalism, supra. This is how the industry and commerce tax operates in Colombian municipalities, J. Vázquez Caro and J. Abelino Ospina, “Evaluación sistémica de la ley impositiva colombiana”, in J.J. Echavarría (ed.), Bases para una reforma tributaria estructural en Colombia (Bogotá: Fedesarrollo y Banco de la República 2006) 385–507. Bird and Slack, “Local Taxes”, supra. Martinez-Vazquez and Timofeev, “Choosing between Centralized”, supra.

218

Bird

base changes by credits and surtaxes to a limited extent—the tax base is still essentially set at the central level and the lines of accountability are somewhat confused. Another obvious, though seldom observed, approach is simply to strengthen the administrative infrastructure of subnational governments and particularly to support them while they accumulate over time the necessary on-the-job experience to do the job properly. It may take a long time. Rome was not built in a day, and its tax administration is no doubt still far from perfect. Still, it has long been one of the world’s great cities, millions have managed to live there not too badly, and much has been done to improve matters over time. Even in very poor countries, it is sometimes possible to improve local administration substantially in a relatively short time if the right people do the right things. In Sierra Leone, for example, a recent careful analysis of why decentralisation worked well in some municipalities and not so well in others points out the great importance of particular local characteristics such as history, social settings, and the will and capacity of specific political figures.96 Reforming taxes and tax administration is never easy, but it can be done, and has been done in many instances in many countries at many times. However, it takes time, patience, and consistent support none of which is readily available in many parts of the world. In South Africa, for example, a major source of revenue at the time the post-apartheid government took power was a local tax (the Regional Services Council levy) that was so poorly structured that it was doomed to fail at some point. It did, with the result that it was then abandoned completely, removing a significant revenue source from local governments, although it could easily have been saved by some restructuring.97 Despite numerous attempts it has not proved possible to reassemble this Humpty-Dumpty once it was pushed over the wall. South Africa’s burgeoning cities thus have no adequate revenue base to cope with their pressing spending needs.98 Impatience for success can be fatal to the success of any aspect of fiscal decentralisation including tax administration. Central politicians and officials, like local voters, seem often to expect too much to work too well too soon—and then react too adversely when their unrealistic expectations are not met. 96 97

98

S.S. Jibao and W. Prichard, “The Political Economy of Property Tax in Africa: Explaining Reform Outcomes in Sierra Leone”, African Affairs, 114 (2015) 404–431. R. Bahl and D. Solomon, “The Regional Service Council Levy”, in R. Bahl and P. Smoke (eds.), Restructuring Local Government Finance in Developing Countries: Lessons from South Africa (Cheltenham: Edward Elgar, 2003) 127–172. N. Steytler, “Governance and Finance in Two South Africa Metropolitan Areas”, in E. Slack and R. Chattopadhyay (eds.), Governance and Finance in Metropolitan Areas of Federal Systems (Oxford: Oxford University Press, 2013).

Fiscal Decentralisation & Decentralising Tax Administration

5

219

Conclusion

This paper assumes that three key points put forward by many previous authors are correct. The first is that for political decentralisation to produce the gains promised by its advocates it must be accompanied by the right kind of fiscal decentralisation. The second is that both theory and experience tell us that the right kind of fiscal decentralisation is for state and local governments to be responsible for raising and spending their own resources at the margin so that changes in spending decided by local politicians and officials are financed by changes in taxes decided by those same people and are as transparent as possible to local citizens. The third point is that this goal may be largely achieved by the combination of a properly designed intergovernmental fiscal transfer system and allowing state and local governments to decide what tax rates they impose (even if only within a limited range on a limited set of taxes) as well as how to spend the revenue. The case for decentralising taxes neither requires nor is necessarily constrained by local tax administration. Tax decentralisation and the decentralisation of tax administration are related but separable decisions. The appropriate­ scope and nature of local administration of local taxes in any particular context can be determined only after careful consideration of the many, sometimes rather nebulous, facets of the issue discussed above. Different countries at different times have reached different conclusions about the appropriate way to mix and match the separable issues of decentralising taxes and decentralising their administration. No country may have it quite right when taking all the relevant factors into consideration, at least when viewed from outside. However, decisions on such matters in the real world are not made outside but inside specific countries, few involved in such decisions are likely to attach the same weights to all factors, and usually no one has the full story in mind when decisions are made. As with many questions of institutional design, there is no one size fits all correct answer in this complex world to either the question of the extent to which taxes should be decentralised or the question of whether such taxes should also be administered in a decentralised fashion. However, thinking through these two distinct questions separately can be a useful step towards achieving better outcomes.

section 2 Intergovernmental Financial Relations

∵

chapter 10

Intergovernmental Financial Relations: Institutions, Rules, and Praxis Elisabeth Alber 1

Introduction: Focus and Structure of the Paper

In order to effectively and efficiently deliver services, central governments assign the lower tiers of government specific powers and responsibilities. To properly deliver services attached to the assigned functions, each order of government needs adequate funding. Thus, the way in which the different orders of government are funded is at the crux of any democratic multilevel State. An appropriate allocation of financial resources to each order of government enables a State to both achieve its overall policy objectives and to regulate the economy across its tiers of government. For the sake of this paper, the design of intergovernmental financial ­relations (IGfRs) is defined as the overarching framework of relations across government orders. In detail, the paper analyses theory and practice as to the constitutional set-up of a State with regard to the allocation of fiscal and financial powers, financial decision-making processes, intergovernmental relations when it comes to schemes of revenue-sharing and equalisation, and the role of governmental and other bodies such as finance commissions or fiscal watchdogs.1 This paper centres on the institutional account of IGfRs, focusing primarily on the constitutional set-up of multilevel States and the composition and functioning of the most relevant institutional mechanisms in charge of designing and handling financial decision-making processes. If not stated otherwise, the paper refers to the central and the subnational level of government (i.e. states, provinces, Länder, autonomous communities). The paper refers to multilevel States in North America, Australia, Asia, Africa and Europe. In particular, it provides insight on States that, in the literature, have been defined

1 See R. Boadway, “International Lessons in Fiscal Federalism Design”, eJournal of Tax Research, 10 (2012) 21–48.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_012

224

Alber

as federal, quasi-federal or devolved.2 The general remarks in Parts 2 and 3 are assessed against the analysis of the main features of IGfRs in some of the oldest federal States (the United States in Part 4.1, Canada in Part 4.2 and Australia in Part 4.3), in two States that, formally, do not define themselves as federal States but that work like a federal State (India in Part 5.1 and South Africa in Part 5.2), in the federation of Germany (Part 6.1), in the quasi-federal State of Spain (Part  6.2) and in the United Kingdom of Great Britain and Northern Ireland, with its devolved governments in Scotland, Wales and Northern Ireland (Part 6.3). In Part 7, the paper engages in an analysis of (recently) established independent fiscal institutions whose mandate is to contribute to the strengthening of fiscal discipline. Against the background of the 2008– 2009 financial crisis, such “fiscal watchdogs” are increasingly being established especially in Member States of the European Union (eu), but not only. Part 7.1 gives evidence on North America, Asia and Africa, while Part 7.3 refers to case studies in the eu’s Eurozone. Finally, Part 8 provides comparative evaluations and concluding remarks. 2

The Design of Intergovernmental Financial Relations: Its Essence and Relevance

From an institutionalist perspective, the core of IGfRs can be summarised as being all those rules and practices that define, establish and govern institutions dealing with, and deciding on, questions as follows: How are tax powers shared? Who is responsible for raising taxes? How is tax revenue shared? Who decides on equalisation schemes and special money flows across government tiers? Principles and details related to these questions are specified in various sources of law and by means of political practice.3 While some multilevel States have very detailed constitutional provisions on taxation and finances (e.g. Germany), others rely on ordinary legislation (often reinforced by special majorities, e.g. Spain). The answers to questions such as how institutions dealing with IGfRs come into existence, which mandate they have and how they de facto work, widely vary across States. The historical and socio-economic ­characteristics of a State, the degree of heterogeneity of its population, the 2 Ex plurimis see R.L. Watts, Comparing Federal Systems (3rd ed., Montreal: McGill–Queen’s University Press, 2008); and A. Fenna and T. Hueghlin, Comparative Federalism: A Systematic Inquiry (2nd ed., Toronto: University of Toronto Press, 2015). 3 For key aspects, see R. Boadway and A. Shah, Fiscal Federalism: Principles and Practice of Multiorder Governance (New York: Cambridge University Press, 2009).

Intergovernmental Financial Relations

225

extent of central government intervention in the economy and the degree of fiscal decentralisation are among the most important factors that influence the design and functioning of IGfRs. In terms of challenges, any multilevel State, while deciding on how to ­arrange both vertical (across government orders) and horizontal (within constituent units) IGfRs, also has to take into account another set of concerns: it has to come to terms with the constructs of symmetry and asymmetry. The American scholar Charles Taylor defined symmetry and asymmetry in 1965, with symmetry being “the level of conformity and commonality in the relations of each separate political unit of the system to both the system as a whole and to the other component units”.4 He contended that a symmetric system would have no major social, economic or political peculiarities claiming for special rights, while this was not the case in an asymmetric system. Moreover, he went on to say that symmetry remains a theoretical construct because socio-economic diversities among constituent units of a State are, first of all, rarely well integrated and, as of late, not spread throughout an entire country. This leads to what has been called, in the literature, the “disparity of power ingredients” because “there is no system in the world in which all the component units are even approximately equal in size, population, political power, administrative skills, wealth, economic development, climatic conditions, predominance of either urban or rural interests, social structure, traditions, or relative geographic location”.5 Such a “disparity of power ingredients” is to be played out politically and may lead to the institutionalisation of asymmetric arrangements, which have to be carefully designed in order to not cause detrimental effects to the entire State.6 As Richard Bird pointed out, in the end, it is the “workings of the myriad of intergovernmental relations that constitute the essence of the public sector in all countries”.7 Thus, it is important not only to establish clarity on how institutions dealing with IGfRs are created and regulated, but it is equally important to examine how they play out (i.e. under which circumstances they operate). 4 C.D. Tarlton, “Symmetry and Asymmetry as Elements of Federalism: A Theoretical Speculation”, Journal of Politics, 27 (1965) 861–874, at 861. 5 I.D. Duchacek, Comparative Federalism: The Territorial Dimension of Politics (New York: Holt, Rinehart, Winston, 1970), at 280. 6 In this regard, see A. Kress’ chapter “Accommodating Diversity while Guaranteeing Stability: The Role of Financial Arrangements” in this volume. 7 R.M. Bird, “Intergovernmental Finance and Local Taxation in Developing Countries: Some Basic Considerations of Reforms”, Public Administration and Development, 10 (1990) 277–288, at 281.

226

Alber

A key question in any multilevel State is the allocation of taxing powers. As a general rule, taxing powers are rather centralised. This means that the central government raises more money than it actually spends for delivering the services it is responsible for, while the lower tiers of government have the opposite problem: they raise too little money compared to the services they are in charge of. To counteract such “vertical gaps”,8 revenues have to be distributed to the lower orders of government by either shared taxes or transfers.9 By contrast to own-source taxes (exclusively set and regulated by the government order they have been assigned to), shared taxes are taxes set by the central level whose revenues are appropriated to government tiers by means of complex formulas. Ideally, such formulas are linked to the fiscal capacity of the interested territories. Federal transfers are flows of money from the central to the subnational level of government that might be earmarked or not.10 There are many different forms of conditional transfers. Depending on which conditions are attached to the transfers, the central government might be enabled to heavily influence subnational policymaking. On the contrary, unconditional transfers may lead to uncontrolled spending. The design of IGfRs is also crucial to addressing horizontal imbalances. Due to socio-economic reasons, some constituent units might be unable to provide their citizens with basic services at the same standard as other constituent units do. Moreover, economies of scale make it more convenient to provide for services in areas with a high population density than in those with population 8

9

10

For the purpose of this paper and in line of what R. Boadway proposes, vertical fiscal gap (vfg) is defined as the asymmetry in revenue raising between the federal and state governments. Cited in C.K. Sharma, “Beyond Gaps and Imbalances: Re-structuring the Debate on Intergovernmental Fiscal Relations”, Public Administration, 90 (2012) 99–128, at 102. The terms vfg and vertical fiscal imbalance (vfi) are sometimes used interchangeably even though they convey different ideas. The definition of vfi for the purpose of this paper is the one proposed by R. Boadway and J. Tremblay: “Vertical fiscal imbalance is a situation in which the size of transfers made by the federal government to the provinces falls well short of the amount of federal tax revenues relative to their expenditures responsibilities, that is, what one might think of as the optimal vertical fiscal gap”, cited in ibid, at 100. Misallocations of funds may also be the cause of high vfi. Please note that this paper does not stipulate a definition of the various terms used. It adopts terminology with regard to I­ GfRs as commonly used in the respective State and its official documents as well as widely recognised doctrine. See R. Boadway, “Mind the Gap: Reflections on Fiscal Balance in Decentralized Federations”, in T. Courchene et al. (eds.), The Federal Idea: Essays in Honour of Ronald L. Watts (Montreal: McGill-Queen’s University Press, 2011) 363–377. See extensively in R. Boadway and A. Shah (eds.), Intergovernmental Fiscal Transfers: Principles and Practices (Washington, dc: World Bank, 2007).

Intergovernmental Financial Relations

227

scarcity. When addressing the issue of balancing territorial differences resulting from fiscal capacities or fiscal needs, multilevel States usually foresee some kind of equalisation mechanisms and special flows of money.11 The provision of such instruments is normally mandated by principles enshrined in the ­federal constitution (i.e. solidarity and equality principles) and it serves the purpose of general objectives of macroeconomic governance.12 For example, Article 72(3) of the German Basic Law refers to “comparable living standards”, Article 158(1) of the Spanish Constitution to “equivalent living conditions” with regard to essential services (health, education and social assistance) and A ­ rticle 275(1) of the Indian Constitution to the concept of “need of assistance”. The motivation to engage in equalisation schemes and special transfers comes from the recognition that excessive disparities in wealth among constituent units are likely to have negative effects on national cohesion and, thus, ultimately, also on vital macroeconomic objectives of a State such as stability, sustainable economic growth, equity in the distribution of income and sound structures in public finance. 3

From Theory to Practice: Institutional Archetypes of Intergovernmental Financial Relations

There is no single model when it comes to deciding who should be responsible for the design of IGfRs. Institutional arrangements vary across multilevel States with wide differences in the operation and membership of the relevant decision-making bodies or consultative commissions.13 With the development of the modern welfare state,14 the design of IGfRs serves the purpose of creating a viable State that provides for the delivery of basic goods and services to all its citizens by allocating resources in a decentralised way.15 Such an approach has been criticised by academics from the public choice school, who stress the fact that a decentralised allocation of resources does not per se lead to success. 11 12 13

14 15

See B. Dafflon, “Solidarity and the Design of Equalization: Setting out the Issues”, eJournal of Tax Research, 10 (2012) 138–164. See Boadway and Shah, Fiscal Federalism, supra, at 464–497. A. Shah, “Introduction: Principles of Fiscal Federalism”, in A. Shah (ed.), The Practice of Fiscal Federalism: Comparative Perspectives (Montreal: McGill–Queen’s University Press, 2007) 3–43, at 35–37. See T.H. Marshall, Citizenship and Social Class: And Other Essays (Cambridge: Cambridge University Press, 1950). See R. Musgrave, The Theory of Public Finance (New York: McGraw-Hill, 1959).

228

Alber

Therefore, any decision-making process related to public goods has to provide for institutional constraints by means of an adequate system of IGfRs.16 Academic debates among economists over the optimal distribution of powers and the costs and benefits of decentralisation in the 1950s and 1960s were accompanied by the development of a more dynamic theory of federalism that, in the end, acknowledges public finance (and thus also IGfRs) as one of the main pillars when it comes to the analysis of the interplay of orders of government in federalism studies, also in disciplines other than economics.17 To begin with, federal scholars, back then, acknowledged that there was too little scope for intergovernmental relations with regard to the complexity of federal reality in the writings of earlier scholars of federalism. Kenneth C. Wheare’s Anglo-American static understanding of federalism,18 with two orders of government being independent but coordinated, was increasingly criticised as being unrealistic.19 The European scholar Carl J. Friedrich introduced the understanding of federalism as a process by stressing the importance of communication systems in politics, whereby separate political communities agree to negotiate solutions or decisions on common problems.20 Likewise, the American scholar Daniel J. Elazar viewed federalism as “polycentric by design” like “a communications network that establishes the linkages that create the whole”.21 Moreover, several scholars urged a shift in the focus of federalism scholarship to exploring the actual operation of federal constitutions in the context of federal societies.22 Ronald L. Watts defined processes and institutions to facilitate intergovernmental collaboration in those areas where 16 17 18

19

20

21 22

See J.M. Buchanan and G. Tullock, The Calculus of Consent: Logical Foundations of Constitutional Democracy (Ann Arbor: University of Michigan Press, 1962). See G.G. Carboni, “Il federalismo fiscale: dalla nozione economica a quella giuridica”, Diritto pubblico comparato ed europeo, 4 (2009) 1417–1442. See K.C. Wheare, Federal Government (New York: Oxford University Press, 1947). For an extensive review of the works of K.C. Wheare, W.A. Livingston, W.H. Riker, C.J. Friedrich and D.J. Elazar, see M. Burgess, In Search of the Federal Spirit (Oxford: Oxford University Press, 2012). Ex plurimis see A.H. Birch, Federalism, Finance and Social Legislation in Canada, Australia and the United States (Oxford: Clarendon Press, 1955); D.J. Elazar, The American Partnership: Intergovernmental Cooperation in the Nineteenth-Century United States (Chicago: University of Chicago Press, 1962). See C.J. Friedrich, “Federal Constitutional Theory and Emergent Proposals”, in A.W. Macmahon (ed.), Federalism: Mature and Emergent (New York: Russell and Russell, 1962) 510–533. D.J. Elazar, Exploring Federalism (Tuscaloosa: University of Alabama Press, 1957), at 13. See W.S. Livingston, “A Note on the Nature of Federalism”, Political Science Quarterly, 67 (1952) 81–95.

Intergovernmental Financial Relations

229

­government responsibilities are shared or inevitably overlap as structural characteristics of a federal system.23 Most importantly, in his work, he continued to point out that in multilevel States the appropriate power for allocating expenditures tends to be the one closest to the people, thus the lower levels, but that the appropriate power for raising revenue tends to be at intermediate and higher levels. While, as a general rule, the share of expenditure responsibilities borne by subnational units tends to be relatively similar across multilevel States, there are significant differences with regard to how those units finance their spending (through their own revenues and the breadth of tax sources to which they have access) and to how the designs of fora where conflicts over IGfRs are to be played out look like.24 In theory, the design and functioning of such fora should be held rather simple in order to be easily understood and to be able to timely adapt to new circumstances in a straightforward manner. In practice, the design and especially the working of such fora cannot be else than a complex task and continuous work in progress because of its contextualisation in political dynamics. Depending on how they are designed, they differently impact the political system in place and, in their functioning, they are differently influenced by the system there are embedded in. This, ultimately, also leads to different answers with regard to the question to what extent budgetary policies are controlled by the executive or the legislature.25 While today the importance of intergovernmental relations in any multilevel State is widely recognised and its various aspects have been extensively studied, relatively little research has been done on classifying the institutional design of IGfRs. Only a few classification systems have been proposed. They especially refer to transfers and equalisation schemes, and they are based on five categories of archetypes that are not mutually exclusive.26 In unitary 23

24

25 26

See R.L. Watts, “The Federal Idea and its Contemporary Relevance”, in Courchene et al., The Federal Idea, supra, 13–27. For a comprehensive analysis of the works of Watts on intergovernmental relations, see R. Agranoff, “R.L. Watts and the Managing of igr in Federal Systems”, in Courchene et al., The Federal Idea, supra, 269–287. See Watts, Comparing Federal Systems, supra, 113–114; R.L. Watts, “Fiscal Federalism in Canada, the usa and Germany”, Working Paper iigr, Queen’s University (2004); and R.L. Watts, “Autonomy or Dependence: Intergovernmental Financial Relations in Eleven Countries”, Working Paper iigr, Queen’s University (2005). See I. Lienert, “Who Controls the Budget: The Legislature or the Executive?” imf Working Paper WP/05/115, (2005). See A. Shah, “Institutional Arrangements for Intergovernmental Fiscal Transfers and a Framework for Evaluation”, in Boadway and Shah, Intergovernmental Fiscal Transfers, supra, 293–317; and J. Boex and J. Martinez Vázquez, “Developing the Institutional

230

Alber

States, one or more government agencies are normally in charge of setting up, regulating and managing institutions that deal with taxing powers, revenuesharing, conditional and unconditional transfers, and equalisation mechanisms. Even though the national legislature has to enact legislation to provide a legal basis for such arrangements, in the “National Government Model”, it is usually the Prime Minister’s Office or the Ministry of Finance or the Ministry of Territorial Autonomies that assumes responsibility. Federal States normally adopt the “Intergovernmental Forum Model”, in which representatives from the lower level of government take part and have authority in co-determining IGfRs at the central level. Even though different in detail, this is the case of the Canadian and German federations. In federal States characterised by the separation of executive and legislative powers within each tier of government (e.g. United States of America, us), the details of IGfRs are primarily tackled within the federal legislature, while in federal States characterised by “fused parliamentary executives”, decisions with regard to IGfRs are primarily taken by means of negotiations between the executives.27 Intergovernmental fora may also be assigned the authority to decide on IGfRs after an independent agency or commission makes its recommendations. Australia, India and South Africa are examples of multilevel States that have adopted such a model. Independent agencies or commissions can be permanent (e.g. in Australia and in South Africa) or brought into existence periodically (e.g. in India). They report to the executive or legislature, and underlying principles related to their operational structure and mandate are normally enshrined in the federal constitution. Their composition may also include civil society representatives (e.g. the Provincial Finance Commission in Pakistan).28 However, guaranteeing such a presence does not necessarily lead to successful negotiations over IGfRs. On the contrary, it might be permanently deadlocked, especially if it is subject to the unanimity rule as in Pakistan. The “Subnational or Local Government Forum Model” is rarely foreseen (e.g. in the Netherlands, Denmark, Estonia and Latvia), mainly because it does not play a fundamental role in designing IGfRs. The same is valid for independent grants’ commissions that are established at the subnational level as advisory bodies for state-local fiscal transfers. Finally ­Framework for Intergovernmental Fiscal Relations in Decentralizing ldtcs”, International Studies Program, Working Paper 04–03, Andrew Young School of Policy Studies, (2001). 27 Watts, Comparing Federal Systems, supra, at 112. 28 See S.J. Ahmed, “Intergovernmental Relations in the Federal System of Pakistan”, in A. Faiz (ed.), Making Federation Work (Karachi: Oxford University Press, 2015) 97–120. See also A.G. Pasha and H.A. Pasha, “Financial Implications of Devolution in Pakistan”, in Faiz, Making Federation Work, supra, 121–146.

Intergovernmental Financial Relations

231

yet importantly, some multilevel States do not foresee any coordinating institutions when it comes to IGfRs (e.g. the us). This “No Forum Model” may be intentional or simply the result of a lack of institutional capacity. 4

Institutions, Rules and Praxis: Evidence from North America and Australia

4.1 United States of America While in Canada and in Australia the design of IGfRs is, for the most part, decided upon by executive fora, in the us, the Congress annually negotiates among representatives of different states over the allocation of grant programmes, and representatives of state administrations act as lobbyists. The us Constitution (in force since 1789) is silent when it comes to intergovernmental transfers: there is no explicit reference to the grants-in-aid system. The primary forum involved in financial issues is thus the federal legislature, even though the us Congress does, in the end, play a small role in monitoring IGfRs.29 Grants-in-aid programmes are the key mechanism that the federal government uses to extend its power into state and local affairs, and by which the federal government, in practice, turned the us system from an originally conceived dual system30 to a rather cooperative one. Grant programmes consist in subsidies that come part and parcel with federal regulations designed to (co-) manage state and local activities. Categorical grants are federal money given to a state for a specific purpose, while block grants are federal money given to a state for a broad purpose. The allocation of grants may be subject to matching components and specific formulas. As the us Constitution does not provide any form of general revenue-sharing, the federal intergovernmental grant system provides for a certain degree of equalisation across the us. It has, however, to be pointed out that the grant system does not at all equalise subnational government service delivery, because most of the money is intended to support low-income individuals.31 29

30

31

T.E. Smith, “Intergovernmental Relations in the United States in an Age of Partisanship and Executive Assertiveness”, in J. Poirier et al. (eds.), Intergovernmental Relations in Federal Systems (Don Mills: Oxford University Press, 2015) 411–439, at 416–417. The term “dual federalism” was coined by Edward S. Corwin. See E.S. Corwin, The Twilight of the Supreme Court: A History of Our Constitutional Theory (New Haven: Yale University Press, 1934), 1–51. W.F. Fox, “The United States of America”, in Shah, The Practice of Fiscal Federalism, supra, 344–367, at 361.

232

Alber

As said and in contrast to the German Basic Law (see further in Part 6.1), the us Constitution is characterised by the absence of a comprehensive body of principles and rules devoted to IGfRs. The us Constitution only includes a few articles in the fields of taxation, spending and borrowing powers. In particular, it grants the Congress the power to impose and collect taxes, duties, imposts and excises with the limitations that all duties, imposts and excises must be uniform throughout the us (Article 1.8.1. us Constitution).32 It does not explicitly enumerate the power of taxation assigned to the states. However, since the states do retain residual powers, they and—at their discretion—local government enjoy broad authority over fiscal policy in the fields of imposing taxes, occupying tax bases, varying tax rates, setting budget priorities and determining their own fiscal rules with regard to balanced budgets and debt. In practice, federal, state and local governments in the us overlap considerably in their use of revenue sources, and states necessarily cooperate in collecting taxes by entering agreements with other states (e.g. through a Multi-State Tax Commission or the Streamlined Sales and Use Tax Agreement).33 Acknowledging that tax policy decisions made by one order of government have implications for other governmental tiers, the question arises to what extent such implications are taken into account by policy-makers. Even if very little, in the us, there is some evidence that such implications are seriously taken into account when policy decisions are made.34 History shows that fiscal interdependence between the states and the federal level has, over time, increased, although the degree of fiscal autonomy of each order of government in the us is unknown to almost all other countries. In the 1930s, the New Deal policies under President Roosevelt ushered in new federal legislation enacted in response to the Great Depression that implemented several programmes and policies geared towards reviving the economy. As a consequence, the federal government started to regulate areas it had not previously been regulating.35 The dualist principle in the division of governmental functions between the federal government and the states was discarded in favour of shared management in fields such as welfare, housing and transportation. 32

33 34 35

Ibid., at 358–361. See also W. Hellerstein, “The Unites States”, in G. Bizioli and C. Sacchetto (eds.), Tax Aspects of Fiscal Federalism: A Comparative Analysis (Amsterdam: ibfd, 2011) 25–78. Fox, “The United States of America”, supra, at 358–361. See also J. Parker, Comparative Federalism and Intergovernmental Agreements (New York: Routledge, 2015), 163–164. Fox, “The United States of America”, supra, at 359ff. See M. Grodzins, The American System: A New View of Government in the United States (Chicago: Rand McNally & Company, 1966).

Intergovernmental Financial Relations

233

Initially, states were against the expansion of federal aid programmes, but their resistance was weakened by the fact that residents would still have to pay federal taxes to support the overall implementation of these programmes. Throughout the 1960s, the number of federal aid programmes reached another high point: ­grants-in-aid programmes were provided for housing, urban renewal, education, healthcare and many more activities. However, mismanagement in service delivery at the state and local level continued, and the federal ­governments of the 1970s and 1980s tried to reverse the trend by reducing grant programmes (known as the era of New Federalism). Today, there are more than 1,000 grants-in-aid programmes, with each programme having its own rules and regulations.36 Altogether, the us system is characterised by what is called “fend-foryourself federalism”,37 in which—even if, in practice, attenuated—states can levy any conceivable tax and spend and borrow money without limit unless there are restrictions imposed by the federal constitution or their own state constitutions.38 Even though the system has become more cooperative (not least because of the Congress’s control over interstate commerce),39 it still gives rise to very diverse patterns of taxing and spending at the state level, with significant variations in both state personal income tax structures and rates, and radical differences with regard to a state’s capacity and willingness to raise tax revenues.40 Today, intergovernmental relations in the us are tremendously fragmented, with narrow functional coordination and very little systemic direction.41 They have been described as so intricate that it is “difficult to generalize succinctly”

36

37

38

39

40 41

See C. Edwards, “Federal Aid-to-State Programs Top 1,100” Tax & Budget Bulletin, (2011), www.cato.org/publications/tax-budget-bulletin/federal-aidstate-programs-top-1000 (accessed 3 August 2016). See T. Conlan et al., “Autonomy and Interdependence: The Scope and Limits of ‘Fend for Yourself Federalism’ in the United States”, in J. Kim and H. Blöchliger (eds.), Institutions of Intergovernmental Fiscal Relations: Challenges Ahead (Paris: oecd Publishing, 2015) 155–179. J. Kincaid, “The Constitutional Frameworks of State and Local Government Finance”, in R.D. and J. Peterson (eds.), The Oxford Handbook of State and Local Government Finance (New York: Oxford University Press, 2012) 45–82, at 76. Article 1, Section 8, clause 3 of the us Constitution gives the Congress the power “to regulate commerce with foreign nations, and among the several states, and with the Indian tribes”. Fox, “The United States of America”, supra, at 354–355. Smith, “Intergovernmental Relations in the United States”, supra, at 433.

234

Alber

about their practical operation.42 Nearly every policy has an intergovernmental aspect characterised by federal coercive features throughout its creation and significant administrative cooperation in its implementing phase. Despite this, the us does not have any agency in charge of monitoring or coordinating intergovernmental relations. From 1959 to 1996, the Advisory Commission on Intergovernmental Relations (acir) existed as a forum that convened federal, state and local officials. The acir was born out of the conviction that a dual system of government is desirable and could function well if supported by a consultative body that, among other tasks, elaborated recommendations on the allocation of revenues and the coordination of tax laws.43 But the acir was never ascribed a truly significant role in the formation or monitoring of us fiscal policies even though it regularly delivered technical assistance on the us grants’ system and on issues of administrative cooperation. From the 1990s onwards, the acir was increasingly considered to be a nonessential body to the Congress, primarily for political reasons related to the management of the federal deficit, and it stopped operating in 1996. 4.2 Canada Canada is characterised by its ineffectual bicameralism,44 and its interstate model of federalism allows for the separation of jurisdictional authority between territorial actors and the predominance of competition and bipolarity according to the principles of dual federalism. With regard to the tax system, Canada has a competitive system, albeit with shared collection arrangements for the income tax as well as a number of other taxes, while Germany has a cooperative one (joint taxation system). On the expenditure side, Canada’s federal government can (quickly) cut expenditures, whereas the German government is constrained by consensus. In contrast to Germany, the Canadian provinces do not have any direct say in federal legislation or its implementation. They are autonomous in using their own legislative powers and in 42

43 44

J. Kincaid, “Intergovernmental Relations in the United States of America”, in J.P. M ­ eekison (ed.), Intergovernmental Relations in Federal Countries (Ottawa: The Forum of ­Federations, 2001) 33–44, at 33. See B.D. McDowell, “Advisory Commission on Intergovernmental Relations in 1996: The End of an Era”, Publius: The Journal of Federalism, 27 (1997) 111–127. The federal legislature consists of the House of Commons (308 members elected for five years) and the Senate (105 members that are appointed by the governor-general upon the recommendations of the prime minister and, as such, it ineffectively represents provincial interests). M.A. Adam et al., “Intergovernmental Relations in Canada: Competing Visions and Diverse Dynamics”, in Poirier et al., Intergovernmental Relations, supra, 135–173, at 141–143.

Intergovernmental Financial Relations

235

c­ arrying out their own policies. When it comes to intergovernmental relations, executives play a pivotal role, while federal and provincial parliaments do not. Intergovernmental relations may entail multilateral relations between federal and provincial executives or bilateral relations between the federal executive and one province.45 Interprovincial cooperation is far less forthcoming with regard to IGfRs, also due to the fact that provinces have the power to control their own natural resources.46 The Canadian Constitution, adopted in 1867 and significantly amended in 1982, devotes Sections 91 and 92 to the allocation of taxing powers. It specifies that it is the exclusive legislative authority of the parliament of Canada to set all kinds of taxes, while the provinces’ exclusive power is to impose direct taxation and license fees. Canadian provinces and cities collect approximately half of tax revenues, while their expenses amount to two-thirds of public spending. Federal transfers more or less make up the difference and amount to close to one-quarter of the federal budget.47 In recent years, federal transfers to Canadian provinces and territories have grown substantially, increasing by 62.3 per cent from 2005–2006 to 2015–2016, a rate much higher than would have been required to keep pace with inflation and population growth.48 The primary legal responsibility for the design of fiscal transfers to the Canadian provinces and territories lies with the federal government. However, the federal government has a weak scope as far as implementation is concerned since the provinces enjoy substantial autonomy. It can gain leverage through negotiations and the provision of incentives, but more recently the strategy has rather been to withdraw from such incentives (e.g. shared-cost programmes) and leave market forces to impose fiscal discipline on the provinces. Like the constitutions of other federal States, the Canadian one also enshrines the equality principle. Section 36.1 refers to the commitment of governments and legislatures to promote equal opportunities for the well-being of Canadians, and, accordingly, Section 36.2 requires the parliament and the government of Canada to provide for equalisation payments to ensure that provincial governments have sufficient revenues to deliver reasonably comparable levels of public services 45

46 47 48

Ibid., at 146. For example, the provinces Alberta as well as Newfoundland and Labrador benefit from large tax revenues (royalties) thanks to abundant natural resources (mainly oil and gas). Ibid., at 141. Ibid., at 159. See extensively in B. Eisen et al., “Are the Provinces Really Shortchanged by Federal Transfers”, Fraser Institute, (2016) www.fraserinstitute.org/studies/are-the-provinces-really -shortchanged-by-federal-transfers (accessed 3 August 2016).

236

Alber

at reasonably comparable levels of taxation. Entitlement to equalisation is designed to compensate the poorer provinces for their lack of fiscal capacity, and it is calculated on the basis of the standard tax rate in the median five provinces of 33 separate taxes.49 Since, in Canada, both the federal and the provincial level have taxing powers that often relate to the same tax base, there is a real risk of overlapping jurisdictions and conflicts.50 Thus, IGfRs do play a big role, and the volume of intergovernmental discussions has constantly increased, particularly in sectoral policy areas. This growth is notably due to the fact that the federal level increasingly intervenes in areas of provincial responsibility by means of its federal spending power (e.g. in health and education). The federal spending power51 is the most important lever of power for the federal government, and it is one of the most controversially discussed instruments in Canada. It is not constitutionally enshrined,52 but it has been widely recognised in the jurisprudence of the Supreme Court of Canada (most notably in Reference Re Canada Assistance Plan 1991 2 s.c.r. 525). In contrast to the us, federal government spending in Canada has, however, not led to an extensive centralisation: in the us, 71 per cent of total government spending is federal, and 29 per cent is state-local, while in Canada it is the reverse, with 38 per cent being federal and 62 per cent being provincial-local.53 Unlike in South Africa (see further in Part 5.2), in Canada, there is no constitutional provision or specific legislation governing intergovernmental relations, and the courts are rather wary of intervening in what they perceive to be a political question. Nevertheless, today, Canadian intergovernmental relations can be considered as rather formalised if compared to other countries (e.g. United Kingdom, see further in 49 A. Trench, “Intergovernmental Relations in Canada: Lessons for the uk”, The Constitution Unit (2003), https://www.ucl.ac.uk/spp/publications/unit-publications/101 .pdf (accessed 3 August 2016), at 11. 50 On the tax system, see B. Alarie and R.M. Bird, “Canada”, in Bizioli and Sacchetto, Tax Aspects of Fiscal Federalism, supra, 79–136. 51 It is worth noting that, in recent decades, the expansion of the federal spending power had detrimental effects on decentralised spending almost all over the world. For a discussion of the spending power in federations, see R.L. Watts, The Spending Power in Federal Systems: a Comparative Study (Kingston: Queen’s University, 1999). For the Canadian case, see E.A. Driedger, “The Spending Power”, Queen’s Law Journal, 7 (1981) 124–134. 52 Peter Hogg argues that the spending power must be inferred from the powers to levy taxes [Section 91(3)], to legislate in relation to public property [Section 91(1A)] and to appropriate federal funds (Section 196). P.W. Hogg, Constitutional Law of Canada (2nd ed, Toronto: Carswell, 1985) at 124. 53 C. Edwards, “Canada’s Fiscal Reforms”, Cato Journal, 22 (2013) 299–306, at 304.

Intergovernmental Financial Relations

237

Part 6.3). Political practice gave birth to a system that, for the most part, consists of numerous exchanges of letters, meetings and conferences among executives of various levels. Next to these forms, intergovernmental agreements have proliferated as an instrument of Canadian collaborative federalism (e.g. tax ­collection agreements).54 Summing up, in the field of IGfRs, three elements are central to Canada: federal-provincial taxation agreements, federal conditional and block grants, and equalisation payments. Coordination takes place primarily by means of political negotiations and is aimed at ensuring that, collectively, governments do not overtax individuals and businesses. In practice, the federal government makes use of its spending power in order to support the delivery of territorial services (e.g. in healthcare and education). Especially provinces that are financially self-sufficient have contested such federal intervention. This triggered the proliferation of agreements. For example, ongoing conflicts led to the constitutionally non-binding 1999 Social Union Framework Agreement (sufa) signed by the federal and provincial governments with the exception of Quebec. Under sufa, the governments agreed to regularly exchange views on the development and implementation of social policies, and, most importantly, with regard to limiting the federal spending power; basically, the federal government agreed not to introduce new social programmes that are funded through intergovernmental transfers without the agreement of a majority of provincial governments.55 4.3 Australia Unlike Canada, the collection of revenue in Australia became centralised over time, and this, in particular, was the reason why Australia turned from being a federation based on dual federalism to a centralising one.56 Judicial interpretation of the Australian Constitution (passed by the British Parliament as part of the Commonwealth of Australia Constitution Act 1900 and in operation since 1 January 1901) has expanded the Commonwealth’s power to collect revenue and distribute financial assistance to the states and territories, also by tying grants for financial assistance to specific conditions.57 The terms of 54 Parker, Comparative Federalism and Intergovernmental Agreements, supra, at 69–70. 55 Adam et al., “Intergovernmental Relations in Canada”, supra, 135–173, at 156. 56 See A. Fenna, “The Character of Australian Federation”, eJournal of Tax Research, 10 (2012) 12–20. 57 The High Court expanded the Commonwealth’s power to collect revenue with the First Uniform Tax Case in 1942 (South Australia v Commonwealth 65 clr 373) and confirmed it in the Second Uniform Tax case in 1957 (Victoria v Commonwealth 99 clr 575).

238

Alber

“financial assistance”58 to poorer constituent units were calculated by reference to a fixed formula that was modified several times up until 1985, when the formula for funding arrangements was replaced by a consolidated funding pool, and distributions were made in accordance with the fiscal strategies of the Commonwealth. Today, the Australian federal government has full control over the biggest revenue levers and collects 85 per cent of all taxation while it is responsible for only half the expenditure on government services.59 For example, Australia’s 10 per cent Goods and Services Tax (gst, introduced on 1 July 2000 and distributed as an untied grant) is collected solely by the federal government, as are all personal and corporate income taxes. Discussions on tax decentralisation rarely pop up in Australia, while in Canada the idea of removing ­provincial autonomy over income and sales taxes by replacing provinces’ ownsource revenues with larger transfers from the federal level is far-fetched. Thus, ­Australia’s system has a very high degree of vertical fiscal imbalance and runs two risks: firstly, states might be encouraged in the direction of wasteful expenditure, and, secondly, the Commonwealth’s interventions might easily lead to conflicts with the states (“blame-shifting game”). To limit such risks, an Intergovernmental Agreement on Federal Financial Relations (iaffr) was agreed upon through the Council of Australian Governments (coag) in 2008. It entered into force on 1 January 2009 and represents the most significant shift in Commonwealth-states relations in decades. Its key objective is to increase the accountability of all government tiers by better clarifying the roles and responsibilities of each jurisdiction. It aims at improving the quality and effectiveness of government services by reducing the Commonwealth prescriptions, aligning grants with the achievement of better governance and giving the constituent units more freedom in deciding how to achieve this aim. Under the iaffr, all payments are processed centrally by the Commonwealth Treasury and paid directly to each state treasury. The states are then responsible for the ­allocation of resources in their jurisdiction.60 Basically, three types of payments are­

58 59

60

In both cases, the High Court established that the Commonwealth was solely responsible for collecting income tax, and thus for collecting the majority of taxation revenue. See P. ­Robinson and T. Farrelly, “The Evolution of Australia’s Intergovernmental Financial Relations Framework”, Australian Journal of Public Administration, 72 (2013) 304–315, at 307. Section 96 reads as follows: “The Parliament may grant financial assistance to any state on such terms and conditions as the Parliament thinks fit”. C. Saunders, The Constitution of Australia: A Contextual Analysis (Portland: Hart Publishing, 2011), at 237. On the tax system, see M. Stewart, “Australia”, in Bizioli and Sacchetto, Tax Aspects of Fiscal Federalism, supra, 137–186. See details at www.coag.gov.au/the_federal_financial_relations_framework (accessed 3 August 2016).

Intergovernmental Financial Relations

239

undertaken by the Commonwealth: (1) general revenue assistance, which includes gst revenue (gra); (2) national specific-purpose payments (nspps); and (3) national partnership payments (npps). Compared to Canada, the volume of all transfers is double the amount, and this results in greater equality between the constituent units. Since the federal government collects a much larger share of total tax revenue than in Canada, subnational economic cycles also play a lesser role with regard to delivering a comparable standard of services within each Australian state. Forms of payments and especially the nspps have long been a contested issue between the two tiers of government because they often came with very strict conditions. By means of the iaffr, the number of nspps dropped from over ninety to five, and each of the nspps is now supported by additional agreements in the areas of healthcare, schools, skills and workforce development, disability services and affordable housing.61 Even though states may spend nspps only in the sectors for which they are provided for, the margin of discretion over how the money can be spent has been increased. This is not the case of npps, payments to the states that relate to specific deliverables. The Commonwealth still has a predominant role in setting precise conditions when it comes to npps. Thus, also with the iaffr, the Commonwealth continues to finance programmes and projects over which it does not have direct legislative or constitutional authority. Similarly to the Canadian case, the processes for adjusting IGfRs in Australia have also been developed beyond the existence of constitutional provisions and to a certain extent also beyond any legislation. They are a result of political negotiations and of jurisprudence. For example, the principle of horizontal fiscal equalisation is not explicitly defined in legislation or in any intergovernmental agreement. Currently, the definition developed by the Commonwealth Grants Commission (cgc) is in use. Accordingly, fiscal equalisation is aimed at enabling each state “by reasonable effort to function at a standard not appreciably below the standards of other states”.62 Australia’s equalisation scheme, a “net scheme”, is thus aimed at adjusting for differences in the cost of providing services, and not only for differences in revenue-raising capacity (like the ­Canadian equalisation programme, which is a “gross scheme”).63 61

62 63

J. Phillimore and J. Harwood, “Intergovernmental Relations in Australia: Increasing Engagement within a Centralizing Dynamic”, in Poirier et al., Intergovernmental Relations, supra, 42–80, at 64–65. Commonwealth Grants Commission Act 1973, Section 5(1). Provinces that are below the standard fiscal capacity are equalised up, but provinces above the standard do not have any corresponding reduction in transfers. This has major implications for the amount of fiscal risk faced by the central government. See P. Boothe,

240

Alber

This, h ­ owever, in reality, is difficult to obtain as states may provide deliberately or by necessity different services and service levels to their inhabitants because of context-bound variables. Likewise to Canada, the strong dualist approach envisaged by the constitutional framers did not foresee the institutionalisation of intergovernmental relations. They developed in practice as a response to the need for more coordination across governmental levels. Today, intergovernmental relations include peak-level fora of government heads, both vertical and horizontal. Vertically, the coag was established in 1992 by an informal agreement between the prime minister and the state premiers.64 Its decisions are not binding and it works by consensus. It has developed its capacity as a co-operative executive body especially since the introduction of the gst. Horizontally, the Council for the Australian Federation was established in 2006 by a memorandum of understanding, and it was modelled on the Canadian Council of Federation (ccf).65 It serves the purpose of providing the leaders of states and territories with an opportunity to discuss matters related both to the coag and to cross-jurisdictional issues.66 Moreover, as in Canada, there are only a few, if any, formal co-operative mechanisms between the Commonwealth parliament and state and territory legislatures, or between state and territory legislatures themselves. Unlike in Canada, Australia has a long tradition of relying on independent agencies in the field of IGfRs, be they jointly created bodies across tiers of government or bodies established at the federal level only. For example, the cgc was established as an independent body in 1933 under federal legislation (Commonwealth Grants Commission Act 1973) and is funded by the federation.67 The Commonwealth Treasurer has always accepted its recommendations on horizontal fiscal equalisation. An important agency mandated to oversee subnational borrowing is the Australian Loan Council. It is a

64

65

66 67

“Taxing, Spending and Sharing in Federations: Evidence from Australia and Canada”, in P. Boothe (ed.), Fiscal Relations in Federal Countries: Four Essays (Ottawa: Forum of Federations, 2003) 5–16, at 13–14 and 16. See G. Anderson, “The Council of Australian Governments: A New Institution of Governance for Australia’s Conditional Federalism”, University of New South Wales Law Journal, 31 (2008) 493–508. See also Parker, Comparative Federalism and Intergovernmental Agreements, supra, at 56. Established in 2003, the ccf enables premiers to work collaboratively to strengthen the Canadian federation by fostering a constructive relationship among the provinces and territories and with the federal government. Phillimore and Harwood, “Intergovernmental Relations in Australia”, supra, at 46 and 53. R.L. Watts, “Processes for Adjusting Federal Financial Relations: Examples from Australia and Canada”, in Boothe, Fiscal Relations in Federal Countries, supra, 17–39, at 32.

Intergovernmental Financial Relations

241

ministerial council comprising the Commonwealth Treasurer as chair and the counterpart of each constituent unit. This body, established in 1927, originally had the task to coordinate state and federal economic and financial policies. Its mandate drastically changed when the federal government took over full control of fiscal matters during World War ii, and federal control over borrowing limits was only loosened again starting from the mid-1970s.68 Today, federal control still exists in the form of global borrowing limits linked to macroeconomic considerations. 5

Institutions, Rules and Praxis: Evidence from Asia and Africa

5.1 India India’s federal experience has evolved as a highly dynamic structure in order to accommodate its socio-economic, linguistic and religious diversity.69 No other recipe would work given the significant asymmetry in socio-functional responsibilities and in revenue-raising powers between both the Union and its constituent units and the constituent units themselves. As it is also the case in South Africa, the word federalism does not appear in the Indian Constitution. India is defined as a “Union of States” with sovereignty resting in the people and its constitution. India’s governance structure was largely modelled on the Government of British North America Canada Act of 1867 and, as such, it is characterised by a strong centre and a detailed distribution of powers between the centre and the states, with residual powers residing with the centre. There are three lists in the Seventh Schedule of the Indian Constitution (in force since 1950) that specify the legislative, executive and fiscal domains of the Union and the state governments in terms of Union, state, and concurrent lists (altogether more than 200 items), also including provisions as to the borrowing powers70 of the centre and the states. The Union and state lists also provide for powers of taxation, with the former including tax on income (other 68

69 70

See C. Saunders and K. Wiltshire, “Fraser’s New Federalism 1975–1980: An Evaluation”, Australian Journal of Politics and History, 26 (1980) 355–371. See also J.R. Madden, “Central Fiscal Dominance, Collaborative Federalism, and Economic Reform”, in J. Wallack and T.N. Srinivasan (eds.), Federalism and Economic Reform (New York: Cambridge University Press, 2006) 85–142, at 108–110. See B. Arora, “Pluralistic Federalism in India: Accommodation of Diversity”, in Faiz, Making Federation Work, supra, 311–333. Article 292 restricts the borrowing power of the Union to the limits set by the parliament, if any exist, and Article 293 restricts the states’ borrowing powers to Indian territory and within the limits set by the legislature along with entailing that the state cannot borrow

242

Alber

than agriculture), excise duties, customs, corporate and service tax, and the latter containing land revenue, excise on alcoholic liquor, tax on agricultural income, estate duty, tax on sales or purchases of goods, tax on vehicles, tax on professions and stamp duties.71 Most importantly, the Indian Constitution does not recognise any concurrent jurisdiction in the field of taxation. Thus, the assignment of taxing powers follows the principle of separation: they are either assigned to the Union or to the state governments. The Indian Constitution makes a broad distinction between the power to levy a tax or duty and the power to appropriate the respective proceeds. Thus, the government level that levies a tax or duty is not necessarily the authority that retains the respective proceeds. When it comes to revenue distribution, the Indian Constitution enumerates in a very detailed way which taxes and duties are levied, collected and retained by the Union (1), which are levied by the Union but collected and appropriated by the states (2), which are levied and collected by the Union even though the net proceeds go to the states (3), which are levied, collected and retained by the states (4), and which are levied and collected by the Union but the net proceeds are shared between the Union and the states (5). Category (5) is known as the “divisible pool” with personal income tax other than the one derived from agriculture and Union excise duty being the most notable ones. It has to be highlighted that with the 80th amendment to the constitution in 2000 all Union taxes were made shareable.72 In this scheme of resource distribution, the Union government ends up having more money than it needs, also because the most relevant sources of revenue are with the centre. Thus, Indian states are financially dependent on the centre to properly fulfil their functions (e.g. healthcare and education). Therefore, Article 275 of the Indian Constitution also provides for grants-in-aid programmes to the states that are assessed to be in “need of assistance”. A Finance Commission (fc), mandated by Article 280 of the Indian Constitution, advises the president on the distribution of financial resources between the Union and the states, including the grants-in-aid programmes. The Finance Commission Act of 1951 firstly defined the terms of qualification,

71 72

funds without the consent of the Union government if there is still any part of a central loan to the state that is outstanding. On the tax system, see S. Poddar and S. Mathur, “India”, in Bizioli and Sacchetto, Tax Aspects of Fiscal Federalism, supra, 187–219. Based on the recommendations of the Tenth Finance Commission, this new scheme for sharing taxes between the Union and the states was introduced with the Constitution (Eightieth Amendment) Act 2000. Overall, the states’ share was increased.

Intergovernmental Financial Relations

243

a­ ppointment and disqualification of the members of the fc as well as the task of the fc. The fc is appointed every five years by the president of India and consists of a chairperson and four other members. The fc has to both assess the Union government’s as well as states’ governments revenue and expenditure projections, and to make recommendations on those issues; moreover, the fc has to pronounce itself also on any other matter which the president may ask for in the specific terms of reference for each fc. As a general rule, the fc makes recommendations with regard to following four items: (1) the distribution between the Union and the states of the net proceeds of sharable taxes and their allocation between the states; (2) the definition of principles that should govern the distribution of grants-in-aid of revenues of the states out of the consolidated fund of India; (3) the measures needed to augment a state’s consolidated fund after having consulted the state finance commission; (4) any other issue aimed at developing sound finances across India.73 The chairperson of the fc is appointed from among persons of public eminence, and its four members should be qualified to become a judge of the High Court, have special knowledge of finances and accounts of the government or have wide experience in financial administrative or economic affairs. The members of the fc must provide full- or part-time service to the fc, as specified in the president’s order. According to Article 281 of the Indian Constitution, “The President shall cause every recommendation made by the Finance Commission […] together with an explanatory memorandum as to the action taken thereon to be laid before each House of Parliament”. At this point, it is worth pointing out that with regard to money bills the upper house, the Rajya Sabha, is a weak chamber.74 Originally, the fc was intended to express itself on all constitutionally enshrined financial transfers from the Union to the states. In fact, the first and second fc recommended financial assistance to cover both the revenue and the capital requirements of the states, but the establishment of the Planning Commission (pc) led to the bifurcation of this function with the fc deciding on tax shares and grants and the pc facilitating the allocation of loans and grants according to Article 282 of the Indian Constitution that allows the Union or a state to make any grants for any public purpose, notwithstanding that the purpose is not one with respect to which the parliament or the legislature of 73 The fc has used equity and efficiency as the two guiding principles. The indicators for horizontal sharing are population, area, fiscal capacity and fiscal discipline. G. Rao, “Republic of India”, in Shah, The Practice of Fiscal Federalism, supra, 151–177, at 164–165. 74 See, Rajya Sabha, “Legislative Functions of Rajya Sabha”, available at http://rajyasabha .nic.in/rsnew/legislation/introduction.asp (accessed 31 March 2017).

244

Alber

the state may make laws. The pc was set up in 1950 by a resolution of the cabinet, and, thus, it lacked any legislative source. Its members were appointed by the prime minister (which was also its chairperson), and they had the task of determining measures for balanced development policies. Over time, against the backdrop of the increasing dependence of states on pc transfers (often earmarked), the pc turned into an oversight commission with the last word on resource distribution. Moreover, the pc had to approve the so-called centrally sponsored schemes (css) from which better off states normally benefitted most as css frequently required co-financing up to 50 per cent by the receiving state. As such, the pc could not but enter into conflict with the functions and recommendations of the fc, also because over time fiscal transfers from the Union to the states became ever more discretionary than formula-based. Initially, the fc assigned the greater share of tax proceeds to the Union, but over time it steadily increased the share of tax proceeds assigned to the states; this trend has obviously to be read against the backdrop of the changing political landscapes in the states (i.e. the emergence of state parties). The respective fcs tried to evolve formulas that balance equity with fiscal efficiency, but, as a general rule, equity considerations prevailed in order to adhere to the equalisation principle. The abolition of the pc in 2014 and the increasing acceptance of the recommendations of the fc have been referred to as the strongest statement that the Modi government was serious about restructuring Union-state relations by cooperative federalism. The pc was replaced by the National ­Institution for Transforming India (niti Aayog), which is also under the direct authority of the prime minister. Where it differs from the pc, however, is that it actually aims to give states a role in its management; in other words, it intends to be a forum in which regional interests are sorted out. Unlike the pc, the niti Aayog has a governing council whose members also comprise chief ministers of states and, as such, it provides a platform that brings the Union and the states together in determining national development priorities. Compared to the pc that was required to work on five years plans regarding the allocation of funds, the niti Aayog’s is essentially an advisory body with, ultimately, the Ministry of Finance deciding on the allocation of funds. Accordingly, the niti Aayog is defined as “the premier policy ‘Think Tank’ of the Government of India, providing both directional and policy inputs”.75 It shall develop strategic and long-term policies and programmes for the Union government and provide relevant technical advice to both the Union and the states. Needless to say that, against the backdrop of its large membership, the success of the niti 75

See National Institution for Transforming India, “Overview”, available at http://niti.gov.in/ content/overview (accessed 31 March 2017).

Intergovernmental Financial Relations

245

Aayog to effectively work on intergovernmental issues primarily depends on political constellations. India’s constitutional set-up coupled with the adoption of a public-sector strategy dominated by planned development at the time of India’s independence led to the creation of a system of IGfRs with a high degree of centralisation, which has to a certain extent reversed itself and become ever more complex.76 Over time, India has undergone four important changes with significant implications for fiscal federalism:77 first, the collapse of one-party rule and the rise of regional parties;78 second, the emergence of regional parties with their agenda of state-centred policies; third, the increase of asymmetric arrangements in fiscal issues due to both strategic alliances of regional parties with the centre and the deregulation of economic policies; fourth, the declining time horizon of political parties and short-term agendas linked with competitive populism. States have started to compete with each other to attract private investment flows, and interstate disparities have started to grow. This has forced the Union and the states to better coordinate economic policies across the country. The adoption of a common Value Added Tax (vat) in 2005 through the mechanism of a group of state finance ministers was a major achievement in this regard, while the failure to agree to replace the vat with a nationwide common Goods and Services Tax (gst) has been seen as a problem for many years.79 The Constitution (One Hundred and First Amendment) Act 2016 finally introduces the gst and, within the first half of 2017, all implementing legislation was passed both at Union and state level (most importantly, with regard to the Central gst,80 the Integrated gst81 and the State gst82). The introduction of the gst (from 01 July 2017 onwards) is the largest-ever tax reform in fiscal 76 77 78

Rao, “Republic of India”, supra, at 152. Ibid., at 155. Despite its parliamentary-federal constitution, India, until 1989, operated as a predominantly centralised parliamentary system because of the predominance of the Indian National Congress. See M.P. Singh and R. Saxena, “Intergovernmental Relations in India: From Centralization to Decentralization”, in Poirier et al., Intergovernmental Relations, supra, 239–271, at 239. 79 S. Cnossen, “Preparing the Way for a Modern gst in India”, International Tax and Public Finance, 20 (2013) 715–723. 80 The cgst is to be levied and administered by the Union on intra-state supply of goods and services. 81 The igst is to be levied and administered by the Union on inter-state supply of goods and services. 82 The sgst is to be levied and administered by the states.

246

Alber

­history in India and overhauls Union-state relations according to the principles of cooperative federalism.83 It requires a paradigm change and, thus, a substantial realignment of taxing powers of the Union and the states.84 The gst replaces various taxes that were levied and collected by the Union (e.g. central excise duty, service tax) and states (e.g. state vat, central sales tax, purchase tax). The reasons for amalgamating a large number of Union and state taxes are manifold and have been intensively discussed for over fifteen years.85 In short, the dual gst regime should both pave the way towards a common national market and reduce the overall tax burden on goods for consumers. A gst Council consisting of both representatives from the Union and states’ governments had the task to advance recommendations with regard to the detailed design of the dual gst (i.e. the taxes to be subsumed under gst, goods and services subject to or exempted from gst, a gst state law model, the rate structure etc.). Moreover, the gst Council shall also provide for the establishment of a mechanism to adjudicate any dispute across governmental levels. The gst Council’s decisions require a majority of not less than three-fourths of the weighted votes of the members present and voting with the Union government having a weightage of one-third of the total votes cast and the votes of the state governments having a weightage of two-thirds of the total votes cast. Compared to the gst type of tax implemented in over 150 countries worldwide, the peculiarity of India’s gst regime lies in the fact that it is not a single national tax but a set of 38 taxes, which are supposed to be fully coordinated to appear like a single tax.86 Noteworthy is also the fact that the states have been assured compensation for possible revenue losses for a five-year period. 5.2 South Africa The Indian fc served as an example for the creation of the South African ­Finance and Fiscal Commission (ffc). An important lesson that South A ­ frica learned from India was the fact that the Indian fc is not a permanent body but appointed every five years. As such, its impact is reduced. This was the 83

Central Board of Excise & Customs, “faqs on Goods and Services Tax”, (New Delhi, 31 March 2017), www.gstindia.com/about/ (accessed 31 March 2017). 84 See M.G. Rao, “Unfinished Reform Agendum: Fiscal consolidation and Reforms—A Comment”, in J. Bhagwati and C.W. Colomiris (eds.), Sustaining India’s Growth Miracle (New York: Columbia Business School, 2008) 104–114. 85 See S. Poddar and E. Ahmad, “gst Reforms and Intergovernmental Considerations in India”, Ministry of Finance Working Paper 1/2009-DEA, (2009). 86 See ey Tax Insights, “gst—A New Era of Cooperative Federalism in India”, available at http://taxinsights.ey.com/archive/archive-articles/gst-a-new-era-of-cooperative-federal ism-in-india.aspx (accessed 31 March 2017).

Intergovernmental Financial Relations

247

major reason why the ffc was created as a permanent body. The South ­African Constitution (Constitution of the Republic of South Africa, Act 108 of 1996) provides for the ffc as an independent advisory body with a mandate to make recommendations regarding several aspects of IGfRs to the parliament, the provincial legislatures and any other authorities determined by national legislation (Section 220).87 Compared to its Australian counterpart, the cgc, the ffc does not limit itself to issuing advice on the distribution of revenue and taxing powers, but it can also advise on taxation issues and any aspect related to financing public services. While recommending an equitable division of revenues among governmental tiers, it has to consider, among other things, the fiscal capacity of provinces and municipalities, economic disparities and, in case of emergencies, the need for flexibility. The fcc is a regular actor in South Africa’s IGfRs because national legislation dealing with the distribution of revenue and grants to provinces and municipalities may not be passed before the parliament both has consulted the fcc and has proven to have considered its recommendations. The members of the fcc—a chairperson, a deputy chairperson and seven other individuals—are, as in India, appointed by the president and must have appropriate expertise (Section 221 of the South ­African Constitution). Initially, the fcc included nine individuals nominated by the provinces, two nominated by organised local government and nine other individuals. The South African Constitution was amended in 2001 (Act No. 61) in order to have a smaller commission because the larger one proved to be unwieldy and difficult to manage. Today, there are three individuals representing the provinces who are selected from a composite list of provincial nominees, the local government representatives are kept at two and are selected after having consulted organised local government, and the chairperson and his or her deputy are appointed by the president. The fcc members can only be removed by the president because of misconduct or incompetence, and this procedure has, so far, never been invoked.88 fcc members may attend the meetings of two other bodies that are of importance to South Africa’s IGfRs: the consultative Budget Council consisting of the minister of finance as its chairperson and the nine provincial members of the executive council (mec) responsible for finances, and the Budget Forum, which was established in order to include the dimension of local government 87 88

See several acts of parliament Financial and Fiscal Commission, “Mandate of the ffc”, available at www.ffc.co.za/index.php/ffc-about-us/mandate (accessed 3 August 2016). See J. de Visser and Z. Ayele, “Intergovernmental Fiscal Relations in South Africa and the Role of the Financial and Fiscal Commission”, Conference Paper, (2014), available at www.ffc.co.za/conference2014 (accessed 3 August 2016).

248

Alber

and which consists of the minister of finance and the nine mecs for finance and representatives of organised local government. It should be noted that the Budget Council is increasingly being used as a management tool for the minister of finance to ensure that national policies are carried out effectively in the provinces.89 Next to an extended cabinet meeting that includes the nine premiers and the mecs responsible for finance, these are the most relevant bodies in the consultative process that leads to the enactment of the annual Division of Revenue Act. As in any other state, senior officials work for the hands-on implementation of IGfRs, ensuring, among other things, the development and implementation of the Medium-Term Expenditure Framework (mtef)90 according to the general principles of intergovernmental relations enshrined in the South African Constitution. In contrast to Canada and the us, South Africa’s Constitution is imbued with the letter and spirit of intergovernmental relations.91 Chapter 3 of the Constitution refers to the importance of “cooperative government” between the respective “spheres” of government and mandates the promulgation of an act of parliament laying down all details as to intergovernmental relations. For political reasons, such an act was enacted only in 2005 (Intergovernmental Relations Framework Act, Act 13 of 2005, hereinafter the “Intergovernmental Relations Act”). Taking inspiration from the principle of German Bundestreue (trust, partnership and comity in federal-state relations),92 the Intergovernmental Relations Act codified the existing informal arrangements in the spirit of the principle of cooperative government. Such a codification of intergovernmental relations is unique within the panorama of modern constitutions. In older federations, intergovernmental relations normally developed over time, and the composition, powers, frequency of meetings and dispute-resolution mechanisms of their bodies are not nearly as institutionalised as to as those in South Africa are. With regard to taxation, the South African Constitution grants revenueraising powers primarily to the national (personal and corporate income tax, vat and customs duties) and local spheres of government, with the provinces having a limited scope on taxation even though they have large expenditure powers (e.g. health and education). Therefore, Section 214 of the constitution 89 90 91

92

D. Brand, “Financial Intergovernmental Relations in South Africa”, Konrad-AdenauerStiftung Johannesburg, Policy Paper No. 1, (2007), at 4. Ibid., at 5. See B. de Villiers, “Codification of ‘Intergovernmental Relations’ by Way of Legislation: The Experiences of South Africa and Potential Lessons for Young Multitiered Systems”, ZaöRV, 72 (2012) 671–694. Ibid., at 679–680.

Intergovernmental Financial Relations

249

mandates the enactment of an act of parliament that provides for an equitable division of revenues among the three tiers of government. In practice, the largest proportion of the provinces’ budgets consists of unconditional and conditional transfers from the national sphere, with all the (negative) implications for the functioning of IGfRs due to the predominance of the African National Congress party.93 The National Council of Provinces’ committee on finance plays a visible role in vetting bills dealing with the distribution of revenue among all spheres of government. This is crucial because over 95 per cent of the funds for delivering provincial programmes are transferred by the national sphere. 6

Institutions, Rules and Praxis: Evidence from Europe

6.1 Germany Germany’s Basic Law (Grundgesetz, 1949) refers to IGfRs and, in general, to its “financial constitution” in Articles 104a-115 of Title x: it lists the taxes that belong exclusively to the federation (Bund) and its constituent units (Länder), mentions the taxes that are shared among the different tiers of government and fixes the percentage of tax sharing with regard to each tier. Article 109(1) of the Basic Law provides that the Bund and the Länder conduct their own autonomous budget policies. However, the Länder’s autonomy in conducting budget policy is weakened for three reasons: first, Länder cannot decide for themselves on their revenues; second, a large amount of the Länder budgets is already predetermined by federal legislation; third, coordination of Länder budgets and financial planning is centralised and not part of the functional autonomy of the Länder. The distribution of revenues between the federal government and the Länder does not reflect the competitive federalism model in which taxing and spending powers are either assigned to the federal or subnational level. On the contrary, the Basic Law provides for a complex system of cooperative federalism with a strongly harmonised tax system in which the Bund has the predominant role, and a four-stage tax redistribution and equalisation system with two vertical and two horizontal components is in place.94

93

94

D. Powell, “Constructing a Developmental State in South Africa: The Corporatization of Intergovernmental Relations”, in Poirier et al., Intergovernmental Relations, supra, 305– 349, at 305–306. See J. Woelk, “I rapporti finanziari nell’ordinamento tedesco”, in F. Palermo and M. ­Nicolini (eds.), Federalismo fiscale in Europa: Esperienze straniere e spunti per il caso italiano (­Napoli: Edizioni Scientifiche Italiane, 2012) 15–42.

250

Alber

The equalisation system functions as a “peak equalisation” and, as such, it has been characterised as a “zero-sum game”.95 The Länder can exert their influence through the Federal Council (Bundesrat), because legislation in fiscal and financial matters requires its approval (Article 105(3) Basic Law).96 Legislation in fiscal and financial matters has to be submitted for approval to the Bundesrat, where the Länder governments are represented. This system enables the participation of each Land and ensures negotiation and consensus in the whole process of shaping IGfRs. However, the flipside is that this system induces decision-making by agreeing on a minimum denominator or preservation of the status quo.97 Thus, on the one hand, Germany’s intrastate federalism allows for decision-making in fiscal federalism at the federal level, where the Länder, which are responsible for implementing federal legislation, have their say. On the other hand, the German system bears the risk of a stalemate when there are different party majorities in both chambers, and Land representatives vote on party lines. The votes in the Bundesrat are weighted with a number of votes ranging from three to six, and representatives from the Land governments are bound by the instructions from their respective Land government when casting their vote en bloc.98 The ability of the Bundesrat to serve as an institution to facilitate intergovernmental relations is conducive to the fact that, unlike other federal second chambers, the German one is composed of instructed delegates representing the Land governments.99 95

96

97 98

99

J.W. Hidien, “Nichts ist unmöglich: Aus dem Kuriositätenkabinett des bundesstaatlichen Finanzausgleichs”, Die öffentliche Verwaltung, 21 (1999) 903–909, at 908. See also H. Scheller, “Der Finanzausgleich zwischen Bund und Ländern als föderatives Sozialversicherungssystem”, in ezff (eds.), Jahrbuch für Föderalismusforschung 2005 (Baden-Baden: Nomos, 2005) 253–269. The scope of this provision is broad. For details, see A. Valdesalici, “Financial Constitutions and Responsibility at the Margin: From Legal Framework to Practice”, available at www.foederalismus.at/foederalismuspreis/foederalismuspreis_valdesalici.pdf (accessed 3 August 2016). See also J. English and H. Tappe, “The Federal Republic of Germany”, in Bizioli and Sacchetto, Tax Aspects of Fiscal Federalism, supra, 273–326. See F.W. Scharpf, “The Joint-Decision Trap: Lessons from German Federalism and European Integration” Public Administration, 66 (1988) 239–278. R. Lhotta and J. van Blumenthal, “Intergovernmental Relations in the Federal Republic of Germany: Complex Co-operation and Party Politics”, in Poirier et al., Intergovernmental Relations, supra, 206–238, at 213–214. On the (sui generis) classification of the Bundesrat, see M. Kotzur, “Federalism and Bicameralism: the German ‘Bundesrat’ (Federal Council) as an Atypical Model”, in J. Luther et al. (eds.), A World of Second Chambers: Handbook for Constitutional Studies on Bicameralism (Milano: Giuffrè, 2006) 257–292. See also R. Sturm, “Der Bundesrat im Grundgesetz:

Intergovernmental Financial Relations

251

Interestingly, fiscal federalism was a main driving force behind the development of intergovernmental relations in Germany, because, historically, in the founding years of Germany in the 19th century, the federation was dependent on the Länder budgets. This shifted in the course of time, and today a centrally coordinated management is in use in budget policy.100 Prior to the establishment of the Stability Council in 2010 (discussed further in Part 7.3), the Financial Planning Council (Finanzplanungsrat), an institution created in 1969 after the Grand Coalition of cdu/csu101 and spd102 adopted a Keynesian economic perspective and amended the rules for taking up Länder debts for the first time in the post-war period, had the task of coordinating fiscal policies in cooperation with the Council for Economic Development (Konjunkturrat). Both institutions consisted of the federal ministries for finance and economy, one member of each Land government and four municipal members. The results of the deliberations of these consultative councils consisted in resolutions comprising guidelines for budgetary decisions for time spans of five years. De facto, those bodies provided for a very loose and ineffective coordination, because a considerable share of tax revenue in Germany is determined by joint decision-making and also because the councils, in practice, did not coordinate instruments to stimulate economic cycles. The ineffectiveness of this system led to an increase in public debts at both the federal and Länder level, and this reached its high point when two Länder claimed a federal bailout asserting that their debt levels were no longer sustainable. The federal government had to bail out Bremen and Saarland following a judgment by the federal constitutional court (Bundesverfassungsgericht, BVerfG).103 The situation with regard to Länder debt levels continued to degenerate,104 and, in 2006, the BVerfG again had to decide whether the federal government is obliged to pay financial assistance or not. Back then, the BVerfG denied Berlin a federal bailout by arguing that a budget crisis only existed when it threatened the entire federal system

100 101 102 103

104

Falsch konstruiert oder falsch verstanden?” in ezff (eds.), Jahrbuch für Föderalismus 2009 (Baden-Baden: Nomos, 2009) 137–148. See D. Heinz, “Coordination in Budget Policy after the Second Federal Reform: Beyond Unity and Diversity”, German Politics, 25 (2016) 286–300. Christian Democratic Union of Germany (cdu) and Christian Social Union in Bavaria (csu). Social Democratic Party of Germany (spd). BVerfGE 86, 148. For details, see S. Korioth, “Rituale im Finanzverfassungsrecht und ihre Folgen—fünf Jahre Stabilitätsrat”, in M. Junkernheinrich et al. (eds.), Jahrbuch für öffentliche Finanzen (Baden-Baden: Nomos, 2015) 299–311, at 301–302. See G. Färber, “Haushaltsnotlagen in der deutschen Finanzverfassung”, föv Discussion Papers, 24 (2006).

252

Alber

as such. While, in the former two cases, the BVerfG ruled in 1992 that they were eligible for federal bailout transfers, it did not rule in favour of this in the case of Berlin in 2006.105 The federal government demanded that Berlin resolves the crisis on its own and required German policymakers to formulate effective borrowing restrictions in order to avoid the need for bailouts in the future. Most importantly, the BVerfG urged the Bund and Länder to come up with a political mechanism for budget issues and, thus, no longer to settle such disputes through the federal constitutional court. Within the most recent German federalism reforms, issues of IGfRs were initially left out, as there was no consensus to include them. The major outcome of Federalism Reform i in 2006 was the improvement of the decision-making capabilities of the Bund and its Länder by redefining the Länders’ participation rights in the Bund legislature (the so-called disentanglement).106 Federalism Reform i did not deal with Germany’s complex equalisation system and made only minor adjustments with regard to IGfRs.107 The revision of IGfRs was postponed to Federalism Reform ii, whose joint parliamentary commission started its work in 2006 and completed it in 2009. However, the joint commission of the Bundestag and the Bundesrat failed to elaborate p ­ roposals on important issues like taxation and equalisation. Both the ­federal and Länder governments agreed to keep the existing intergovernmental transfer scheme (Finanzaus­ gleichsgesetz), which will expire in 2019. The major outcome of Federalism Reform ii was the revision of the constitutional rules that govern subnational and federal borrowing. Against the need to comply with the revised European Growth and Stability Pact, a debt brake (Schuldenbremse) was constitutionally entrenched to secure balanced budgets (Article 109 and 115 of the Basic Law).108 This reshaped the constitutional indebtedness rules for both the Bund and the Länder, reserving deviations on the Land level until 2019. Moreover, Federalism Reform ii institutionalised an early-warning system to prevent and remedy future budget crises, the Stability Council. ­Under the new rules, Berlin, Bremen, Saarland as well as other Länder, such as ­Schleswig-Holstein, will still receive transfers, but they are subject to controls. In December 2016, the f­ederation 105 BVerfGE 116, 327. 106 See S. Burkhart et al., “A More Efficient and Accountable Federalism? An Analysis of the Consequences of Germany’s 2006 Constitutional Reform”, German Politics, 17 (2008) 522–540. 107 See A. Gunlicks, “German Federalism Reform: Part One”, German Law Journal, 8 (2007) 111–132. 108 See Deutsche Bundesbank, “The Debt Brake in Germany—Key Aspects and Implementation”, Monthly Report (October 2011).

Intergovernmental Financial Relations

253

and the 16 Länder after several years of negotiations ultimately reached an agreement to overhaul the equalisation system. It requires the amendment of 13 articles of the Basic Law. On 1 and 2 June, 2017 the two chambers of parliament agreed to the reform package. In a nutshell, horizontal equalisation between the Länder will be abolished and will be replaced by an equalisation system that is valid from 2020 to 2030 and exclusively of vertical nature. To this end, the federal level guarantees the transfer of additional funds to the 16 Länder and receives more rights to intervene (among others, the right to closer scrutiny of spending in the Länder through the Stability Council, and the prerogative to coordinate tax collection). Interestingly, the Länder unanimously agreed to the reform package. They did so because of political convenience. First, at the end of a strenuous negotiation process, the federation was able to satisfy the interests of all Länder by bearing additional costs. Second, the new system will (apparently) ease current tensions between contributors and recipients. Third, under the new regime, the Länder will keep the possibility to voice their discontent with the agreed formulas and, ultimately, also to overhaul them. The impact that the new system will have on the relationship between the federation and the Länder is still to be seen but, in general, the issue has been increasingly discussed in academia under the thesis of “fiscal control paradigm”.109 While it seems to be clear that an exclusively vertical system of equalisation will in no way reduce incentive problems, the extent to which the new regime will render the Länder more dependent on the Bund remains to be seen. At this stage, it is important to recall that once the debt brake enters into force in 2020, the Länder will depend on tax revenue over which they do not have control (the reform package did not deal with tax assignments). 6.2 Spain In Spain, the Senate, the chamber of territorial representation,110 lacks significant policymaking powers, and its relations with the constituent units of Spain, the autonomous communities (acs), are essentially symbolic because of its composition and the fact that in the legislative process the last word

109 See H. Scheller, “Auf dem Weg zum fiskalischen Kontrollföderalismus”, dms—der moderne staat, 7 (2014) 137–156. See also T. Lenk and P. Glinka, “Der neue bundesstaatliche Finanzausgleich: Eine Reform und viel Reformaufschub, Wirtschaftsdienst”, Zeitschrift für Wirtschaftspolitik, 97 (2017) 506–512. 110 The Spanish parliament comprises two chambers: the Congress of Deputies and the Senate, the house of territorial representation. Senators are partly directly elected and partly appointed by the legislative assemblies of the autonomous communities.

254

Alber

b­ elongs to the Congress.111 Intergovernmental relations—predominantly vertical ones—have initially played out rather informally. Against the backdrop of strong regionalist party claims in fiscally robust acs, IGfRs instruments such as bilateral sectoral transfer commissions, sectoral conferences, joint planning bodies and cooperative agreements became essential elements in financial decision-making processes. As a consequence, Spain turned from a system based on centralised tax collection and conditional transfers to a system that increasingly relies on regions’ own-source tax revenues, revenue-sharing and some unconditional equalisation grants,112 in which, however, regional taxes are heavily regulated and inflexible.113 The Spanish Constitution of 1978 contains only a few articles that refer to IGfRs. Articles 156–158 provide for general principles with regard to the financial autonomy of the acs (such as coordination, solidarity, sufficiency),114 to the resources of the acs and to interregional compensation schemes. In particular, Article 157 lists the revenue resources of acs; they consist of (1) taxes wholly or partially assigned to them by the State; surcharges on State taxes and other shares in State revenue; (2) their own taxes, rates and special levies; (3) transfers from an interregional clearing fund and other allocations to be charged to the State budget; (4) revenues accruing from their property and private law income; and (5) the yield from credit operations. Article 158(1) provides for a Basic Public Services Guarantee Fund aiming at ensuring all acs the same resources per unit of need in order to finance the essential public services in the fields of health, education and social services. The overall amount of the fund is calculated annually by taking into account the fluctuation in terms of fiscal capacity. The sum assigned to each ac is determined using the index of the weighted population through several variables: some of them are 111 M.J.G. Morales and X.A. Marín, “Intergovernmental Relations in Spain: An Essential but Underestimated Element of the State of Autonomies”, in Poirier et al., Intergovernmental Relations, supra, 350–378, at 355–356. 112 On the tax system, see F.A. García Prats, “Spain”, in Bizioli and Sacchetto, Tax Aspects of Fiscal Federalism, supra, 361–402. 113 C. Colino and M. Kölling, “Parallel Lives: Unsolved Problems and Reform Initiatives in Spanish and German Fiscal Federalism”, in S. Lütgenau (ed.), Fiscal Federalism and Fiscal Decentralization in Europe: Comparative Case Studies on Spain, Austria, the United Kingdom and Italy (Innsbruck: Studienverlag, 2014), 253–287, at 257. 114 With regard to the principle of financial autonomy, its nature and how it has to be balanced against principles of coordination and solidarity, see Valdesalici, “Financial Constitutions and Responsibility”, supra. See also J. Zornoza, “New Trends in Fiscal Decentralization: A Spanish View”, in Lütgenau, Fiscal Federalism and Fiscal Decentralization, supra, 105–126.

Intergovernmental Financial Relations

255

linked to the population such as an individual’s age, while some others go beyond the spread (dispersion), the surface and the insularity criteria. By doing so, the equalisation system reflects the effective conditions of each ac and the resource distribution matches real expenditure needs.115 Details as to IGfRs are regulated by a special law that is adopted according to a reinforced procedure. The Organic Act on the Financing of the Autonomous Communities (Ley Orgánica de Financiación de las Comunidades Autónomas, lofca), originally approved by Organic Law No. 8/80 and subsequently amended several times, regulates the relationship between the State and the acs that are vested with an ordinary statute. Accordingly, ordinary regional financing rests on three pillars: devolved taxes, shared taxes and equalisation transfers.116 Even though, with the current lofca, the portion of the acs in shared taxes was raised, they still lack relevant regional taxing power. Regional financing differs in acs with a special regime (a foral regime): Navarre and the Basque Country enjoy high fiscal autonomy and, by contrast to acs with an ordinary financing system, they make lower contributions to the interregional solidarity fund.117 They hold large autonomy in tax collection with the exception of customs tariffs and they sent a pre-arranged share of revenue calculated against their relative income and population to the central government (respectively aportación and cupo). The lofca also provides for the creation of specific funds for different purposes as well as for details as to the implementation of the equalisation schemes: the Basic Public Services Guarantee Fund, a horizontal mechanism that receives resources from the acs and then distributes them among them (this fund represents the lion’s share of total resources available in the system); the Global Sufficiency Fund and the two convergence funds, the Competitiveness Fund and the Cooperation Fund, are funds with a vertical nature and they channel economic transfers from the State to the acs. The equalisation scheme was changed from a system of total equalisation based on a static assessment of relative needs to a system that provides for partial equalisation, with frequent adjustments. It has, however, to be pointed out that the amount of funds that is appropriated to each ac cannot in any case be lower than the one appropriated under the preceding financial regime.

115 E. Alber and A. Valdesalici, “Reforming Fiscal Federalism in Europe: Where Does the Pendulum Swing?”, L’Europe on Formation, 1 (2012) 325–366, at 341. 116 A. Herrero Alcalde, “The Spanish Way of Fiscal Federalism”, in Lütgenau, Fiscal Federalism and Fiscal Decentralization, supra, 15–39, at 29. 117 Ibid., at 27.

256

Alber

Details and reforms of the lofca are de facto elaborated by an intergovernmental body composed of ministers of the State and the acs, the Fiscal and Financial Policy Council (Consejo de Política Fiscal y Financiera, cpff). The cpff was the first sectoral consultative conference created in Spain and it is responsible for the coordination of IGfRs. Its members are the Ministries of Finance and Public Administration of the central government and the finance counsellors of the acs.118 Some scholars have argued that regional alliances in the cpff tend to form more around regional income levels than around political-party affiliation.119 However, this evaluation has been contested as according to other scholars the party logic weighs heavily on the cpff members. As a consequence, voting in the cpff would occur according to orders received from parties.120 If one looks at how decisions are taken in the cpff, then the central government ultimately controls the cpff because agreements are decided by a simple majority and the central government does hold 50% of the votes. Therefore, the central government, in order to carry a vote, only needs the representative of one of the acs to vote in favour or abstain. H ­ owever, in practice, any reform as to the funding of the acs needs a certain degree of political commitment in each acs in order to play out successfully. Overall, the functioning of the cpff has been increasingly criticised for its lack of transparency and the inadequate way in which it communicates about its work. 6.3 United Kingdom of Great Britain and Northern Ireland (uk)121 In the early years of devolution in the uk from 1999 up to 2007, decentralisation to three of its four constituent nations (Scotland, Northern Ireland and Wales) was undertaken in a major way in the fields of legislation and administration, and less so in the field of fiscal powers.122 Today, taxes are still collected to a very large extent by the centre, and revenue is shared by means of an unconditional block grant system, which, in the absence of a written constitution, is based on a complex web of instruments of soft and statutory law. The block grant system consists of money allocated on an annual basis by the uk government, 118 See J. Calvo Vérgez, “El Consejo de Política Fiscal y Financiera en el Nuevo Modelo de Financiación Autonómica”, Crónica tributaria, 139 (2011) 7–44. 119 See J. López Laborda et al., “Kingdom of Spain”, in A. Shah (ed.), The Practice of Fiscal Federalism, supra, 287–316. 120 Morales and Marín, “Intergovernmental Relations in Spain”, supra, 350–378, at 369. 121 This part was finalised before uk Prime Minister Theresa May has triggered the formal two-year process of “Brexit” negotiations by invoking Article 50 of the Lisbon Treaty on 29 March 2017. 122 J. Frosini, “La devolution fiscale nel Regno Unito: verso l’apertura del vaso di Pandora?” in Palermo and Nicolini, Federalismo fiscale in Europa, supra, 109–123, at 109ff.

Intergovernmental Financial Relations

257

the amount of which is determined by a technical formula known as the Barnett formula. The Barnett formula was first introduced in 1978 as a short-term solution; it has no statutory basis and, being merely based on a convention, it could be changed at any time at the will of the uk Treasury.123 The formula applies to all expenditure issues the devolved governments are responsible for, and changes to its amount are calculated by taking into account the changes made to spending in England on those services that, in the case of the devolved governments, are decentralised.124 Put differently, any increase or reduction in expenditures in England leads to a differentiated but proportionate increase or reduction in expenditures in Scotland, Northern Ireland and Wales, calculated on the basis of all the devolved competencies. Thus, the formula links the financial allocations to each constituent unit to decisions the uk parliament makes on public spending in England. It is worth noting that no account is made in the calculations to the revenue raised in each constituent units or to a set of needs-based indicators. Beyond population weighting, calculations only relate to the spending needs in England. This bilateral distributive mechanism has been widely criticised because Scotland seems to be overfunded in relation to its needs while Wales and England are not. Reform debates have revolved around the fact that the current block grant system should be turned from a distributive into a redistributive mechanism that is based on relative needs. This because the adoption of a distributive approach made both inequalities in the uk’s constituent units and inequities in how public money is distributed across the uk ever more visible, and it is questioned how long a distributive approach based on bilateral negotiations will be. While the devolved governments have very little or no responsibility on the revenue side, they enjoy broad freedoms on the expenditure side, with the Scottish government being responsible for 70 per cent, the Welsh government for 56 per cent and the Northern Ireland executive for 53 per cent of public spending.125 Scotland, Northern Ireland and Wales deal with IGfRs in different ways (e.g. claims for tax devolution), and the uk government’s approach in designing IGfRs is to react on a bilateral basis when forced to do so, supported by the help of ad hoc commissions. 123 Ibid., at 119. See also P. Leyland, The Constitution of the United Kingdom: A Contextual Analysis (Portland: Hart Publishing, 2012), especially 246–266. 124 E. Gwilym, “United Kingdom”, in Lütgenau, Fiscal Federalism and Fiscal Decentralization, supra, 61–71, at 65. 125 A. Trench, “The uk’s Devolution Finance Debates”, in Lütgenau, Fiscal Federalism and Fiscal Decentralization, supra, 321–339, at 321.

258

Alber

An example is the Commission of Scottish Devolution, the Calman Commission (2007–2009). It consists of 15 members that include nominees of the three Unionist parties and representatives of business, trade unions, academia and organised civil society. According to its recommendations, the Scottish Parliament should get substantially greater control over raising revenue primarily through sharing responsibility with the uk government over the income tax rate, but also through the devolution of taxes (e.g. landfill tax, stamp duty, land tax, air passenger duty). Moreover, the Calman Commission recommended that the block grant system should continue to be based on the Barnett formula but it should be replaced as soon as possible by a UK-wide mechanism based on needs assessments. In line with these recommendations, the Scotland Act 2012 assigned the Scottish Parliament certain fiscal powers, but this did not put a stop to debates over fiscal autonomy. The Smith Commission, established in the wake of the no vote on Scottish independence on 18 September 2014 and consisting of 10 members nominated by political parties, recommended that new powers over some taxes and welfare payments should be devolved. Its recommendations were the basis for almost a full year of negotiations between the Scottish and the uk governments. The results of these negotiations led to the Scotland Act 2016, which grants Scotland further fiscal powers and “upgrades” the Scottish Parliament and government as permanent institutions within the uk’s set-up.126 In 2007, the Welsh government established the Holtham Commission, consisting of three economics professors. In 2010, the Commission concluded that the Barnett formula had resulted in unsatisfactory funding for Wales. It proposed a new funding scheme calculated on the basis of relative needs and advanced suggestions as to the decentralisation of borrowing powers and certain taxes. In 2011, the uk government mandated the Silk Commission127 (seven members including representatives of political parties from the Welsh Assembly) to review possibilities for fiscal decentralisation and to consider whether 126 See at hm Treasury et al., “The Agreement between the Scottish Government and the United Kingdom Government on the Scottish Government’s Fiscal Framework”, (2016), available at www.gov.uk/government/publications/the-agreement-between-the-­scottish -government-and-the-united-kingdom-government-on-the-scottish-governments -fiscal-framework (accessed 3 August 2016). For assessments, see Institute for Fiscal Studies, “Scotland, Wales and Northern Ireland”, available at www.ifs.org.uk/research _areas/198/201?year_published[start]=&year_published[end]=&page=1& (accessed 3 August 2016). 127 Trench, “The uk’s Devolution”, supra, at 328–333. See also National Assembly for Wales, “The Silk Commission and Wales Act 2014”, available at www.assembly.wales/en/bus -home/research/Pages/research-silk-commission.aspx (accessed 3 August 2016).

Intergovernmental Financial Relations

259

further non-fiscal powers should be devolved to Wales. Some of the Silk Commission’s recommendations were implemented by the Wales Act 2014. In concrete terms, devolution of tax-raising powers is to begin with smaller taxes such as stamp duty in 2018, and, by 2020, the Welsh government will also have the power to change its income tax following a referendum. Moreover, the Tax Collection and Management Act 2016 created the Welsh Revenue Authority. Northern Ireland has, until recently, largely opted out of major discussions on fiscal devolution, because since the Good Friday Agreement of 1998 it has enjoyed generous treatment, also within the block grant system. For example, borrowing powers were devolved in 2002 and generous access to the uk Reserve was granted in 2010 when services in justice and policing were devolved.128 From 2010 onwards, the most discussed issue started to be the corporation tax. The devolution of the corporation tax was argued as being necessary to give a substantial boost to the growth of the private sector in Northern Ireland.129 In 2014, the Stormont House Agreement put in place the conditions for the government of Northern Ireland to exert greater influence over the structure of its economy by means of levying a corporation tax. The Stormont House Agreement also binds the Northern Ireland executive to ensure a balanced budget and to implement a comprehensive reform of its weak public sector.130 Based on the agreement’s implementation plan,131 the Northern Ireland Assembly should be enabled to change the corporation tax rate as of April 2017. As illustrated, any discussion about IGfRs in the uk has evolved in a ­disjointed manner. It is driven, first of all, by political negotiations and s­ upported by ad hoc commissions. Since 2012, the Office for Budget Responsibility (discussed further in Part 7.3) has forecast some tax streams that have been devolved to the Scottish Parliament and, more recently, also with regard to the other devolved constituent units.132

128 Trench, “The uk’s Devolution”, supra, 333–335. 129 For a review of the fiscal powers of the Northern Ireland Assembly, see nicva, “Review of the Fiscal Powers of the Northern Ireland Assembly”, (2013) available at www.nicva.org/ resource/review-fiscal-powers-northern-ireland-assembly (accessed 3 August 2016). 130 See oecd, “Northern Ireland (United Kingdom): Implementing Joined-up Governance for a Purpose”, oecd Public Governance Reviews (Paris: oecd Publishing, 2016). 131 Northern Ireland Office and T. Villiers, “A Fresh Start for Northern Ireland”, (2015), available at www.gov.uk/government/news/a-fresh-start-for-northern-ireland (accessed 3  ­August 2016). 132 For forecasts, see Office for Budget Responsibility, “Norther Ireland Tax Forecast— Legilsation”, (2016), available at http://budgetresponsibility.org.uk/topics/scotland-wales -and-northern-ireland/northern-ireland-tax-forecasts/ (accessed 3 August 2016).

260 7

Alber

Strengthening (Post-Crisis) Fiscal Credibility: (New) Fiscal Watchdogs

Against the backdrop of the persistence of deficits and the unprecedented indebtedness of the public sector in the aftermath of the global financial crisis of 2008–2009, the number of Independent Fiscal Institutions (ifis) housed in the legislature or outside the executive as stand-alone expert committees is increasing.133 There are a few countries where “fiscal watchdogs” had already been established earlier,134 but it is especially over the last two decades that governments have allowed for reviews of budgets and budgetary targets by ifis as a response to both their loss of credibility with regard to budget forecasts and cost assessments, and to supranational pressures (see the case of eu Member States discussed in Part 7.3). Academically, the idea of establishing ifis gained fresh momentum among economists towards the end of the 1990s, when fiscal decentralisation was at stake. It is seen as an effective way to address bias towards spending and deficits, and, more generally, to enhance fiscal discipline by raising the quality of policymaking and scrutiny processes.135 Today, the database of the International Monetary Fund (imf) tracks ifis in 39 states.136 Of these, 26 are classified as free of any political interference, almost half of them have a mandate to produce economic forecasts, while the others have a mandate to analyse forecasts by the institutions responsible for budget submission. In accordance with the imf’s criteria and the recommendations of the Organisation for Economic C ­ o-operation and Development (oecd),137 an ifi—which comes u ­ nder 133 See G. Kopits, “Independent Fiscal Institutions: Developing Good Practices”, oecd Journal on Budgeting, 11 (2011) 1–18. See also X. Debrun and T. Kinda, “Strengthening PostCrisis Fiscal Credibility: Fiscal Councils on the Rise—A New Dataset”, imf Working Paper, WP/14/58 (2014). 134 For example, the us Congressional Budget Office has been in operation since 1974 (discussed further in Part 7.1), the Belgium High Council of Finance since 1936 and the Netherlands Bureau of Economic Policy Analysis since 1945 (the latter two were without a clear remit in fiscal policy until respectively 1989 and 1986). 135 See X. Debrun et al., “The Functions and Impact of Fiscal Councils”, imf Policy Paper (July 16 2013). See also J.A. Frankel and J. Schreger, “Over-optimistic Official Forecasts in the Eurozone and Fiscal Rules”, Review of World Economics, 149 (2013) 247–272. 136 Source: imf, “Fiscal Council Dataset”, available at: www.imf.org/external/np/fad/council/ (data refers to the end of December 2014). If not stated otherwise, the paragraph drawn on the information available at the imf web page (accessed 2 August 2016). 137 oecd, “Recommendation of the Council on Principles for Independent Fiscal Institutions”, (2014). See also the oecd network at: www.oecd.org/gov/budgeting/­oecdnetwork ofparliamentarybudgetofficialspbo.htm (accessed 3 August 2016).

Intergovernmental Financial Relations

261

d­ ifferent names such as parliamentary budget office, fiscal council, office for budget responsibility or stability council—is a permanent agency with no decision-making powers that provides non-partisan analysis of public-sector finances and performs at least one of the following functions independently: • assessing the government’s fiscal policies, plans and performance against macroeconomic objectives related to the long-term sustainability of public finances, as well as short- to medium-term macroeconomic stability or other objectives; • elaborating unbiased macroeconomic and budgetary forecasts in the different phases of budget cycles; • compiling sensible fiscal policy analysis and possibly also policy recommendations; • facilitating the implementation of fiscal rules and objectives of economic governance set forth at national and supranational level. The organisational structure and mandate of ifis widely vary from State to State. In order to properly perform their functions in parliamentary systems, ifis should be independent from both the executive and the legislature, while in presidential systems independence from the executive is sufficient. ifis are headed either by an individual or by a council that is either appointed by the government or elected by the legislature (or a legislative committee) for a period spanning beyond the term of the government. This is to guarantee independence and impartiality. Moreover, ifis must have their own staff and get full access to information from the respective public offices in order to properly fulfil their mandate. Unlike public audit institutions, ifis are not “backwardlooking”, but they rather perform a “forward-looking diagnostic task”.138 Their work is intended to contribute to transparency in policymaking processes and to increase people’s trust in government actions. As independent institutions, they may limit the “scope for politicization of fiscal decisions”139 and, ultimately, enhance the credibility of partisan budgetary decision-making.140 7.1 Fiscal Watchdogs: Evidence from North America, Asia and Africa In the us, conflicts over how to manage finances used to be very common because of the country’s political system, whereby the executive branch, the 138 Kopits, “Independent Fiscal Institutions”, supra, at 2. 139 X. Debrun and M.S. Kumar, “The Discipline Enhancing Role of Fiscal Institutions: Theory and Empirical Evidence”, imf Working Paper WP/07/171 (2007), at 4. 140 See also B. Anderson, “The Changing Role of Parliament in the Budget Process”, oecd Journal on Budgeting, 1 (2009) 1–11.

262

Alber

­president, is ascribed overall responsibility for budget planning, and the legislative branch lacks institutional capacity to enforce budgetary priorities.141 Conflicts between the executive and the legislative branch came to a head in 1973, when the administration of Richard Nixon impounded congressional appropriations for some programmes. The resulting dispute led to the enactment of the Congressional Budget and Impoundment Control Act of 1974, which (1) prevented the executive from refusing to spend budget appropriations approved by the Congress, (2) limited delays of approved expenditures to the next fiscal year, (3) mandated the creation of the Congressional Budget Office (cbo), which has the task to provide the Budget Committees and Congress with non-partisan information about budgetary and economic issues (Section 202 of the Congressional Budget and Impoundment Control Act 1974). The cbo produces an annual Budget and Economic Outlook and is required to produce a formal cost estimate for nearly every bill (the only exceptions are appropriations bills, which do not receive formal written estimates, but their budgetary effects are estimated by the cbo for the Appropriations Committees).142 Today, the cbo staff consists of approximately 250 analysts and is headed by a director who is appointed jointly by the president of the Senate and the speaker of the House of Representatives for a renewable four-year term. The cbo is not tasked with the formulation of policy recommendations, but it produces a wide range of analyses from forecasts of the budgetary impact of policies to estimates of the cost of bills and their effect on macroeconomic variables. This enables the Congress to compare the government’s budget plan with the cbo’s estimates. Interestingly, in the us, a number of states created cbos before the cbo was established at the federal level. California’s Legislative Analyst Office was, for example, created in 1941. Today, almost all states have an independent office, but the remits of ifis at the state level differ widely from state to state. The Canadian Parliamentary Budget Officer (pbo) was established in 2006 by an amendment of the Parliament of Canada Act 1985 (Section 79.1). It was created as a response to serious criticism surrounding the accuracy and credibility of the federal government’s fiscal projections and forecasting process.143 141 See J.R. Blöndal et al., “Budgeting in the United States”, oecd Journal of Budgeting, 3 (2003) 7–53. 142 For details, see the website of the Congressional Budget Office: www.cbo.gov/ (accessed 3 August 2016). The paragraph draws on the information available on the website. 143 See G.A. Beaumier, “The Accountability Act and the Parliamentary Budget Officer” (Ottawa: Library of Parliament, 2006). See also G. Levy, “A Parliamentary Budget Officer for Canada”, Canadian Parliamentary Review, (2008) 39–44; and B. Jeffrey, “The Parliamentary Budget Officer Two Years Later: A Progress Report”, Canadian Parliamentary Review, (2010) 37–45.

Intergovernmental Financial Relations

263

In sum, the pbo is tasked with the elaboration of independent analysis and assessments on the state of the nation’s finances, on the federal government’s estimates, on trends in the national economy and, upon request, on the financial cost of any proposal in matters over which the federal parliament has jurisdiction. The Governor in Council may select the pbo from a list of three names submitted in confidence, through the leader of the government in the House of Commons, by a committee formed and chaired by the parliamentary librarian (one of the key officials in the parliament of Canada). The pbo is an officer of the Library of Parliament, appointed for a renewable five-year term and supported by almost 20 people, who are employees of the Library. Unlike the director of the cbo, the pbo’s position is weaker because its mandate can be revoked at any time. In carrying out his/her work, the pbo has encountered several problems. In 2012, these problems reached a high point when several government departments denied the pbo access to all relevant information in order to be able to analyse the impact of the 2012 budget. The pbo sought a reference from the federal court of Canada in order to clarify its mandate. The federal court of Canada dismissed the pbo’s reference on a technicality (basically, because the departments did not yet decline the provision of necessary information). The federal court of Canada, however, held that if the federal parliament wished to limit the pbo’s work, then it would have to abolish or reframe the pbo’s mandate. The judge’s obiter dictum upheld both the right of the pbo to seek the necessary information and the right to take the government to court for the refusal to provide the requested information.144 Thus, the independence of the pbo and its right to resort to judicial review where departments unlawfully refuse to disclose information was affirmed.145 Currently, the Trudeau government intends to provide the pbo with adequate funding and even to expand its mandate by including the formulation of estimates of party platforms in election campaigns. Unlike in the us, where almost every state has established its own independent budget office, in Canada only Ontario has established such an institution in the form of its Financial Accountability Office. Many other Canadian provinces tried to create such a body but have not yet been successful (for example, British Columbia, Saskatchewan, Alberta, New Brunswick). In 2012, the Australian Parliamentary Budget Office (Australian pbo) was created as one of four parliamentary departments supporting the Australian Parliament. Its mandate is to provide for independent analysis of the budget 144 L. von Trapp et al., “Canada”, oecd Journal on Budgeting, 2 (2015) 77–91, at 85. 145 See T. Yalkin and B. Segel-Brown, “Demystifying the Mandate of the Parliamentary Budget Officer”, Office of the pbo, paper prepared for a parliamentary seminar on 29 May 2014.

264

Alber

cycle, fiscal policy and the financial implications of proposals (Section 64B of the Parliamentary Services Act 1999). “Enhancing the credibility and transparency of Australia’s already strong fiscal and budget frameworks” was the motivation for the establishment of the Australian pbo, which is understood as a “significant institutional innovation”.146 According to Section 64E(1) of the Parliamentary Services Act 1999, the mandate of the Australian pbo comprises: (1) the preparation of policy costing if requested by members of parliament or other authorised members; (2) the response to requests by senators or members of the House of Representatives that are related to the budget; (3) the preparation of submissions requested by parliamentary committee inquiries; and (4) own research and analysis regarding the budget and fiscal policy settings. The Australian pbo’s mandate does not include the elaboration of any economic forecasts or budget estimates at any level and, in this respect, it significantly differs from the uk Office for Budget Responsibility, which provides forecasts for the economy and makes projections about both short- and long-term fiscal targets that the central government has set rather than costing ­policy for parliament. The Australian pbo is chaired by the parliamentary budget officer, who is appointed for a renewable term of four years. The approval of the Joint Committee for Public Accounts and Audit (jcpaa) must be obtained before the officer can be appointed. In contrast to the Canadian pbo officer, who can be dismissed by the prime minister at will, the Australian officer can only be removed on grounds of misbehaviour, incapacity or if insolvent under administration. As also in the case of pbos in other countries, a memorandum of understanding with all relevant departments and agencies is made to ensure that the Australian pbo will have access to all information needed. Moreover, the terms of reference both as to confidentiality and the release of information to the public and the interaction of the officer with the Department of Treasury and Finance and the jcpaa are outlined in the Parliamentary Service Act 1999. According to an audit in 2014, the Australian pbo has “effectively undertaken its statutory role” and has made a “significant contribution to levelling the playing field for all parliamentarians”.147 Likewise, the jcpaa defines the pbo as a “permanent and indispensable part of [the] architecture 146 M. Stewart and H. Jager, “The Australian Parliamentary Budget Office: A Sustainable Innovation in Fiscal Decision-making?” Melbourne Legal Studies Research Paper, (2013), at 2. As to the motivations for establishing the Australian pbo, see at 18–19. 147 Australian National Audit Office, “The Administration of the Parliamentary Budget Office”, Commonwealth of Australia report, no. 36 (2014), at 18. The pbo welcomed the recommendations of the audit to include in all costings estimates of administrative expenses where significant (at 30).

Intergovernmental Financial Relations

265

of good government, contributing to transparency and accountability in fiscal policy, policy costings and the budget”.148 In South Africa, the establishment of a parliamentary budget officer was accompanied by debates on whether such an institution would be the only appropriate means to enhance fiscal credibility.149 The Money Bill Amendment Procedure and Related Matters Act No. 9/2009 provides for the establishment of a Parliamentary Budget Office. Its Section 15(1) specifies that such an office headed by a director has to “provide independent, objective and professional advice and analysis to Parliament on matters related to the budget and other money bills”. In sum, it has to support the committees on Finance and Appropriations in the National Assembly and National Council of Provinces by annually providing reviews and analysis of the documentation tabled in parliament by the executive in relation to the Money Bill Act, give advice on proposed amendments of the fiscal arrangements and on policy proposals and their budgetary implications, monitor and report on potential unfunded mandates, and give advice on the correlation between South Africa’s budgetary allocation and its policy priorities. Its first director was appointed in June 2013. In India, fiscal forecasts are prepared by the Ministry of Finance and scrutinised by the parliamentary Standing Committee on Finance. There is presently no independent stand-alone institution that regularly reviews forecasts and does research on fiscal issues. The comptroller and auditor-general, an independent body authorised by the Indian Constitution (Article 148), evaluates and reports ex post on the government’s fiscal performance, but it does not pronounce itself ex ante on the government’s fiscal policy.150 7.2 Fiscal Watchdogs in eu Member States: Preliminary Remark In the eu, the Treaty on Stability, Coordination and Governance in the ­Economic and Monetary Union, which regulates the so-called “Fiscal C ­ ompact”, recommends that independent bodies at the national level monitor compliance with national and supranational fiscal policy rules in each Member State of the Eurozone [Article 3(2)]. From 11 September 2015 onwards, ifis within the eu are pooled together in a network called “eu Independent Fiscal Institutions”.

148 jcpaa Report 446, “Review of the Operations of Parliamentary Budget Office”, Commonwealth of Australia (2014) at 46. 149 See E. Callitz et al., “Enhancing the Credibility of Fiscal Forecasts in South Africa: Is a Fiscal Council the Only Way?” Stellenbosch Economic Working Papers 25 (2013). 150 S. Gupta, “Budget Institutions in G-20 Countries—Country Evaluations”, imf (2014), at 43–44.

266

Alber

The network is a voluntary and inclusive institution open to all independent fiscal oversight bodies operating in the eu.151 7.3 Fiscal Watchdogs: Evidence from eu Member States In Germany, in order to prevent a budget crisis, the Bund and the Länder in Federalism Reform ii agreed on the creation of a consultative mechanism that monitors all budgets. The German Financial Council152 was “upgraded” to what is now called the Stability Council (Stabilitätsrat). It monitors compliance with the new borrowing rules and provides for procedures aimed at containing the indebtedness of all government tiers (restructuring programmes).153 The Stability Council (sc), mandated by Article 109a(3) of the Basic Law and regulated by the Stability Council Act 2009, is composed of the federal finance minister, the finance ministers of all Länder and the federal economy minister; it meets twice a year and is alternatively chaired by the federal minister of finance and the chairperson of the financial ministers’ conference. Its mandate is to monitor fiscal policies and to evaluate the Länder budgets according to financial indicators, which are defined by taking the Länder average as a benchmark (structural financial balance, credit financing ratio, debt level per capita, and ratio of interest expense to tax revenue).154 The sc publishes its decisions and all corresponding materials on its website.155 Based on the calculated position of the regular economic situation, the sc decides whether the Bund or a Land is facing a recession or is experiencing an economic boom, and, in a second step, whether new debts can be taken on or not. The evaluation procedure by the sc does not trigger any financial assistance from the federation or the other Länder; aid is left to the current equalisation system (excluded from the scope of Federalism Reform ii and expiring in 2019). The starting point for budget reviews are the stability reports that both the Bund and the Länder have to submit annually. They include a report of the current and short-term budget situation and a medium-term budget projection. The latter is limited by its qualitative political nature and by the different financial planning methods in use in the Länder. The sc restructuring programmes are implemented 151 See at www.euifis.eu (accessed 3 August 2016). 152 Prior to the establishment of the Stability Council, there was no close cooperation among the Länder themselves or between the Länder and the Bund. The Financial Planning Council provided only for loose agreements and not for any kind of preventive instrument. 153 See Research Briefing, Deutsche Bank Research (ed.), “Stability Council: Financial Inspector of Germany’s Länder”, (2011), available at www.dbresearch.com (accessed 3 August 2016). 154 Heinz, “Coordination in Budget Policy”, supra, at 293–296. 155 See www.stabilitaetsrat.de for more details (accessed 3 August 2016).

Intergovernmental Financial Relations

267

under each Land’s own directive, and the Land is obliged to regularly report to the sc. In theory, consolidation measures could equally occur on the expenditure side and on the income side. In practice, the Stability Council Act stresses that measures are only suitable if they are solely in the jurisdiction of the affected level of government; as a consequence, the restructuring measures focus on the expenditure side owing to the limited revenue autonomy of the Länder. Overall, the sc is considered to be a watchdog that exerts public pressure by making budgetary and fiscal policy transparent while activating soft coordination mechanisms of “naming and shaming” that lead to an implicit performance comparison among Länder based on an average benchmarking of efficiency (a sort of horizontal competitive coordination). As specified in the Stability Council Act, “The resolutions of the Stability Council must obtain the vote of the Federation and a majority of two-thirds of the Länder. The vote of the Federation shall be cast by the Federal Minister of Finance”. If a decision with regard to a single Land is subject to a vote, then the respective Land is not allowed to participate in the vote. When decisions affect the federal level, decisions are taken by a majority of two thirds of all voting members. The fact that the sc does not have any scope to impose ultimate direct sanctions if any restructuring programme fails (e.g. the denial of voting rights in the Bundesrat) is viewed as its biggest weakness next to the fact that the sc, for the time being, does not follow up on budget offenders in a strict way.156 The sc’s “mild responses” to budget offenders might as well have implications for future jurisprudence of the constitutional court in the areas of federal bailouts, provided that a Land will claim them after 2019. Proposals on how to strengthen the position of the sc are controversially discussed, especially against the backdrop of the Länder autonomy in budget policy but also with regard to the question whether any reform of the sc would be democratically legitimised or not.157 In Spain, Organic Law No. 6/2013 provides for the creation of the Independent Authority for Fiscal Responsibility (Autoridad Independiente de Responsabilidad Fiscal, AIReF).158 Its mandate is to ensure strict compliance with the principles of budgetary stability and financial sustainability enshrined in 156 Korioth, “Rituale im Finanzverfassungsrecht und ihre Folgen”, supra, at 301–302. 157 See S. Thomasius, “Der Stabilitätsrat: Ein fiskalpolitisches Gremium zwischen Kontinuität und Neuanfang”, in K. Schmidt et al. (eds.), Staatsverschuldung in Deutschland nach der Föderalismusreform ii—eine Zwischenbilanz (Hamburg: Bucerius Law School, 2012) 189–222. See also M. Thye, “Die demokratische Legitimation eines unabhängigen Stabilitätsrats aus Sachverständigen: Eine Untersuchung mithilfe einer politikfeldbezogenen Demokratietheorie”, in Schmidt et al., Staatsverschuldung in Deutschland, supra, 223–244. 158 See www.airef.es/ (accessed 3 August 2016).

268

Alber

­Article 135 of the Spanish Constitution through ongoing monitoring of the budget cycle, public indebtness and the analysis of economic forecasts. Similarly to other ifis in the eu, the creation of AIReF was the result of two interlinked processes: the external eu reform process leading to the new framework of the eu’s economic and fiscal policy governance, and the internal reform process leading to the amendment of Article 135 of the Spanish Constitution in 2011. The main purpose of the new Article 135 is to assure that public administration complies with the principle of budgetary stability, and, thus, to forbid them to incur a deficit position above the thresholds defined in the eu’s Stability and Growth Pact. AIReF is attached to the Ministry of Finance and Public Administrations. However, none of its board members or staff may request instructions from any public or private entity and, like other ifis, AIReF in its work is bound by the principles of transparency and accountability. AIReF’s advisory board was created in March 2015 and it is formed by a total of nine members who are appointed by the president of AIReF for a three-year term. The president of AIReF is designated for a non-renewable six-year term of office prior approval by the committees for public finance and public administrations, committees of the Congress and the Senate. Most importantly, the president cannot be dismissed without a hearing in front of those committees. The president of AIReF has to report at least once a year to the committees in parliament, but also more often shall it be requested. The authority’s role in Spain’s system of IGfRs is reinforced by the principle of “comply or explain”. This means that any institution receiving an AIReF report that is considered mandatory (i.e. in the case of macroeconomic forecasts, draft budget, stability programme) has to provide an explanation if it disregards the AIReF recommendations. However, as this principle does not come along with any strict time limits or sanctions, it has, for the time being, played out unsatisfactorily.159 Noteworthy as regards the AIReF is also the fact that it is subject to external evaluation not only with regard to its institutional reports but also in relation to its own governance and operational structure, and even its mandate. Moreover, although AIReF fully operates only for a few years, it is considered to be an institution that, compared to many of its equally young peers, has already issued a significant amount of reports and whose functioning is a model of transparency.160 In the uk, a non-political body, the Office for Budget Responsibility (obr),161 was set up by the Budget Responsibility and National Audit Act in April 2011 159 L. von Trapp et al., “Spain”, oecd Journal on Budgeting, 2 (2015) 213–229, at 218. 160 Ibid, at 227. 161 See www.budgetresponsibility.independent.gov.uk (accessed 3 August 2016). This part was finalised before uk Prime Minister Theresa May has triggered the formal two-year

Intergovernmental Financial Relations

269

with a mandate to monitor fiscal policies and examine the sustainability of public finances. Beforehand, this function was vested in the financial services authority, accountable to the treasury ministers and, through them, to parliament. Accountable to the chancellor of the exchequer and to parliament (Treasury Selection Committee, tsc),162 the obr has four tasks: (1) it is responsible for formulating the official five-year forecasts for the economy and public finances; (2) on the basis of these five-year forecasts, it assesses whether the government has a better than 50 per cent chance of achieving the fiscal targets that it has set itself in different policy fields; (3) it scrutinises the fiscal cost or savings that would result from each tax and welfare spending measure set forth in the budget statement; and (4) it provides both for analyses of the health of the public-sector balance sheet and for analyses with regard to the long-term sustainability of public finances (50-year projections of revenues, spending and financial transactions that also take into account changes in demographics).163 Interestingly, the obr has also been increasingly producing forecasts related to fiscal and financial issues in relation to tax devolution in Scotland, Northern Ireland and Wales. By international standards, the obr is considered to be the most extreme case of positive analysis. It restricts itself to the analysis of the current policies of the current government and does not provide analysis with regard to the impact of different policy options,164 whereby specifying current government policies is not always as straightforward as it sounds, particularly when making projections over five decades rather than five years. By contrast, the South African pbo, for example, also assesses economic and fiscal consequences of political parties’ manifestoes ahead of general elections. Moreover, from a comparative perspective, the obr has a relatively narrow remit focusing very much on fiscal rather than broader policy analysis (see the case of the German sc). The obr publishes its macroeconomic forecasts without drawing process of “Brexit” negotiations by invoking Article 50 of the Lisbon Treaty on 29 March 2017. 162 The tsc also plays a role in appointing the members of the obr. It holds confirmation hearings and has veto powers after a candidate has passed the formal application and interview process organised by the civil service, and has been nominated by the chancellor of the exchequer. Put simply, the parliament has a veto over appointments even though this safeguard for independence might be easily questioned because the majority of the tsc Committee come from government parties. 163 R. Chote, “Britain’s Fiscal Watchdog: a View from the Kennel”, Institute and Faculty of Actuaries Spring Lecture, (2013), at 6–7, available at http://budgetresponsibility.org.uk/ britains-fiscal-watchdog-view-kennel/, (accessed 3 August 2016). 164 See S. Wren-Lewis, “Fiscal Councils: The uk Office for Budget Responsibility”, CESinfo dice Report No. 3 (2011).

270

Alber

any wider macroeconomic policy conclusions and without comparing them to government forecasts. This is due to two reasons: first, it is up to the treasury to provide macroeconomic advice to the chancellor of the exchequer, and, second, the government no longer publishes economic and fiscal forecasts itself. This implies that the obr has to work closely with ministers and officials. With regard to the operational structure and financial resources, the obr works under the aegis of the treasury. Therefore, it could be easily subject to political and financial pressures. But the fact that the obr has been given a multi-year funding profile makes the exertion of pressure more difficult. An external review of the obr has found that the obr has not, up until now, encountered major problems and has succeeded in its forecasts.165 8

Comparative Evaluations and Concluding Remarks

The main goal of this paper has been to give evidence on international experiences in the field of IGfRs from an institutional account. Eight multilevel States served as case studies—three of the world’s oldest federal States, the us, Canada and Australia; one of Europe’s oldest federal State, Germany; a European quasi-federal State, Spain; the European devolved system of the uk; and two States that formally are not called federal but do work as federal States in order to dynamically accommodate their socio-economic, linguistic and religious diversity, India and South Africa. The paper explored relevant country-specific factors that influenced the creation and current functioning of institutions dealing with IGfRs, and it provided answers to a set of questions that are crucial to IGfRs. To this end, it, initially, offered general remarks as to the design and relevance of IGfRs by pointing out that the primary task of any design of IGfRs is to allow a multilevel State to effectively manage mechanisms aimed at ensuring the delivery of service throughout its whole territory and across governmental levels, and, thus, to counteract “disparities of power ingredients” and to best handle asymmetries while respecting the autonomy of its constituent units. Answers as to how details of IGfRs (i.e. the allocation of fiscal and financial powers as well as the distribution of revenues and equalisation schemes) look like and play out in practice, widely vary from State to State. While in unitary States IGfRs are, first of all, an administrative issue (“National Government Model”), in federal, quasi-federal or devolved States the design of IGfRs is political (“Intergovernmental Forum Model”). Actors and mechanisms 165 K. Page, “External Review of the Office for Budget Responsibility”, (2014), at 12, available at www.gov.uk/government/publications (accessed 3 August 2016).

Intergovernmental Financial Relations

271

deciding on the questions to what extent and under which conditions revenues are distributed from the central government to the lower ties of government are the linchpin of federal-state relations, and of state-state relations. Put differently, at the heart of IGfRs there are institutional mechanisms through which different actors enter into relations with each other in order to share information, coordinate policies and distribute wealth, both vertically and horizontally. The paper has given evidence of the functioning of such mechanisms by putting them into the respective socio-political contexts. Ideally, institutions dealing with IGfRs should not only offer a platform for negotiating financial matters at stake, but they should be the primary forum where the right sorts of incentives are developed in order to best meet a State’s overall macroeconomic governance objective: sustainable economic growth. With regard to the analysed case studies, the paper highlighted if and to what extent principles, institutions and procedures of IGfRs are explicitly enshrined in the constitution of eight States. For example, the constitution of the us is a prime example of a constitution that is silent when it comes to revenue-sharing schemes and intergovernmental transfers. Moreover, the us constitution only includes a few articles in the fields of taxation, spending and borrowing powers. On the contrary, the German Basic Law and the Indian constitution are examples at the other end of the spectrum: they provide a comprehensive body of principles and rules devoted to IGfRs. The other case studies occupy intermediate positions, in which the design of IGfRs is the product of political negotiations, jurisprudence and legislation. When it comes to the question of who does decide upon the design of IGfRs, the case studies in this paper suggest that the key issues as to IGfRs are to a large extent negotiated in executive fora, with the us being an exception and Canada being the prime example to this regard. In Canada, executives play a pivotal role in any aspect of intergovernmental relations, also because its underlying principles and rules are neither enshrined in the constitution nor in any specific legislation (unlike in South Africa with its unique codification of intergovernmental relations). Even though intergovernmental relations in Canada are in essence of political nature, they can be considered as rather formalised when compared to, for example, the uk. In the uk, the principle of bilateral relations reigns between Westminster and its devolved governments in Scotland, Wales and Northern Ireland. In short, the uk government’s approach in designing and handling IGfRs is to react on a bilateral basis when forced to do so by establishing ad hoc commissions that elaborate recommendations. As a consequence, de jure asymmetries come into being by “necessity” and through “political negotiations”. Constitutional asymmetry can also be inherent to the institutional State engineering itself, either because cultural pluralism has

272

Alber

always been the hub of State building or State consolidation, or because of other historical reasons. For example, in the case of Spain, the particular political and fiscal arrangements in Navarre and the Basque Country had survived centralizing attempts of different Kingdoms, and, they are still in place. Bilateral negotiations have also characterised the development of IGfRs between the central government and the acs having an ordinary regime of regional financing, not least because the central government ultimately controls the sectoral consultative conference. Spain, Canada and Germany are three examples of States that have experienced a rather substantial decentralisation of their public finances, either on the spending side, the tax revenue side or on both sides. Canada stands out for being a federation with a high level of owntax revenues in relation to the totality of revenues of the central government, while Germany’s and Spain’s constituent units have a significantly lower level of tax autonomy and tax revenues in relation to the central government total tax revenues. While the acs have little leeway with regard to the decision on how the regional financing system has to look like, in Germany, the Länder can exert their influence through the Federal Council when it comes to taking decisions on how the complex system of cooperative federalism with its strongly harmonised tax system should look like. Even if different in detail, Australia, India and South Africa are examples of States in which finance commissions are regularly consulted when it comes to deciding on how revenues should be shared across governmental levels. Australia, a federation that unlike Canada initially was characterised by a concurrent taxation system but gradually drifted towards a centralised one, has the longest tradition of relying on independent agencies in the field of IGfRs. While India’s finance commission is appointed every five years by the president, South Africa opted for the creation of a permanent finance commission to further enhance its independence. Even though the remits of finance commissions widely differ across States, they all face a similar challenge: they have to elaborate formulas that balance equity with fiscal efficiency. Generally, equity considerations prevail, also against the backdrop of solidarity and equality principles that are enshrined in most constitutions. Such principles are increasingly also referred to by the federal level when it expands its spending power to the detriment of the spending power of the lower tier of government. The federal spending power keeps to be the most important lever of power of the federal government and it is one of the most controversially discussed instruments of IGfRs, most prominently but not only in Canada, where it did, however, unlike in the us, not lead to an extensive centralisation. While evidence from some States has shown that there is a compelling need to either make clarity on IGfRs (e.g. us) or totally overhaul them (e.g. India),

Intergovernmental Financial Relations

273

in other States IGfRs are regularly on the political agenda with reforms coming along in a piecemeal manner (e.g. Germany, uk). Against the backdrop of the 2008–2009 financial crisis and, within the eu, against the backdrop of its new economic and fiscal governance schemes, all case studies give evidence of either the creation or reinvigoration of so-called independent fiscal institutions. The structure and remit of the ifis widely vary from State to State. As a general rule, in order to properly function in parliamentary systems, ifis should be independent from both the executive and the legislative, while in presidential systems independence from the executive is sufficient. Moreover, ifis must have access to all relevant information to be able to carry out their functions. What is common to all ifis analysed in this paper, is the fact that they are consultative bodies and that their recommendations can be easily disregarded, also in presence of the “comply or explain” principle. Summing up, evidence from the case studies has shown that there is no single best institutional approach to designing, organising and handling IGfRs. A few experts could surely design in an optimal manner formal or informal mechanisms as well as define the remit of fora dealing with IGfRs, but, once they are in use, they have to come to terms with the characteristics and dynamics of the political system they are embedded in, at central and subnational level. Moreover, they are continuously challenged by changing socio-political frameworks. Their design has to be regularly adapted to shifting economic and socio-political conditions. The institutionalisation of IGfRs might undoubtedly help to enhance their durability and efficiency, but, as the case studies have shown, it does not present any guarantee with regard to their proper functioning. In the end, the robustness and sustainability of IGfRs largely depends on political practices. The evidence shown in this paper raises several implications for federal comparative research. Undoubtedly, the possibility to explicitly highlighting political and constitutional elements within a State in order to better understanding patterns of (institutionalised) IGfRs. Most importantly, however, the findings of this paper show that there is no single best approach to analysing and interpreting IGfRs, not least due to the fact that the essence of IGfRs is very difficult to grasp because, in its praxis, manifold political and economic factors do influence it. This implies that any comparative generalisations with regard to how IGfRs play out are a rather daunting task the outcome of which might nevertheless be of inspiration for academia, practitioners and the interested public. Ultimately, the most important lesson for comparative federal research might be that it should focus on the analysis of differences in particular institutional features and their cumulative effect rather than to assume that the form of government (broadly speaking) is inherently decisive of how IGfRs play out.

chapter 11

Accommodating Diversity While Guaranteeing Stability: The Role of Financial Arrangements Annika Kress 1

Introduction

The stability of any federal system essentially relies on the appropriate balances of the contrary (but complementary) concepts that lie at the very core of federalism.1 Among others, it needs to balance unity and autonomy, “­shared-rule” and “self-rule”,2 equality and diversity, solidarity and competition, symmetry and asymmetry. Beyond balancing and complementing their contraries, these concepts are also highly interdependent. It has been argued, that to some degree, symmetry relates to shared-rule as asymmetry relates to self-rule,3 and—to take the argument further—also relates in a similar way to equality and diversity, to unity and autonomy and so on. Tilting the balance too far in either direction could cause national–subnational conflicts that, in turn, could result in the instability of the system as a whole. Therefore, each federal system in its own right has to figure out the legal-constitutional and institutional set-up that caters to these balances, departing from its own local, regional and trans-regional realities. The importance of financial relations in federal systems can hardly be overstated. They are an essential, defining characteristic of any federation and directly relate to the autonomy of subnational units. Subnational revenue and expenditure powers are fundamental in determining the degree to which s­ubnational entities can autonomously exercise their subnational 1 The definition of federal systems that has been adopted in this chapter follows F. Palermo and K. Kössler, Comparative Federalism: Constitutional Arrangements and Case Law (Oxford, Hart Publishing, 2017). Therefore, it does not only include classical federations but also systems that may be referred to as regional or devolved systems, see footnote xx of the Introduction to this volume. 2 D.J. Elazar, Exploring Federalism (Tuscaloosa: University of Alabama Press, 1987), at 12. 3 F. Palermo, “Asymmetries in Constitutional Law: An Introduction”, in F. Palermo, C. Zwilling and K. Kössler, Asymmetries in Constitutional Law: Recent Developments in Federal and Regional Systems (Bolzano-Bozen: Eurac Research, 2009) 11–17, at 11.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_013

Accommodating Diversity While Guaranteeing Stability

275

­competences.4 Without sufficient resources, subnational units would simply not be able to carry out their responsibilities nor to pursue policies that reflect their own priorities autonomously. It is, however, a two-way relationship. The effectiveness of the federal system depends on financial relations while financial relations, in turn, derive from the very constitutional and institutional frameworks that define subnational units and their competences.5 As noted by Alice Valdesalici, a static system can hardly guarantee this effectiveness.6 A federal system needs to adapt continuously to the challenges stemming from global, national, regional and local dynamics. One of these challenges is, certainly, the constant negotiation of necessary degrees of unity with sufficient autonomy of subnational units. In other words, the particular financial arrangement, the “financial constitution”, crucially affects but also effects the fundamental federal principles that have been outlined above.7 This chapter will focus on the relationship between diversity and asymmetry in federal design. While asymmetries are to a certain degree unavoidable in any given system, they appear to be an especially prominent tool in attempts to mitigate tensions that stem from regional diversity.8 This chapter will examine the asymmetries in financial designs of five regionally diverse Western federal systems: Italy, Spain, Belgium, the United Kingdom, and Canada. These ­systems were chosen for the traits that they have in common: their status as welfare states, their considerable regionally rooted cultural and political identities, and the existing tensions that stem from these identities in relation to the federal system as a whole and the financial arrangements in particular. It will especially look at the ways in which these systems accommodate difference by constitutionally or politically allowing for financial asymmetries in the federal attempts to strike a working balance between diversity and unity.

4 See J. Rodden, Hamilton’s Paradox: The Promise and Peril of Fiscal Federalism (Cambridge: Cambridge University Press, 2006). 5 R.D. Congleton, “Asymmetric Federalism and the Political Economy of Decentralization”, in E. Ahmad, and G. Brosio (eds.), Handbook of Fiscal Federalism (Cheltenham: Edward Elgar, 2006) 131–153, at 131. 6 A. Valdesalici’s chapter, “Defining Fiscal Federalism” in this volume. 7 For the concept of financial constitution, see S. Parolari’s chapter in this volume, or G. Färber, “Fiscal Equalization in Germany”, in G. Pola (ed.), Principles and Practices of Fiscal Autonomy: Experiences, Debates and Prospects (Farnham, Ashgate, 2015) 113–134, at 118. 8 The approach taken to diversity in this chapter follows that of Luis Moreno and César Colino in L. Moreno and C. Colino, “Introduction”, in L. Moreno and C. Colino, Diversity and Unity in Federal Countries (Montreal: McGill-Queen’s University Press, 2010) 3–15, at 3.

276 2

Kress

On Asymmetries

While for a long time federal studies have neglected the study of asymmetrical features in favour of the study of symmetrical features of subnational units’ status, at this point, asymmetries have lastingly made their way into the spotlight and firmly established themselves in the debates of federalism. Social, political, cultural or economic differences among subnational units virtually rule out complete symmetry and make asymmetries part and parcel of every federal system.9 Today, the scholarly debate focusses more on the impact asymmetries have on the stability of the federal systems. In what can be considered one of the seminal theoretical articles for the study of symmetries and asymmetries in federal systems, Charles D. Tarlton argues that asymmetrical federal features are directly related to conflicts ­between the federal level and subnational units.10 He contends that the simultaneous occurrence of national–subnational conflict across most or all units— an indication of symmetry—is unlikely and more generally occurs “where the relationship between local and central authorities corresponds to the image of an asymmetrical situation, and where that asymmetry is characteristic only of a few of the states in their relation to the whole”.11 According to this line of argumentation, a federal system strongly characterised by subnational diversity would thus require the dominance of the “forces of unity” over those of separatism in order to remain stable, which would require a strong central authority.12 Michael Burgess criticises this approach in that it fails to differentiate between what he calls “the preconditions of asymmetry” and “asymmetrical outcomes”.13 The “preconditions of asymmetry” would be the cultural, ­societal, economical, financial, territorial conditions, the existing empirical ­circumstances, whereas the “asymmetrical outcomes” are the political and institutional arrangements, legislation and intergovernmental relationships that result from these factors. By recognizing diversity through autonomy arrangements, the economics of fiscal federalism thus need to address the different 9 10 11 12 13

I.D. Ducachek, Comparative Federalism: The Territorial Dimension of Politics (Lanham: University of America Press, 1987). C.D. Tarlton, “Symmetry and Asymmetry as Elements of Federalism: A Theoretical Speculation”, The Journal of Politics, 27 (1965), at 871. Ibid. Ibid, at 874. M. Burgess, Comparative Federalism: Theory and Practice (London: Routledge, 2006), at 217–221.

Accommodating Diversity While Guaranteeing Stability

277

s­ ubnational realities. The needs of citizens in different parts of the country inevitably differ and local public administration have to adjust accordingly.14 This justifies the introduction of mechanisms of financing, distribution and ­management of resources that enable the system to cope with a plurality of autonomous choices by the subnational units. In this context, the literature usually differentiates between de facto—or political—and de jure asymmetries.15 Simply put, de jure asymmetries are enshrined in law and in that sense an outcome; a legal reaction to a perceived need for differentiation. De facto asymmetries can result from these legal provisions, as well as from public policy and reversible intergovernmental agreements and even be related directly to the “preconditions of asymmetry”. As we will see in the case studies below, these preconditions—for our purposes mainly the regional and crossregional diversity—can have strong implications for political and ideological landscapes based on subnational identities, which is why some scholars argue that asymmetries may even constitute a necessary tool for achieving federal stability.16 Notwithstanding, asymmetrical territorial frameworks do pose additional questions and challenges, especially when it comes to the adoption of specific financial mechanisms to match differentiated institutional patterns. A differentiated system of financing has to be weighed against equity–uniformity of resource distribution; otherwise a special treatment could easily evoke the impression of discrimination based on the place of residence. Here, individual fundamental rights especially raise equality concerns. How can social and civil rights be guaranteed equally throughout an asymmetrical system? Turning the question around, however, may raise equity concerns. Can subnational diversity be sustained in too symmetrical a system? In terms of a system’s comprehensive stability, unity and sustainability, the regulation of financial relations should aim to reconcile subnational financial responsibility for own powers and state-wide solidarity with structurally and economically weaker subnational units. In other words, a compromise is required between autonomy demands and solidarity obligations. 14

For more on this argument, see W. Oates, Fiscal Federalism (Cheltenham: Edward Elgar, 2011 [1972]); C. Tiebout, “A Pure Theory of Local Expenditures”, The Journal of Political Economy, 64 (1956) 416–424; M. Olson, “The Principle of “Fiscal Equivalence”: The Division of Responsibility among Different Levels of Government”, American Economic Review, 59 (1969) 479–487, at 479. 15 See, e.g. R.L. Watts, Comparing Federal Systems (2nd edn; Montreal & Kingston: McGill– Queen’s University Press, 1999), at 63. 16 Burgess, Comparative Federalism, supra, at 221.

278 3

Kress

Case Studies

Federal systems’ asymmetrical designs rely on a wide variety of factors. ­ ccording to Ronald L. Watts, they can be politically motivated or based on A differential subnational capacities, can be transitional or lasting features of the system, can be de jure or de facto, relate to peripheral and full-fledged subnational units.17 It is important to take into consideration the circumstances within the single federal system, its subnational social, political, territorial make-up and the resulting diversity, as well as the specific asymmetries that have been introduced in order to be able to evaluate their contribution to or subtraction from federal stability.18 In turn, fiscal federalism can translate into a huge variety of asymmetric solutions in the regulation of special territories’ public finances, as the diverse reasons for differentiation of the financial system call for different financial mechanisms and instruments. Ultimately, financial relations should ensure adequate subnational funding, while ensuring the political accountability of subnational entities for the decisions they adopt. Each system therefore has to decide how to distribute resources (and responsibilities) so that it caters to the specific needs of the subnational entities and to what extent to limit subnational financial autonomy in favour of a more centralised, symmetrical system that would presumably take a more neutral approach to the redistribution of funds. Too much centralisation, however, would subdue the exercise of self-government and the objective of a government closer to the people. Financial responsibility as an incentive for greater efficiency could also be lost.19 Too much asymmetry, on the other hand, could put into question the principle of equality of citizens. Thus, the legal tools that regulate intergovernmental financial relations should ensure adequate funding for responsibilities at all government levels without centralizing all decision-making powers relating to revenue and expenditure in order to provide for the possibility of subnational differentiation.20 17 18

19 20

Watts, “Asymmetrical Decentralization”, supra. R.M. Bird and R.D. Ebel, “Subsidiarity, Solidarity and Asymmetry: Aspects of the Problem”, in R.M. Bird and R.D. Ebel, Fiscal Fragmentation in Decentralized Countries: Subsidiarity, Solidarity and Asymmetry (Cheltenham, Edward Elgar, 2007) 3–25. See Olson, “The Principle of “Fiscal Equivalence”, supra. For more on this argument, see W.E. Oates, “An Essay on Fiscal Federalism”, Journal of Economic Literature, (1999) 1120–1149; R.H. Gordon, “An Optimal Taxation Approach to Fiscal Federalism”, Quarterly Journal of Economics, 98 (1983) 567–586; J. Rodden, “The Dilemma of Fiscal Federalism: Grants and Fiscal Performance around the World”, American Journal of Political Science, 46 (2002) 670–687.

Accommodating Diversity While Guaranteeing Stability

279

The following section will consider symmetrical and asymmetrical approaches to revenue autonomy, expenditure autonomy and equalisation mechanisms in order to better understand their role in sustaining the unity of federal systems. 3.1 Italy In total, Italy’s regional system consists of 20 regions that form the first-level of subnational government. After World War ii, the drafters of the Italian Constitution were confronted with the challenge of reconciling the principle of national unity with the substantial diversity among its regions, especially those at the margins. In light of its international obligations, socio-political tensions and geo-political practicality, five regions were awarded special constitutional status: the autonomous regions.21 Italy’s autonomous regions are a case in point to show that de jure and de facto asymmetries are inseparable categories and can indeed be part and parcel of each other. At the outset, its five special regions—Aosta Valley, FriuliVenezia-Giulia, Sardinia, Sicily, and Trentino-Alto Adige—all enjoy the same potential for autonomy that is provided by Article 116 of the Italian Constitution. It states that they “have special forms and conditions of autonomy pursuant to the special statutes adopted by constitutional law”. Their special regimes thus rest upon their respective Autonomy Statutes that hold constitutional rank, making exceptions from the general constitutional and regional design possible; special autonomy is considered a fundamental constitutional principle. The degree to which their specialty results in differentiated arrangements compared to those of ordinary regions, however, depends vastly on the particular political culture in the respective bilateral negotiations. It could even be argued that Trentino-South Tyrol and the Aosta Valley are the only truly autonomous regions in Italy,22 as both have successfully, through negotiation with the State (government), acquired new matters that go beyond the p ­ owers

21

22

J. Woelk, “Autonomie speciali e integrazione, riforma costituzionale (e degli Statuti)”, in F. Palermo and S. Parolari (eds.), Il futuro della specialità regionale alla luce della riforma costituzionale (Napoli: Edizioni Scientifiche Italiane, 2016). R. Bin, “L’autonomia e i rapporti tra esecutivo, legislativo e le commissioni paritetiche”, in A. Di Michele, F. Palermo and G. Pallaver (eds.) Fine di un conflitto: Dieci anni dalla chiusura della vicenda sudtirolese (Bologna: Il Mulino, 2003) 205–218, as cited in F. Palermo, “Autonomy and Asymmetry in the Italian Legal System: The Case of Autonomous Province of Bolzano/Bozen”, in G. Pola (ed.), Principles and Practices of Fiscal Autonomy: Experiences, Debates, and Prospects (London: Routledge, 2015) 228–241.

280

Kress

stipulated in their autonomy statutes.23 Sardinia’s and Sicily’s institutional autonomy, on the other hand, has been limited and shaped by the “creation of political ties with the central government and on the benefits arising from the considerable financial aid granted by the state”, which has resulted in regimes that closely resemble those of regions with ordinary statute.24 In other words, it is the political will and regional demand for self-government that have translated the constitutional provision of specialty of the region into the statutory de jure and de facto asymmetrical manifestations of this autonomy and the resulting asymmetrical subnational systems. Considering financial relations, only Article 119 of the Italian Constitution states some very general principles of subnational revenue and expenditure autonomy. While the financial regimes of the special regions are outlined in their autonomy statutes, ordinary regions depend on national legislation. While this does not mean that the special fiscal regimes remain completely outside of the national system and enjoy full and independent taxing power, each special statute does, however, contain a financial guarantee or even indicate a quota of the national taxes generated in the respective territory that is to remain in the region for the exercise of autonomous responsibilities; as well as a list of shared taxes.25 Furthermore, the central government cannot unilaterally modify the special regions’ revenue regimes, as each of them has contractual power in financial matters.26 In Trentino-South Tyrol, for example, Chapter 6 of the Autonomy Statute covers “The Finances of the Region and the Provinces” and stipulates, for example, the percentage of national taxes collected in the region ceded to the region, such as revenue from the inheritance tax, the vat, the gambling tax, etc. The special regions’ financial systems are therefore highly disconnected from the one that applies to the 15 “ordinary” regions. Effectively, the special regions’ financial arrangements translate to 25–90% of revenue raised on their territory returned to the regions over which they have complete spending power.27 This does, however, also mean that 23 24 25

26 27

G. Cerea, “The Italian Regions with Special Autonomy and the Case of Trentino-Alto Adige” in Pola (ed.), Principles and Practices (2015) 243–263. Palermo, “Autonomy and Asymmetry in the Italian Legal System”, supra, at 232. See art. 36 of the Statute of Sicily; arts. 8 and 10 of the Statute of Sardinia; art. 12 of the Statute of the Aosta Valley; art. 49 of the Statute of Friuli-Venezia Giulia; art. 75 of the Statute of Trentino–South Tyrol. See F. Guella, “Il principio negoziale nei rapporti finanziari tra livelli di governo”, Le Regioni, 1–2 (2014), at 131. A. Valdesalici, “Features and Trajectories of Fiscal Federalism in Italy”, in S.A. Lütgenau (ed.), Fiscal Federalism and Fiscal Decentralization in Europe: Comparative Case Studies on Spain, Austria, the United Kingdom and Italy (Innsbruck: Studienverlag, 2014) 73–101, at 94.

Accommodating Diversity While Guaranteeing Stability

281

the funding levels are dependent on the state of the regional economy. While special regions can benefit a lot from this system in times of well-functioning economy and economic growth, during times of economic crises, the opposite applies. Italy’s asymmetric financial relations signify first and foremost asymmetric constitutional protections of financial resources, both in quantitative and in qualitative terms. Quantitatively, available resources are higher for those special entities vested with broader legislative and administrative competences. Qualitatively, each special entity bindingly negotiates its own fixed share of national tax revenue. The latter is also the case for the special regions involvement in the national solidarity and equalisation system, for which the agreements with the northern regions were signed following the passing of Law 42/2009.28 A very important consequence of these negotiations is a further increase in autonomy. They can contribute to the national solidarity system by taking over central state responsibilities at their own expense. In other words, they can opt for managing new powers autonomously with the same financial resources they had negotiated before. Once again, this possibility has created asymmetries among asymmetrical regions as only the Northern, richer special regions have taken this road and enlarged their sphere of autonomous powers by assuming additional responsibilities at their own expense. As a matter of fact, direct central government expenditure in Trentino-South Tyrol and Aosta Valley has at this point become limited to managing taxes, the judiciary and matters of public security.29 Over the years, the bilateral negotiations between the central government and these regions have resulted in vast areas of what in general would be central government responsibilities now being financed and managed with these special regions’ budgets, including aspects of education, social security and transport.30 Regarding the degree to which Italian special regions have shaped their autonomies and transformed de jure asymmetrical provisions into de facto 28 29 30

Ibid, at 91–92. Ibid. For recent examples of the procedures and bilateral negotiations with regard to financial questions, as envisioned by the autonomy statutes, see the Accordo di Milano (2009) that implemented law 42/2009 regarding fiscal federalism, or the Patto di garanzia (15 October 2014) between the Italian State and the Autonomous Provinces of Trento and BolzanoBozen/South Tyrol. See also A. Valdesalici, “Il nuovo accord finanziario per il TrentinoAlto Adige/Südtirol”, Il Commercialista Veneto, 222 (2014) 23–24; and F. Guella, Sovranità e autonomia finanziaria negli ordinamenti composti: La norma costituzionale come limite e garanzia per le dimensioni della spesa pubblica territoriale (Trento: Università degli studi di Trento, 2014).

282

Kress

asymmetries highlight the important difference Burgess makes between “the preconditions of asymmetry” vs. “asymmetrical outcomes”. They especially emphasise the importance of political culture and tradition, social cleavages and economic aspects that influence the bilateral political negotiations that bring to life constitutional provisions. 3.2 Spain Spain’s first subnational level comprises 17 autonomous communities. Regarding the general principles of the territorial organisation of the state, Article 138.1 of the Spanish Constitution guarantees “the effective implementation of the principle of solidarity […] by endeavouring to establish a fair and adequate economic balance […]”, which is based on the “[recognition and guarantee of] the right to self-government of the nationalities and regions of which [the Spanish Nation] is composed and the solidarity among them all” (Article 2). Importantly, Article 138.2 goes on to proclaim that “[d]ifferences between Statutes of the different Autonomous Communities may in no case imply economic or social privileges”. Concurrently, the Spanish Constitution provides for substantial de jure asymmetries in intergovernmental financial relations. Additional Provision 1 states that “[t]he Constitution protects and respects the historic rights of the territories with traditional charters (fueros)”. These fueros or régimen foral (foral regime) grant the autonomous communities of Navarre and the Basque Country the “historic rights” to specific areas of ­self-government, among them special fiscal arrangements that had remained in place for most of the history of Spain as a country. Under these fiscal arrangements,31 Navarre and the Basque Country are granted extensive tax autonomy that are defined in the Basque Country’s economic accord32 and Navarre’s economic convention.33 Essentially, what are generally national taxes become taxes of the two autonomous communities. These include, among others, substantial revenue sources, such as the personal income tax, the wealth tax, and the corporate tax. While they do not have taxing powers over the value added tax that the Spanish state sets the tax rate for, they also collect the revenue from this tax in their respective territories. Furthermore, 31

32 33

See, Ley Orgánica 8/1980, de 22 de septiembre, de Financiación de las Comunidades Autónomas (lofca), Organic Law 8/1980 of 22 September 1980, additional regulations 1 and 2. Concierto Económico con la Comunidad Autónoma del País Vasco, Law 12/2002 of 23 May 2002. Convenio Económico entre el Estado y la Comunidad Foral de Navarra, Law 28/1990, of 26 December 1990.

Accommodating Diversity While Guaranteeing Stability

283

both ­autonomous ­communities have complete spending power of the revenues they collect.34 Since this means that the Spanish state collects none of the taxes in their territories except for customs duty, the Basque Country and Navarre transfer additional amounts of money to the Spanish state for the services that it carries out in their territories. Furthermore, Navarre and the Basque Country do not participate in the equalisation mechanism, but only in the regional development fund through their transfers to the Spanish states.35 The autonomous communities under the common Spanish regime do not at all enjoy the same level of financial autonomy as the ones under the foral regime, although the system of financing has become somewhat more decentralised over the last 25 years. While initially, only minor taxes were assigned to the communities under the common system and their financing depended to a large degree on unconditional grants from the Spanish state, reforms in the 1990s first introduced small percentages of tax shares (15% of the personal income tax). Subsequent reforms have augmented these shares and following the reform in 2009, the communities under common regime now collect considerable shares of tax revenue from the personal income tax, the value added tax and excise taxes.36 Nevertheless, it is still the Spanish state that mostly determines tax rates and tax bases and all common regime communities participate in both solidarity mechanisms. The special treatment of the communities under the foral regime have certainly led to discontent in other communities, especially so in Catalonia. Much of the recent debate surrounding Catalonia’s position in the Spanish system indeed relates to the Spanish financial system and the financial burden that Catalonia bears. As an autonomous community of common regime and one of the economically strongest, Catalonia does contribute significantly higher 34

35 36

For more information on the foral regimes, see J. López Laborda and C. Monasterio Escudero, “Regional Governments: Vertical Imbalances and Revenue Assignments”, in J. Martinez-Vázquez and J.F. Sanz (eds.), Fiscal Reform in Spain: Accomplishments and Challenges (Cheltenham: Edward Elgar, 2007) 422–452; V.R. Almendral, “The Asymmetric Distribution of Taxation Powers in the Spanish State of Autonomies: The Common System and the Foral Tax Regimes”, Regional and Federal Studies, 13 (2004) 41–66; C. Monasterio Escudero, “Un análisis del sistema foral desde la perspectiva de la teoría del federalismo fiscal”, in C. Monasterio Escuderio and I. Zubiri Oria (eds.), Dos ensayos sobre financiación autonómica (Madrid: Funcas, 2009) 167–246; and F. de la Hucha Celador, Introducciòn al regimen jurìdico de las haciendas forales (Madrid: Civitas, 1995). J. Ruiz-Huerta, “Spanish Decentralization and Fiscal Federalism”, in Pola, Principles and Practices (2015) 185–202, at 195. Ibid, at 194 and Law 22/2009 of 18 December 2009, regulating the financial system of the Autonomous Communities of the common regime.

284

Kress

amounts of money to the Spanish system than the average ac, while it has around the same amount of resources per capita available.37 It has, indeed, been suggested repeatedly that changing the Catalonian fiscal arrangement into one that resembles that of Navarre and the Basque Country under the foral regime could potentially add to the overall stability of the Spanish system.38 The Spanish case, thus, raises interesting questions about differential accommodation of regional diversity in financial systems and whether different treatment of difference itself leads to instability when it comes to “preconditions of asymmetry” that are based on history rather than on current social, political, economic or cultural givens.39 3.3 Belgium Even compared to other federations, the Belgian federal system is highly complicated and entangled. Departing from the unitary set-up that characterised Belgium for more than a century after its independence of 1830, six major state reforms have shaped the country into today’s complex federal system.40 Ignoring the tensions stemming from the refusal to officially recognise the Dutch language well into the 20th century, the rising demands for cultural autonomy of Flemish community more and more showed the inadequacy of the unitary system in dealing with Belgium’s regional diversity. In order to finally meet these demands and to prevent the situation from further deteriorating, the first state reform was passed in 1970, which set the path for Belgium’s development into the “federal state, composed of communities and regions” that it is today.41 Throughout the reforms, the communities and regions have gained substantial 37

38 39 40

41

A. Castells, “The Relation Catalonia/Spain: Some Financial and Economic Aspects”, in Pola, Principles and Practices (2015) 203–225, at 214. For more on the discourse surrounding the balances fiscals (fiscal imbalances) in Spain, see N. Bosch and M. Espasa, “Les balances fiscals: Concepte, mètode i aplicacions”, El Clip, 35 (2006); A. Rodríguez Bereijo, “Constitución Española y financiación autonómica”, in F. Pau i Vall (ed.), La financiación autonómica: xvi Jornadas de la Asociación Española de Letrados de Parlamentos (Madrid: Editorial Tecnos, 2010); and A. de los Ríos Berjillos, “Los objetivos de financiación autonómica: Una perspectiva desde los principios de autonomía, solidaridad, coordinación y suficiencia”, Presupuesto y gasto público, 26 (2001) 173–188. V. Ferreres Comella, The Constitution of Spain: A Contextual Analysis (Oxford: Hart Publishing, 2013), at 192. See A. Herrero Alcalde, “The Spanish Way of Fiscal Federalism”, in Lütgenau, Fiscal Federalism and Fiscal Decentralization in Europe, supra, 15–39, at 16. These took place in 1970, 1980, 1988–89, 1993, 2001–03 and 2011–14. For a comprehensive overview over the evolution of the Belgian federal system, see M. Gérard, “Economic Aspects of Constitutional Change: The Case of Belgium”, Oxford Review of Economic Policy, 30 (2014) 257–276, Table 5. Belgian Constitution, Article 1.

Accommodating Diversity While Guaranteeing Stability

285

autonomy in major policy fields and in 1993 with the fourth state reform federalism became the official form of government of the state. Indeed, the Belgian case has been characterised as, “[lacking] any true national foundation” and its federal system to have been established in order to ensure the preservation of a single state that suits all of its subnational communities.42 As Popelier and Koen argue, however, “[p]resumed complexity, consensus-thinking and compromises have too often led to a patchwork approach to constitutional changes, with the Belgian population paying the price in the present for a lack of long-term vision in the past”.43 The process of government formation after the general elections in 2010 that lasted a record-long 542 days exemplifies the politically fragmented landscape and divided society of Belgium today, as well as the capacity of finally achieving compromise.44 According to Article 1 of the Constitution, three communities and three regions make up Belgium’s overlapping subnational entities. The communities have been split more or less along language borders and are the Flemish, the French and the German-speaking community. The bilingual area of Brussels, in turn, is composed of Dutch-speakers and French-speakers belonging to their respective linguistic communities, according to the logic of personal autonomy. The regions instead follow the logic of territorial autonomy and are the Brussels-Capital Region, the Flemish Region and the Walloon Region. One of the major asymmetries within the Belgian system is that the Flemish Region and the Flemish community have merged to become one entity with one parliament and one government, whereas the French community and the Walloon region remain separate entities that deal with their respective competencies. According to Arts. 127–129 of the Constitution, the parliaments of the Flemish and French communities have the power and responsibility over regulation of, e.g. cultural matters; education (with a few exceptions); cooperation between the Communities; person-related matters; international cooperation; language-use in public matters. Regions’ competencies, according to Article 39 of the Constitution, are outlined in specific laws but generally relate to topics of the economy and territorial management. In large parts, the state reforms were necessary in order to overcome continuous institutional crises and instability and pursued the goal of mitigating existing conflicts and reducing the potential for future discords within the 42 43 44

B. Bayenet and P. de Bruycker, “Belgium: A Unique Evolving Federalism”, in Bird and Ebel, Fiscal Fragmentation (2007) 169–207, at 171. P. Popelier and K. Lemmes, The Constitution of Belgium: A Contextual Analysis (Oxford: Hart Publishing, 2015), at v. K. Deschouwers and M. Reuchamps, “The Belgian Federation at a Crossroad”, Regional & Federal Studies, 23 (2013).

286

Kress

socially, linguistically, culturally and economically diverse country. Especially the tensions between the French-speaking and Dutch-speaking Belgians have played a major role in the negotiations that have shaped to the current system. Furthermore, “[i]n the exercise of their respective responsibilities, the federal State, the Communities, the Regions and the Joint Community Commission act with respect for federal loyalty, in order to prevent conflicts of interest”.45 Considering the relatively short period in which the Belgian federal system has evolved, the amount of competencies that have been devolved to the subnational level and the increase in subnational autonomy is very substantial. Far-reaching autonomy of the subnational entities has been the recipe for maintaining the State. The sixth and latest reform has seen further extensive power transfers and major autonomy gains for the regions, which for the first time includes transfers of social security matters. Arguably, most powers now lie at the subnational level, rather than at federal level.46 Together with the asymmetrical evolution of the subnational entities, this obviously has a considerable impact on the symmetry of the subnational arrangements and fosters further differentiation of the subnational institutional developments. Through the state reforms however, it has become clear that the de jure relationship of the Flemish community and region and the French community and Walloon region with the federation is rather symmetrical, indeed. The devolution of powers from the federation has applied to the French and the Flemish subnational entities equally, and their constitutionally assigned role and standing is, if not equal, as well, then at least more or less proportional.47­ The asymmetries that follow from these de jure provisions rather stem from the regions’ and communities’ own exercise of powers and interpretations of the laws at hand, as well as from the peculiar asymmetric pre-­conditions. Considering the rationale behind the state reforms, this does indeed make sense. In a small federal system with two major subnational groups that both have strong negotiating powers, an arrangement that is considered to be more advantageous to one group and disadvantageous to the other would of course be heavily disputed by the disadvantaged group, thus being an obstacle to 45 46

47

Article 143 Belgian Constitution, emphasis added. For more, see J. Goossens and P. Cannoot, “Belgian Federalism after the Sixth State Reform”, Perspectives on Federalism, 7 (2015), at E-45; and B. Bayenet and J.-F. Husson, “Le Financement des entités décentralisées dans la nouvelle Belgique fédérale”, in La dotation globale de fonctionnement en question, éléments d’évaluation et perspectives: Contributions pour un débat d’actualité (Paris: acuf/amgvf/apvf, 2015) 112–129. See Articles 67 and 99 on senate members and members of the council of ministers, as well as articles 127ff and 135ff. that outline regional and communities’ powers.

Accommodating Diversity While Guaranteeing Stability

287

c­ ompromise. ­Differentiation by the exercise of autonomous powers does not have the same effect—at least in legal terms. These kind of dynamics are quite typical of binational federations. The financial relations between the center and the subnational entities follow a similar rationale. Title v of the Constitution assigns the establishment of the system of financing of the communities and regions to a special federal law, i.e. the 1989 Special Law on the Financing of the Communities and Regions, as most recently amended by the Special Finance Act of 6 Januar 2014. From an institutional perspective, this law treats the Flemish and French communities and the Flemish and Walloon Regions symmetrically, despite the considerable differences in economic strength on the subnational level. Some of its reforms even foresee and attempt to mitigate their own potential disadvantages. The 2014 Special Finance Act, for example, foresees a major increase in subnational fiscal autonomy, which, implemented by itself and all at once would likely have been to the detriment of the Walloon Region. Therefore, a temporary mechanism was introduced through which a transitional amount will soften the financial impact of the reform on regions and communities over ten years after the introduction of the law.48 The actual de jure asymmetry in the Belgian fiscal system concerns the much smaller German-speaking community. Article 176 provides for a separate system of financing for the German-speaking community, which is mainly financed through federal grants as regulated in the (ordinary) Law on the Institutional Reform concerning the German-speaking Community of 31 December 1983 and transfers from the Walloon region with regard to the regional competences that it has devolved to the German-speaking community. Rather than political frictions, this financial asymmetry is attributable to the small size of the community that only counts a population of approximately 75 000, less than 1% of the total national population. 3.4 United Kingdom Including the United Kingdom in a list of federal systems is certainly debatable. Nonetheless, considering the devolution of powers over the last 20 years, the resulting increase in subnational autonomy and moreover the abovementioned­“preconditions of asymmetry”, it does seem justified, in particular with regard to the chapter’s focus on asymmetrical arrangements. Until the end of the 20th century, there would have been no discussion at all about the nature of the United Kingdom’s categorisation as a centralised 48

J. Goossens and P. Cannoot, “Belgian Federalism after the Sixth State Reform”, Perspectives on Federalism, 7 (2015), at 43.

288

Kress

unitary state. The top-level administrative division of the United Kingdom of Great Britain and Norther Ireland are the four countries England, Scotland, Wales and Northern Ireland that until the first devolution legislation was introduced in 1998 held no legislative powers of their own.49 This changed with the Welsh and Scottish referendums on devolution in 1997 and the provisions of the Good Friday Agreement in 1998 that—in various steps—led to the establishment of the devolved legislatures of the Scottish Parliament, the National Assembly for Wales and the Northern Ireland Assembly and the devolved administrations of the Scottish government, the Welsh government and the Northern Ireland Executive. Each of these political bodies has been devolved powers as laid out in the respective devolution acts. England, in turn, does not have its own regional parliament, a point of contention that several proposals for the formation of such have attempted to modify.50 Devolution has been financed almost entirely through the transfer of grants. At the basis stands a national and unitary tax system. The necessary amounts of revenue to cover expenditure needs have been guaranteed through a mechanism that is known as the Barnett Formula rather than through considerable own taxing powers of Scotland, North Ireland and Wales.51 It calculates the increases in the amount of expenditure that are available for the English administration for a public service against the proportion of the population, while also factoring in the extent to which the devolved public service is comparable.52 The uk treasury continues to use this formula to adjust the funds available to the devolved administration, year by year. As such, the Barnett Formula can be considered as the United Kingdom’s equalisation mechanism, in that it attempts to close the vertical gap between devolved expenditure responsibilities and own fiscal capacity in the territories in order to provide for similar service standards throughout the union.53 49 50

51

52 53

Administrative decentralisation on the other hand, had been quite advanced already. See, V. Bogdanor, “The West Lothian Question”, Parliamentary Affairs, 1(2010) 156–172 and, more generally, P. Leyland, The Constitution of the United Kingdom: A Contextual Analysis (Oxford, Hart Publishing, 2012). See D. Heald and A. McLeod, “Beyond Barnett? Financing Devolution”, in J. Adams and P. Robinson (eds.), Devolution in Practice: Public Policy Differences within the uk (London: Institute for Public Policy Research, 2002) 147–175; D.N.F. Bell “Territorial Finance and the Future of Barnett”, The Political Quarterly, 86 (2015) 209–216; J. Barnett, “The Barnett Formula: How a Temporary Expedient Became Permanent”, New Economy, 7 (2000) 69–71. See, T. Edmonds, “The Barnett Formula”, House of Commons Library, Research Paper 01/108 (2001). J. Martinez-Vazquez and B. Searle, Fiscal Equalization: Challenges in the Design of Intergovernmental Transfers (New York: Springer, 2007), at 294.

Accommodating Diversity While Guaranteeing Stability

289

Subnational financing is thus construed to be neither dependent on discretionary decision-making by the central government nor on the local capacity to administer own taxes.54 The core of the system is, on the contrary, a special grant system built on and perpetuated by an automatism based on convention that has received significant criticism: We have concluded that the Barnett Formula should no longer be used […]. Although the annual increment in funds is made on the basis of recent population figures, the baseline […] does not reflect today’s population in the devolved administrations. The Barnett Formula also takes no account of the relative needs of any of the devolved administrations.55 As an exception, in the 1998 Scotland Act, Scotland had also been given tax raising power with the so-called Scottish Variable Rate. It could adjust the uk’s personal income tax rate by up to three percent. The Scottish parliament, however, never made use of this power, the additional administrative cost and the likely political consequences were considered to outweigh the economic benefits.56 The Scotland Acts 2012 and 2016 further devolved this tax raising power and now, “[t]he Scottish Parliament may by resolution (a “Scottish rate resolution”) set the Scottish basic rate, and any other rates, for the purposes of Section 11A of the Income Tax Act 2007 […]” (Scotland Act 2016, Section 13(2)). Since the election of the Scottish National Party to (minority) government in 2007, the debate on how to accommodate Scotland’s regional diversity has flared up again and gained even further ground after it won by absolute majority in the 2011 elections. Similar to the Catalonian case, much of the debate surrounding Scottish accommodation leading to the 2014 referendum on independence, has focused on fiscal decentralisation.57 In an attempt to sway voters towards “remain”, the uk government assured further devolution of powers that was introduced in the 2016 Scotland Act. The Scottish referendum on independence is a prime example for the mitigation of secessionist claims

54 55

56 57

See B. Ashcroft et al., “Flaws and Myths in the Case for Scottish Fiscal Autonomy” Fraser of Allender Quarterly Economic Commentary, 31 (2006) 33–39. House of Lords—Select Committee, “The Barnett Formula. Summary” (2009), https:// publications.parliament.uk/pa/ld200809/ldselect/ldbarnett/139/13903.htm (last accessed 08 March 2017). S. Eden, “United Kingdom”, in Bizioli and Sacchetto, Tax Aspects (2011), at 545. C. Jeffrey, “The Debate on Fiscal Decentralization in Scotland”, in Pola, Principles and Practices (2015), at 155.

290

Kress

through the promise of increased autonomy, or “devolution-max”.58 Next to other important stipulations, such as the newly introduced permanence of the Scottish legislature and administration, financial relations between Scotland and the uk have changed significantly. Sections 13–15 of the 2016 Scotland Act, devolve further taxing powers over the income tax, Section 16 reserves substantial amounts of the value added tax collected on Scottish territory to Scotland and Sections 17 and 18 devolve the “tax on carriage of passengers by air” and the “tax on commercial exploitation of aggregate”. Neither Wales, nor Northern Ireland have been devolved any comparable financial powers. In the light of “Brexit”, the future of intergovernmental relations in the United Kingdom and further devolution is the topic of much debate with the Scottish independentist movement re-gaining lost ground or the apparently increased political attractiveness of a federalising process.59 3.5 Canada Looking beyond the European borders, Canada’s federal system provides an interesting case of financial asymmetries and the accommodation of subnational diversity. Among the chosen cases, it is the only one that has formally been a federation from the foundation of the country; today, it is subdivided into ten provinces and three territories, of which only the provinces derive their powers from the constitution, whereas it is the Canadian Parliament that delegates powers to the territorial governments.60 While most provinces are majoritarian English-speaking, the vast Canadian territory is home to a large number of minority nations and minority languages. Especially in the light of federal stability, the only majoritarian French-speaking province of Quebec will be focussed on in this chapter. Similar to the Belgian case, the Canadian (financial) Constitution remains rather symmetrical in the treatment of its provinces. While it does provide for small number of asymmetries between the provinces, it could hardly be argued that the Constitution itself treats any single one of them as extraordinarily different.61 The asymmetries between the Canadian subnational financial 58

59

60 61

For a definition of the concept, see M. Keating, “Rethinking Sovereignty. IndependenceLite, Devolution-Max and National Accommodation”, Revista d’estudis autonòmics i federals, 16(2012) 9–29, at 11. S. Carrell, “Scottish Parliament Votes for Second Independence Referendum” The Guardian (28 March 2017); S. Carrell, “Jeremy Corbyn pledges to consider more federalised uk” The Guardian (15 May 2017). Considering the scope of this chapter, we will only take into consideration the systems of financing of the provinces. R. Iacovino, “Partial Asymmetry and Federal Constitution: Accommodating Diversity in the Canadian Constitution”, in M. Weller and K. Nobbs (eds.), Asymmetric Autonomy and

Accommodating Diversity While Guaranteeing Stability

291

set-ups are rather the result of a contractual approach to federalism that allows provinces to opt out of federal programs that are widely used to finance the provinces within a highly developed cooperative federal design.62 Nevertheless, Quebec’s adversarial role in the evolution of the Canadian federal system has without doubt left its mark also on the system of financial relations. The Constitution Act 1867 bestows on the Canadian parliament the legislative authority “to make Laws […] in relation to […] 3. The raising of Money by any Mode or System of Taxation” (Section 91).63 Subsequently, it limits the taxing power of the provincial parliaments to “[d]irect Taxation within the Province in order to the raising of a Revenue for Provincial Purposes” (­Section 92(2)), and certain license fees “in order to the raising of a Revenue for Provincial, Local, or Municipal Purposes” (92(9)). As Alarie and Bird argue, the way in which the federal and provincial taxing powers are phrased and the broadness of the purposes they concern, indicate not only a concurrent power over direct taxation but also only a vague delimitation of provincial powers over indirect taxes.64 Considering the issue of concurrent tax bases in the light of the burden it can place on the taxpayer, it is no surprise that there has been much litigation over the constitutionality of tax laws.65 The Canadian solution to the challenges imposed by concurrency of tax bases has been to regulate them through intergovernmental agreements.66 This became especially important with the increasing weight of the direct personal and corporate income taxes in overall tax revenue. The first such agreements were the 1941 Wartime Tax Agreements (also known as “tax rental” agreements) that were introduced during World War ii. Under these agreements, the federal government “rented” the provinces’ income taxes in exchange for fixed annual federal transfers. Along with the intergovernmental negotiations following the expiry of these agreements came the asymmetries that now characterise the Canadian fiscal landscape. While Quebec is by far not the only province that has decided to opt out of federal programs, its autonomy claims can most certainly be considered the main driver against centralizing trends in Canada. Emblematic of this is the

62

63 64 65 66

the Settlement of Ethnic Conflict (Philadelphia: University of Pennsylvania Press, 2010) 75–96, 82. See H.R. Ouimet, “Quebec and Canadian Fiscal Federalism: From Tremblay to Séguin and Beyond”, in Canadian Journal of Political Science, 47.01, 2014, pp. 47. See also D. Béland, A. Lecours, “Accommodation and the Politics of Fiscal Equalization in Multinational States: The Case of Canada”, in Nations and Nationalism, 20.2, 2014, pp. 337. Own emphasis. B. Alarie and R.M. Bird, “Canada”, in G. Bizioli and C. Sacchetto, Tax Aspects of Fiscal Federalism. A Comparative Analysis (Amsterdam: ibfd, 2011) 79–136, at 86. For an extensive review of the jurisprudence concerning taxation powers, see ibid, at 91. Palermo and Kössler, Comparative Federalism (2017).

292

Kress

establishment of the Royal Commission of Inquiry on Constitutional Problems (the “Tremblay Commission”) by the Quebec government in 1953 in response to the continuation of the tax rental agreements and other constitutional issues.67 In the years of the Quiet Revolution,68 Quebec demands were often accommodated, also through changes in intergovernmental fiscal agreements and bilateral treaties that granted Quebec asymmetrical powers over health insurance, education, welfare, pension plans and other federally administered programs, such as students loans and youth allowances.69 Subsequent governments, however, would actively seek to avoid awarding Quebec any more of a special status. Subsequent intergovernmental agreements would treat all provinces symmetrically and all would have the same possibility to opt out as Quebec. This trend culminated in the 1997 Calgary Declaration (signed by all provincial Prime Ministers, except Quebec’s), which explicitly states that “if any future constitutional amendment confers powers to one province, these powers must be available to all provinces”. It furthermore reaffirms the equal status of the provinces and emphasises the cooperative nature of the Canadian system, but also the flexibility it provides in order to function efficiently and effectively. Nevertheless, Quebec’s objection to federal encroachment into areas that Quebec argued were constitutionally exclusive provincial jurisdiction in the mid-20th-century, has lastingly shaped the Canadian federal system. Provinces can since opt out of such federal programs where they already have own provincial ones and, in turn, are compensated fiscally, e.g. through federal tax cuts or increased equalisation payments. In general, the lee-way that Canadian provinces enjoy when it comes to taxing allows provinces to adapt the system to specific local circumstances. Quebec, especially, has made use of this discretion and is one of three provinces to collect its own corporate income tax (Ontario and Alberta being the other two), is the only province to collect the federal sales tax, collects an own valueadded tax—the Quebec Sales Tax—and is as well as the only province to collect its own personal income tax.70

67 68

69 70

J. Webber, The Constitution of Canada: A Contextual Analysis (Oxford: Hart Publishing, 2015), at 36. Referring to a time of rapid and profound societal and political changes in Quebec in the 1960s. For more, see K. McRoberts, Social Change and Political Crisis (Toronto: McClelland and Stewart, 1988). Iacovino, “Partial Asymmetry” (2010), at 84. Bird and Vaillancourt, “Reconciling Diversity with Equality” (2007), at 72; And Alarie and Bird, “Canada”, at 107–124.

Accommodating Diversity While Guaranteeing Stability

293

Besides the possibility that the contractual approach to (fiscal) federalism provides Canadian provinces in their exercise of provincial autonomy, another mechanism of fiscal federalism seems to foster stability, especially with regard to the Province of Quebec: the equalisation mechanism. First established in 1957, it was incorporated into the Constitution Act 1982, where it states that “Parliament and the government of Canada are committed to the principle of making equalisation payments to ensure that provincial governments have sufficient revenues to provide reasonably comparable levels of public services at reasonably comparable levels of taxation”. Quebec, due to its large population, has continuously been the province to receive the largest sum in equalisation transfers from the federal government since the establishment of the mechanism.71 As Béland and Lecours argue, this has served as an important argument against Quebec secessionism.72 Although a consequence of the use of power constitutionally assigned to subnational entities, asymmetries in the Canadian financial system are mainly the result of federal-provincial negotiations and effective differentiation becomes a question of political strengths and factual guarantees rather than a special feature of constitutional fiscal autonomy. Quebec’s strong (linguistic and political) identity has led to highly differentiated financial arrangements as compensation for avoiding secession.73 The Canadian system that is characterised by cooperation and intergovernmental negotiations has enabled its provinces to adapt the financial system to provincial needs—a system, of which Quebec has notably made use. In the Canadian financial system, the final degree of asymmetry thus depends mostly on political negotiation and political will, rather than constitutionally enshrined de jure asymmetries. 4

Financial Asymmetries—Political Realities: Comparative Observations

The above case studies show very different approaches to the institutionalisation of asymmetries in fiscal arrangements for accommodating subnational diversity. Italy, Spain and the uk constitutionally establish differentiated treatment of certain subnational entities, and negotiate the details bilaterally. By contrast, the constitutions of Belgium and Canada treat their subnational 71 72 73

See, Bird and Vaillancourt, “Reconciling Diversity with Equality” (2007), at 66. Ibid, at 338. See R.A. Young, “The Political Economy of Secession: The Case of Québec”, Constitutional Political Economy, 5 (1994), at 221.

294

Kress

units rather symmetrically74 and the existing asymmetries in their financial systems rather stem from differential political interpretation and implementation of these constitutional provisions, as well as from the results of intergovernmental negotiations; specifically so in order to grant their subnational units the same legal point of departure. In other words, asymmetric financial policies based upon autonomous choices can be realised even within a symmetrical financial constitutional framework. Evidently, these different approaches do not directly determine the possible degree of financial asymmetry within the system. This degree still appears to depend mostly on the political realities, the “preconditions”, in which intergovernmental negotiations about financial relations take place. It highlights the important role bilateral relations play in systems in which subnational entities are granted substantial autonomy and how these can effectuate asymmetric arrangements. Taxing powers, tax administration, expenditure powers and equalisation mechanisms all play a role in the balancing of unity and autonomy. Differential taxing powers and tax administration can lead to asymmetries in tax rates based on the plurality of political preferences of the sub-national units and cause a differentiated fiscal burden on the citizens. Quebec’s personal income tax is a case in point; its impact is softened, however, by the reduction of the federal personal income tax. The protection of adequate financial endowment to fulfil subnational units’ expenditure responsibilities is technically also possible without vesting them with taxing powers. While more extensive subnational taxing powers could be an option for guaranteeing financial autonomy of asymmetric units, Spain and Italy have opted for a constitutional guarantee that links the subnational revenue at disposal to the fiscal capacity of the respective territory and ensure, at the same time, substantial expenditure autonomy for subnational responsibilities. This pattern, furthermore, grants a comparably higher degree of autonomy, without relinquishing the overall coordination of the tax and finance system of the state. At the same time, however, practically all national and subnational experiences implement equalisation mechanisms that aim at the reduction of fiscal imbalances between the systems. Such schemes are a common feature also in the context of asymmetric systems like those analysed in this chapter. A reduction of financial differentiation, however, is to a certain extent ­inconsistent with the rationale behind de jure asymmetries, so that applying solidarity schemes on a nation-wide basis needs to be considered carefully. As mentioned above, asymmetric experiences often provide for differentiated legal 74

Not considering Belgium’s German-speaking community.

Accommodating Diversity While Guaranteeing Stability

295

tools to reconcile the special autonomy with general equalisation schemes. The relationship between asymmetric autonomy and equalisation can be arranged in many different ways, from complete exemption from equalisation to complete inclusion in the general regime—which are often represented as the former posing the threat of being perceived as an unjust privilege and the latter as carrying the danger of homogenisation. Both extremes would upset the overall balance and at the same time challenge the asymmetric system of autonomy. The Italian special regions represent an example, in which alternative models of involvement in national equalisation have been developed, turning solidarity into an opportunity for increasing their responsibility and autonomy. By assuming more and more legislative and especially administrative competences at their own expense, they contribute differentially to the balancing of State accounts. Other experiences show that too much centrally mandated solidarity can cause tensions between the subnational units that feel disproportionately burdened by it. Strong sentiments of subnational distinctiveness from the rest of the country and ineligibility for the same asymmetric exemptions that other units enjoy can exacerbate these effects, as is the case for Catalonia. At the same time, equalisation can also serve to mitigate independentist claims, for instance in Canada, where Quebec as the largest receiver of equalisation transfers benefits immensely from such a set-up. What finally becomes apparent from all examined cases is that asymmetrical financial arrangements (providing for asymmetries de jure, or permitting them) indeed can play and have played important roles in the accommodation of subnational diversity and provide meaningful, if not forceful arguments in the political discourses surrounding both autonomy and unity. Especially in systems, where constitutions allow for more flexibility in the negotiation of fiscal arrangements, fiscal asymmetries appear to provide an effective resource both for subnational units with strong own identities and national governments for reaching political compromises that on the one hand cater to subnational claims of autonomy while on the other hand stabilise federal unity. At least for the time being.

Part 3 New Perspectives on Fiscal Federalism

∵

chapter 12

Local Governments in African Federal and Devolved Systems of Government: The Struggle for a Balance between Financial and Fiscal Autonomy and Discipline Nico Steytler and Zemelak Ayitenew Ayele 1

Introduction

Across the federations of the world, local government is increasingly forming a constituent component of the constitutional framework of governments.1 Starting with Germany after World War ii, local government, and often its powers, were entrenched in constitutions. Following on from the German Basic Law (1949), local government features in the constitutions of Spain (1978), Brazil (1988), India (1992), Mexico (1999) and Switzerland (1999). Although constitutionally entrenched in one form or another, local governments were not elevated to the same status as subnational governments, but were at best “half” or “junior” partners in multilevel government systems.2 A very strong indicator of their status as an order of government has been their fiscal and financial autonomy. The status and place of local government are described, to varying degrees, in the constitutional architecture of some African countries with a federal or devolved system of government, including South Africa, Ethiopia, Nigeria and Kenya. In South Africa, local government was firmly entrenched in the 1996 Constitution as a “sphere of government” alongside the national and provincial governments. The Nigerian Constitution of 1999 also recognises local government, but places it firmly under the control of the subnational governments. In Ethiopia, there is no explicit mention of local government in its federal Constitution of 1995. The Kenyan Constitution of 2010 says nothing about a third (local) level of government; instead, there are only two levels of government—national and county—with no democratically elected local 1 N. Steytler, “Comparative Conclusion”, in N. Steytler (ed.), Local Government and Metropolitan Regions in Federal Systems (Montreal: McGill–Queen’s University Press, 2009) 393–436. 2 Ibid, at 435.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_014

300

Steytler and Ayele

g­ overnment required below the counties. As the latter largely perform local government functions, they are therefore included in this analysis. Depending on their level of constitutional recognition, local governments have varying degrees of responsibility for the provision of public services. The South African municipalities and the Kenyan counties are responsible for at least a quarter of all government expenditure, while in Nigeria and Ethiopia local government’s contribution is between 5 and 8 per cent. The key determinants of their role and how they perform it are their access to resources and their spending discretion. The necessary skills may be lacking, however, and corruption indices show that this is a significant problem in all four countries, requiring both national and subnational supervision over the lowest level of government. The issues that this chapter addresses are: (1) how constitutional recognition of local government (or in the case of Ethiopia, the lack of constitutional recognition) has shaped its political autonomy; (2) how fiscal and financial autonomy are measured with respect to revenue and expenditure (which underpins political autonomy); and (3) how a balance is struck between the measure of autonomy and financial and fiscal discipline exercised by “senior” partners in government. Overall, this chapter examines whether any patterns emerge from the four African countries. By financial autonomy, we understand the power of local authorities to raise their own revenue and spend it at their own discretion. Fiscal discipline, on the other hand, is the imposition of federal/national or subnational rules that structure both revenue-raising and spending decisions, as well as the measures available to ensure that such rules are complied with. 2

South Africa

With 54 million inhabitants as of 2014, South Africa is more than half (52 per cent) urbanised. It is regarded as a middle-income country with a gdp of usd 12,900 per capita in 2014 (gdp of usd 350 billion), but it has one of the highest levels of income disparity (its income Gini coefficient is 0.69),3 with nearly half of the population living in poverty spread across urban and rural areas. It has a high unemployment rate of 25 per cent, with the public administration being the largest employer: all public services and parastatals combined p ­ rovide

3 oecd Economics Department, “oecd Economic Surveys: South Africa 2015” (2015), http:// www.treasury.gov.za/publications/other/OECD%20Economic%20Surveys%20South%20 Africa%202015.pdf (accessed 11 December 2015), at 7.

Local Governments in African Federal and Devolved Systems

301

employment for more than 1.5 million people.4 Since the dawn of democracy in 1994, the country has been governed by the African National Congress (anc), which has dominated political life, controlling all but one province and one metropolitan council until 2016; in the local elections of that year, the anc lost three major metropolitan councils, including Johannesburg and Pretoria, to opposition parties. In so far as corruption perceptions are concerned, South Africa ranks 67th out of 175 countries with a score of 44/100, with a 0 indicating very corrupt and 100 very clean. When it comes to budget transparency, however, it scores 92 out of 100, indicating the provision of extensive information.5 2.1 Local Government in South Africa’s System of Devolved Government The 1993 interim Constitution, which entered into force after the country’s first democratic elections in April 1994, introduced constitutional supremacy and limited government after centuries of racial oppression. The Constitution also entrenched devolution in the form of provinces and local government—there are nine provinces and, since 2016 there have been 275 municipalities, which include eight large metropolitan municipalities and numerous small rural local ones. The new constitutional dispensation radically changed the status of local government. Before 1994, municipal lawmaking powers were delegated powers, derived from provincial statutes. As a result, any municipal by-law was merely an administrative act and thus subject to administrative-law review. Since 1994, the powers of the three levels of government have been derived from the Constitution: a municipal council was “no longer a public body exercising delegated powers, but was a deliberative legislative assembly with legislative and executive powers recognised in the Constitution itself”.6 Consequently, because the adoption of the budget is deemed a legislative act, it can no longer be challenged on administrative-law grounds. Section 156.2 of the Constitution provides that a municipality has the power to pass by-laws on matters listed in Schedules 4B and 5B, which include electricity distribution, water, sanitation, waste removal, roads and public transport, and municipal health services. However, both national and provincial governments also have powers over the same schedules but in a restricted 4 National Treasury, “Annexure 1—Summary of the Provincial Budgets and Expenditure Review (pber)” (2015), http://www.treasury.gov.za/publications/igfr/2015/prov/PBER%20Sum mary.pdf (accessed 11 December 2015), at 13. 5 Transparency International, “Corruption by Country/Territory: South Africa”, https://www .transparency.org/country/#ZAF (accessed 11 December 2015). 6 Fedsure Life Assurance Ltd and Others v Greater Johannesburg Transitional Metropolitan Council and Others 1998 (12) bclr 1458 (cc), at para 44.

302

Steytler and Ayele

manner:7 they only have regulatory powers that are restricted to framework legislation, setting norms and standards, and not determining particular outcomes (s. 155.7 of the Constitution). 2.2 Local Government Revenue In comparison with provinces, which have very limited constitutionally entrenched taxing powers and are reliant on national transfers for 97 per cent of their revenue, municipalities raise, on average, 73 per cent of their revenue (with metropolitan municipalities, hereinafter metros, raising 83 per cent).8 The principal reason for this disparity is the constitutionally entrenched municipal taxing powers. However, the implicit finance model for municipalities is that if their own revenue raising is insufficient, there is a system for equalisation through transfers from the revenue raised nationally. The main tax source is a property tax (called property rates), followed by a surcharge on fees for services rendered. Although property rates are an original tax, the national government may regulate its imposition. Thus, the Municipal Property Rates Act of 2004 provides an elaborate framework for the method of value assessment, the classification of property, the compilation of a property valuation roll and the processes by which the levying and collecting of the rates must comply. The minister of finance may also prescribe what the ratio between the tax rate on various types of property should be. However, the minister may not determine the actual rate to be applied (which is the municipalities’ prerogative), although a maximum annual percentage increase may be imposed. On average, the property taxes contribute 15 per cent of local governments’ own revenue collected, although metros collect much more than rural municipalities, which cannot impose tax on communal land. A similar legal framework is applicable for the imposition of surcharges on service fees. The bulk of local governments’ own revenue comes from service fees (mainly electricity, water, sanitation and waste removal) and surcharges on such fees (41 per cent of total municipal revenues).9 The Municipal Systems Act of 2000 provides a framework for the charging of service fees, including making the provision of free basic services for indigent persons compulsory. The rate of increases on electricity tariffs is also controlled by an independent national regulator. 7 See, generally, N. Steytler and J. Visser, “Chapter 5—The Powers of Local Government”, in Local Government Law of South Africa (Durban: LexisNexis, 2014) 5-16–5-29. 8 National Treasury, Budget Review 2014 (Pretoria: National Treasury, 2014), 93. 9 B. Khumalo, G. Dawood and J. Mahabir, “South Africa’s Intergovernmental Fiscal Relations System”, in N. Steytler and Y.P. Ghai (eds.), Kenyan–South African Dialogue on Devolution (Cape Town: Juta, 2015) 201–227.

Local Governments in African Federal and Devolved Systems

303

These revenue sources have produced variable results: some metros and large cities are 90 per cent self-sufficient, while in the case of rural municipalities, where property taxes are not lucrative and few trading services are provided, this drops to as low as 10 per cent. The power of a municipality to borrow money is protected in the Constitution; however, it may do so only for capital expenditure projects or for bridging finance purposes repayable within a year. Furthermore, this power may be exercised only in accordance with national legislation (s. 230A, Constitution). Such legislation in the form of the Municipal Finance Management Act of 2003 regulates this power in detail. Only large metropolitan councils have been able to successfully float bonds for capital projects. Given the great disparities in the ability of municipalities to be financially self-reliant, a system of fiscal equalisation lies at the heart of the system of intergovernmental fiscal relations. The provinces and local governments are entitled to “an equitable share of revenue raised nationally” to enable them to provide basic services (s. 227.1 of the Constitution). As the national government has all the main taxes (except property rates), it collects the bulk of revenue, which then must be divided vertically among the three spheres of government and horizontally among the provinces and municipalities. The Finance and Fiscal Commission, an independent constitutional body, advises Parliament on both divisions of revenue. To ensure equalisation between municipalities, a formula-driven distribution is used, which takes into account their fiscal capacity and effort.10 In addition to the untied equitable share transfer, the national government may provide further conditional grants to municipalities. Of all the transfers to local government in 2014–2015, 40 per cent were conditional grants mainly for infrastructural expenditure and capacity-building. A more recent phenomenon has been the so-called “indirect” conditional grants, where the national government spends such funds on behalf of a struggling municipality. 2.3 Local Government Expenditure Provinces and municipalities were responsible in the 2014–2015 fiscal year for over 60 per cent of total government expenditures: the national government was responsible for 38 per cent, provincial governments for 36 per cent and local governments for 26 per cent.11 If the national government’s debt servicing costs and contingency reserve are added, subnational governments are still responsible for more than half of government expenditure (national government 43.3 per cent, provinces 32.6 per cent and local government 10 11

Ibid., at 204. Calculations based on National Treasury, Budget Review 2014, supra, at 93 and Table 7.1.

304

Steytler and Ayele

24.4 per cent).12 Of the expenditures by local government, the combined budgets of the eight metros account for 67 per cent of all local government budgets. Given that municipalities (and metros in particular) collect the lion’s share of their revenue, they have far greater spending autonomy than provinces. Although the latter have much larger budgets, they are constrained by nationally bargained salary levels for staff in the education and health sectors, which comprise 87 per cent of their workforce. Nevertheless, expenditures by all municipalities are subject to stringent conditions to ensure financial discipline. However, the framework does not entail obtaining prior provincial or national permission for spending decisions, but reckless behaviour can prompt a provincial intervention. Municipalities must have balanced budgets, as they cannot borrow money for recurrent expenditure. Although a municipal council must submit its draft budget to the provincial and national treasuries for comment, no approval is sought or required. However, if no budget has been approved by the beginning of the financial year, the provincial government must intervene and take corrective steps, which may include the dismissal of the council, the appointment of an administrator until a new council has been elected and the adoption of a temporary budget (s. 139.4 of the Constitution). Spending autonomy flows from a municipality’s own revenue and its untied equitable share of revenue raised nationally. No external approval of spending decisions is required, but there has been a slow erosion of local spending autonomy through conditional grants in pursuit of national priorities. That some form of discipline is necessary has become increasingly apparent. The minister responsible for local government declared in 2014 that a third of municipalities were carrying out their tasks adequately, a third were just managing, and the last third were “frankly dysfunctional” because of poor governance, inadequate financial management and poor accountability mechanisms.13 This assessment is borne out by the auditor-general’s report on the 2013–2014 fiscal year, which showed that only 58 per cent of the municipalities received an unqualified audit (which was a huge improvement on previous years); the rest were either qualified (22 per cent) or received a disclaimer 12 13

Ibid. Ministry of Cooperative Government and Traditional Affairs, “The Back to Basics Approach Concept Document” (2014), www.cogta.gov.za/sites/cogtapub/B2BDOCS/ The%20Back%20to%20Basics%20Approach%20Concept%20Document.pdf (accessed 11 December 2015).

Local Governments in African Federal and Devolved Systems

305

(16 per cent).14 There was also a direct link between the poor financial systems of the provinces overseeing municipalities and the latter’s poor audit outcomes. The audit reports show high levels of unauthorised, irregular and wasteful expenditure. A further worrying trend has been the increased expenditure on personnel. The salary bill for political office holders and senior managers of the largest 27 cities grew over the period 2010–2014 4 per cent faster than municipal revenue did.15 Corruption is also reported as being rampant in many municipalities. Corruption Watch, an ngo, reported that their own statistics have shown that the highest incidence of graft happens at the local government level, accounting for 22 per cent of reports received from the public in 2013.16 To deal with a lack of discipline, provincial and national governments have a number of supervisory powers. Both the national and provincial treasuries have extensive financial monitoring systems. The provincial government must intervene “when as a result of a financial crisis” a municipality cannot comply with its service delivery obligations or meet its financial commitments (s. 139.5 of the Constitution). Such a measure must include a financial recovery plan and may also entail the dismissal of the council. Should the province fail to perform this function, the national government may do so instead. Despite clear incidences of “financial crises” in a number of municipalities, this specific legislation has not been used. Provinces did, however, intervene under a discretionary power in 67 municipalities between 2007 and 2014. A further disciplinary tool in the hands of the National Treasury is the cutting off of equitable share transfers due to serious or persistent material breaches of financial regulations (s. 216.2 of the Constitution). In 2015, the National Treasury did this to 58 municipalities that were in arrears with their payments to the national electricity utility and water boards, until they made suitable arrangements for the repayment of their debt. 14

15

16

Auditor-General South Africa, “Media Release: Auditor-General Reports a Noticeable, Improving Trend in Local Government Audit Results” (2015), https://www.agsa.co.za/ Portals/0/Media%20release/2015%20media%20release/MFMA%202013-14%20 Media%20release.pdf (accessed 11 December 2015). D. Powell, M. O’Donovan and T.C. Chigwata, “Capable Cities Index: Performance”, Dullah Omar Institute, (2015), dullahomarinstitute.org.za/multilevel-govt/mlgi/the-capable-cit ies-index-performance/at_download/file (accessed 11 December 2015). Corruption Watch, “Corrupion News: Local Government is the Weakest Link” (2013), http:// www.corruptionwatch.org.za/local-government-the-weakest-link/ (accessed 11 December 2015).

306

Steytler and Ayele

2.4 Conclusion The constitutional entrenchment of local government powers and revenue sources has provided the space for relative autonomy. It has enabled capable metropolitan and city governments to exploit this space and through their financial self-reliance, it could provide pockets of innovative and effective governance. The same cannot be said of many rural municipalities, the bottom third of which are dysfunctional and have become financially dependent on transfers. Local government is, however, treated as a monolith that has to operate within the same national legal framework and that aims to curb poor financial discipline among municipalities regardless of their capability or past conduct. The result has been overregulation, “strangulating” the capable by depriving them of innovative space and subjecting them to an onerous compliance burden.17 For the less capable municipalities, it has often meant stumbling over the numerous legal tripwires in their quest to impose and collect their own revenue from recalcitrant ratepayers.18 Whether extensive regulation actually achieves better governance is open to question. In many cases, it has led to greater lawlessness. It is evident that financial discipline is necessary. There is a clear tendency for municipal institutions to become vehicles for the benefit of incumbents rather than serving the broader community, as both significant expenditures for personnel costs and rampant corruption suggest. There is thus a difficult balance to be struck between the value of local autonomy and the need for national/provincial supervision. The law cannot, however, achieve this alone. The answer lies in a political culture at all three levels of government that seeks accountable and prudent financial management, a characteristic that is sadly missing in many parts of the South African body politic. 3

The Federal Republic of Nigeria

Nigeria, with over 178 million people, is the most populous country in Africa. Although more than half of its population is at present rural (53 per cent), the annual rate of urbanisation of 4 per cent is among the highest in the

17 18

N. Steytler, “The Strangulation of Local Government”, Journal of South African Law, 3 (2008) 518–535. J. de Visser and N. Steytler, “Confronting the State of Local Government: the 2013 Constitutional Court Decisions”, Constitutional Court Review (2016), 1–23, at 7.

Local Governments in African Federal and Devolved Systems

307

world.19 With a gdp of usd 568 billion, it has the largest economy on the continent, but is regarded as a “lower-middle-income” country with its usd 2,970 per capita gdp.20 Yet, due to uneven income distribution, over 46 per cent of its population lives below the national poverty line.21 A federation since independence in 1962, Nigeria has lurched back and forth between democracy and military rule ever since. After its return to civilian rule in 1999, President Buhari from the All Progressive Congress ousted the incumbent Goodluck Jonathan of the People’s Democratic Party in the 2015 presidential elections. Nigeria is also among the most corrupt countries in the world: it ranks 136th out of 176 countries and received a score of only 27 out of 100 in the 2014 Corruption Index.22 On budget transparency, its score is worse: at 18 out of a possible 100, there is scant or no budget openness.23 3.1 Local Government in the Nigerian Federation The Nigerian federation is composed of a federal government at the centre, 36 states and a Federal Capital Territory (s. 3.1and 2 of the Constitution of the Federal Republic of Nigeria, 1999). The Constitution also recognises democratically constituted local government made up of 768 local government areas and six area councils in the federal capital, Abuja (s. 7.1 of the Constitution). Local government is thus seen as an integral part of a “three-tier federation”.24 This view, however, belies the fact that under the Constitution the states retain broad authority to legislatively determine the “structure, composition, finance and functions” of local governments within their jurisdiction (s. 7.1 of the Constitution), a power often used to undercut the political and fiscal autonomy of the latter. Schedule 4 of the Constitution contains a list of local government functions that the states may assign to local governments, including cemeteries, abattoirs, parking, roads, streets, street lighting, public highways, parks, gardens, open spaces, sewerage and refuse disposal. These functions can therefore not be considered “original” constitutional powers, because they are assigned at the discretion of the states. For this reason, there is a significant variation in 19 20 21 22 23 24

The World Bank, “Data: Urban Population (% of Total)” (2015), http://data.worldbank.org/ indicator/SP.URB.TOTL.IN.ZS (accessed 10 December 2015). The World Bank, “Data: Nigeria” (2015), http://data.worldbank.org/country/nigeria (accessed 10 December 2015). Ibid. Transparency International, “Corruption Perception Index 2014: Results” (2014), https:// www.transparency.org/cpi2014/results (accessed on 09 December 2015). Ibid. See O.M. Ikeanyibe, “Three-tier Federative Structure and Local Government Autonomy in Brazil and Nigeria”, Mediterranean Journal of Social Sciences, 5 (2014) 559–569, at 559.

308

Steytler and Ayele

“the degree of autonomy afforded to local government within and across states”.25 Many states retain local government functions and revenue sources for themselves, thus seriously eroding the autonomy of local government. 3.2 Local Government Revenue Under the Constitution, local government has limited revenue-raising autonomy.26 The internal sources of local government revenue, also listed in Schedule 4, include marketing and trading-license fees, property taxes and rating, motor park duties, advertising fees and licensing fees for television and wireless radio. As with the case of their legislative powers, local governments cannot claim authority directly from the Constitution to raise revenue from these sources; instead, their taxing powers fall to the discretion of the states. The states seldom use their discretion to foster the revenue-raising autonomy of local governments. For instance, none of the states allow local governments to raise revenue from all those powers listed in the Constitution.27 However, most states do allow local government to raise revenue from property rates, while retaining the power to determine the rate at which properties are taxed. Moreover, the listed taxing powers are the least buoyant revenue sources, yielding minimal revenue. Even in the case of property rates, the Nigerian land tenure system of communal landholding under the custody of traditional leaders excludes such lands from property taxes.28 There are several user charges and service fees from which local governments could raise revenue. However, not only do these have little potential for generating revenue, local government also loses most of what it could collect due to the prevalence of corruption,29 as local officials collect revenue only to pocket the money.30 Moreover, the lucrative service fees derived from electricity and water supply fall outside the competences of local government.31 25

26

27 28 29

30 31

H. Galadima, “Federal Republic of Nigeria”, in N. Steytler (ed.) Local Government and Metropolitan Regions in Federal Systems (Montreal: McGill–Queen’s University Press, 2009) 234–266, at 247. S.T. Akindele, O.R. Olaopa and A.S. Obiyan, “Fiscal Federalism and Local Government Finance in Nigeria: An Examination of Revenue Rights and Fiscal Jurisdiction”, International Review of Administrative Sciences, 68 (2002) 557–577, at 565. S.I. Otinche, “Fiscal Policy and Local Government Administration in Nigeria”, African Research Review, 8 (2014) 118–137, at 130. Ibid. O.C. Olatunji, O. Asaolu Taiwo and J.O. Adewoye, “A Review of Revenue Generation in Nigeria Local Government: A Case Study of Ekiti State”, International Business Management, 3 (2009) 54–60, at 55. Ibid. Electricity generation and distribution are the exclusive domain of the federal government. See ss. 13–15 Constitution.

Local Governments in African Federal and Devolved Systems

309

It also seems that local governments make only a minimal effort to raise revenue from internal sources and depend mainly on federal and state transfers,32 with the result that the total amount of revenue that local governments raise internally forms only about 5 per cent of their total expenditure.33 The Investments and Securities Act of 2007, which regulates borrowing by the three tiers of government, as a rule allows a local government to borrow from internal sources by issuing securities in the form of registered bonds or promissory notes (s. 223). However, the Act imposes several substantive and procedural restrictions that make borrowing by local government practically impossible. Borrowing needs prior approval of at least four federal institutions,34 but such approval is predicated on the existence of state legislation authorising local government to borrow. The legislation must also guarantee, at the same time, that the state would service the debt, should the former default, by deducting it from the statutory allocation for the local government. According to Ekpo, at the end of December 2014, the 36 states together owed 21 per cent of Nigeria’s total public debt, which stood at usd 67.84 billion, while the federal government owed 79 per cent, leaving local government with 0 per cent debt.35 The three external sources of local government revenue are: federal grants, sharing a value added tax (vat) revenue and an entitlement to 10 per cent of the revenue a state collects from its internal sources. Local government is entitled, as are the federal and state governments, to a formula-based, statutory allocation from the revenue paid into the Federation Account, an account into which revenues collected by the federal government, after significant prior deductions, are paid (s. 162.3 of the Constitution). The portion of revenue that goes to local governments (the vertical division) varies from time to time and is within the discretion of the federal government. Presently, the local tier of government receives about 20 per cent of the revenue that is paid into the ‘Federation Account’, while the federal and state governments take 56 and 24 per cent, respectively.36 It should be noted, 32 33 34

35 36

Galadima, “Federal Republic of Nigeria”, supra, at 250. Ibid. These are “the Securities and Exchange Commission (sec), Ministry of Finance (Minister of Finance), Debt Management Office and where borrowing involves commercial banks […] the Central Bank of Nigeria”, see A.H. Ekpo, “Issues in Sub-national Borrowing in Nigeria”, Open Knowledge Repository Working Paper, (2015), https://openknowledge .worldbank.org/bitstream/handle/10986/21787/953450WP00PUBL0MENT0BORROWING 0final.txt?sequence=2 (accessed 11 December 2015). Ibid. A.C. Akujuru, “Revenue Allocation in Nigeria and the Dependency on Oil Revenue: The Need for Alternative Solutions”, Global Journal of Arts Humanities and Social Sciences, 3 (2015) 19–36, at 23.

310

Steytler and Ayele

however, that the federal government makes numerous deductions before the vertical split is made.37 Moreover, there are several other “off-budget accounts” that the federal government runs, including an “oil surplus account” and a “special debt account”.38 Thus, the federal government controls up to 75 per cent of total public expenditure. Of the remaining 25 per cent, local governments control about a quarter, making their overall public expenditure a mere 8 per cent of the total expenditure.39 With crude oil accounting for about 90 per cent of Nigeria’s export earnings and 70 per cent of total federal revenue,40 such revenue depends on the price of oil on the global market. According to Ahmed and Mottu, the whole scheme of revenue sharing is based on oil price assumptions that are embedded in the country’s budgets. Thus, changes in the oil price have an impact on the ‘Federation Account’ and hence on the revenue going to local government. As most of local revenue is indeed “oil-related”,41 the sharp fall in oil prices in 2014–2015 resulted in a 28 per cent drop in Nigeria’s revenue, dramatically impacting local government’s share of the ‘Federation Account’.42 The National Assembly determines the formula for the distribution of the statutory allocations among the local governments in each state based on the proposal of the president after considering the advice of the Revenue Mobilisation Allocation and Fiscal Commission (rmafc) (s. 162.2 of the Constitution). The distribution factors in the formula include population size and density, equality of the states, internal revenue, land mass and terrain.43 The share for local governments in the ‘Federation Account’ is not, however, directly transferred to local governments but is paid instead into the State Joint Local Government Account (sjlga), which the states create and control (s. 162.6 of the Constitution). This arrangement allows the states to reduce the amount local authorities receive by making numerous deductions, thereby enabling state governors to disburse the allocation “on a master-servant basis”.44

37 38 39 40

41

42 43 44

Ibid. Ibid. Ibid. O. Chima and J. Emejo, “Nigeria’s Revenue Shrinks by 28% amid Oil Price Slump”, This Day, (2015), http://www.thisdaylive.com/articles/nigeria-s-revenue-shrinks-by-28-amid -oil-price-slump/203084/ (accessed 06 December 2015). E. Ahmad and E. Mottu, “Oil Revenue Assignments: Country Experiences and Issues”, imf Working Paper, 2/203 (2002), www.imf.org/external/pubs/ft/wp/2002/wp02203.pdf (accessed 11 December 2015), at 18. Chima and Emejo, “Nigeria’s Revenue Shrinks”, supra. Galadima, “Federal Republic of Nigeria”, supra, at 251. Otinche, “Fiscal Policy and Local Government”, supra, at 125.

Local Governments in African Federal and Devolved Systems

311

vat is a revenue source under the exclusive authority of the federal government but is shared among the three levels of government according to a fixed percentage. Local government is entitled to 15 per cent of all receipts, which are distributed among all the local authorities based on equality (50 per cent), population (30 per cent) and derivation (20 per cent).45 The federal government also transfers conditional grants for the purpose of ensuring “that federal based policies like public health, primary education and public works are implemented at the local government level”.46 However, what each local government receives often depends on the support it gives to the ruling party at the federal level during elections, whether the president’s party controls the local government units and whether the president and the governor of the state within which a local government is located are from the same party and on good terms.47 The states are also required to transfer to their local authorities 10 per cent of the revenue they collect from their internal sources (s. 162.7 of the Constitution). The total transferable amount is distributed among the local authorities according to a formula determined by the state House of Assembly. Not only is this source of revenue insignificant, but most states simply refuse to transfer the revenue due.48 3.3 Local Government Expenditure Local government has insubstantial spending autonomy given that the states exercise extensive control over their expenditure both before and after local budget approval. Although the annual budget must be approved by the local council,49 it does not have the final say; instead, the budget is submitted to the state Bureau for Local Government for review, an approval process that is complex and lengthy, taking as long as three months, often undermining local governments’ ability to deliver services on a timely basis.50 In some states, the executive of a local authority is required to submit its draft budget to the 45 46

Galadima, “Federal Republic of Nigeria”, supra, at 252. F. Akpan and O. Ekanem, “The Politics of Local Government Autonomy in Nigeria: Reloaded”, European Scientific Journal, 9 (2013) 193–205, at 196. 47 Ibid. 48 J.O. Fatile and G.L. Ejalonibu, “Decentralization and Local Government Autonomy: Quest for Quality Service Delivery in Nigeria”, British Journal of Economics, Management & Trade, 10 (2015) 1–21, at 15. 49 C. Ugoh and W.I. Ukpere “Problems and Prospects of Budgeting and Budget Implementation in Local Government System in Nigeria”, African Journal of Business Management, 3 (2009) 836–846, at 840. 50 Adedokun, Local Government Tax Mobilization, http://visar.csustan.edu/aaba/Adedokun .pdf (accessed 05 December 2017), at 14.

312

Steytler and Ayele

Bureau even before the budget is submitted to, and approved by, the local council.51 This is done under the pretext of ensuring fiscal discipline by making sure that the local government “budget estimate meets with the laid down procedure and practice [and] that the budget is in conformity with the developmental plans of both the local and state governments”.52 There are also several institutional mechanisms through which the states impose fiscal discipline after local budgets have been approved. Much hinges on the fact that all federal and state transfers are paid into the sjlga, which allows the states to scrutinise and control local expenditure decisions, which significantly undermines the spending autonomy of local governments.53 There was an attempt to amend Sections 7 and 162 of the Constitution to abolish the sjlga and instead create a “Special Federal Account” that would allow local governments to access its federal transfers directly without involving the states. The states saw this as an attempt by the federal government to undermine their autonomy by directly interacting with local government, and they made sure that the amendment was not adopted.54 The states also exercise post-budget oversight measures through the Office of the State Auditor-General (osag), which has the power to exercise oversight over the management of local government finance by inspecting contracts awarded by a local government council.55 Furthermore, the office of the relevant state governor has the power to undertake a “periodic post-payment audit check […] on local government accounts”.56 Despite the extensive array of state oversight mechanisms over local finances, Nigerian local governments remain fraught with fiscal indiscipline. There is widespread corruption, often resulting in budget deficits. Corruption is so rife at the local level that Ugoh and Ukpere laconically observe that “a mere mention of the local government exudes corruption”.57 It has often been alleged that local accounting officers issue false certificates of job completion, inflate contract prices for procured items or make unauthorised variations of contract.58 They even “effect payment for services not rendered or poorly rendered and for items not purchased or not delivered”.59 The internal mechanisms for 51 52 53 54 55 56 57 58 59

Ugoh and Ukpere, “Problems and Prospects of Budgeting”, supra, at 840. Ibid. Ikeanyibe “Three-tier Federative Structure”, supra, at 567. Akpan and Ekanem, “The Politics of Local Government Autonomy”, supra, at 201. Otinche ,“Fiscal Policy and Local Government”, supra, at 120. Ibid, at 132. Ugoh and Ukpere, “Problems and Prospects of Budgeting”, supra, at 841. Otinche, “Fiscal Policy and Local Government”, supra, at 133. Ibid.

Local Governments in African Federal and Devolved Systems

313

ensuring fiscal discipline, such as oversight by the local council, are extremely weak in terms of tackling corruption, chiefly due to the political relations existing between members of local councils and the executive.60 Moreover, members of local councils have little incentive to restrict excessive expenditure by the executive since this is a mechanism for ensuring re-election. Internal auditors do not have the necessary skill or capacity to tackle corruption, and they themselves are often involved in it. The external mechanisms for spending oversight by the states are equally ineffective in curbing poor discipline because they are not free of corruption or other forms of fiscal indiscipline themselves.61 3.4 Concluding Remarks In terms of both Nigeria’s constitutional design and its political practice, the scale is tilted decidedly in favour of state oversight over local fiscal autonomy. The political autonomy that the Constitution envisages has been rendered hollow due to, among other reasons, excessive state fiscal control. To make matters worse, while denying local government the necessary flexibility to deliver on its mandates, the fiscal controls failed spectacularly to bring about the desired fiscal discipline, leaving local governments mired in fiscal indiscipline and corruption. The latter is, according to the Nigerian National Planning Commission, a systemic problem and “the major source of development failure” in the country.62 Without addressing the problem at all levels of government, excessive state control of local government finances is not likely to achieve discipline. 4

The Federal Democratic Republic of Ethiopia (fdre)

In 2014, Ethiopia had an estimated population of 96.6 million people, making it the second most populous country in Africa.63 With less than 20 per cent of its population urbanised, it has one of the lowest levels of urbanisation in the world.64 With a per capita gdp of usd 550 (a gdp of usd 54.8 billion), 60 61 62 63 64

Ibid. Ugoh and Ukpere, “Problems and Prospects of Budgeting”, supra, at 841. Quoted in F. Khan, “Understanding the Spread of Systemic Corruption in the Third World”, American Review of Political Economy, 6 (2008) 16–39, at 17. The World Bank, “Data: Ethiopia” (2015), http://data.worldbank.org/country/ethiopia (accessed on 11 December 2015). Ibid.

314

Steytler and Ayele

Ethiopia is considered a low-income country. About 29 per cent of the population lives under the national poverty line.65 Since the ousting of the Derg regime in 1991, a single political party, the Ethiopian People’s Revolutionary Democratic Front (eprdf), has been in power, controlling all levels of government. In the May 2015 general election, the eprdf claimed a 100 per cent electoral victory, leaving not even a single seat to any other political party at any level of government.66 With regard to the question of clean government, Ethiopia ranked 110th out of 176 countries in 2014 with a score of 33/100 on Transparency International’s Corruption Perception Index.67 4.1 Local Government in the Ethiopian Federation The Ethiopian federal system, formally adopted in 1995, is often referred to as an ethnic federal system since it is principally aimed at managing the ethnic diversity of the Ethiopian people. The federal system has a federal government at the centre, nine states organised along ethnic lines, two federal cities (Addis Ababa and Dire Dawa) and several ethnic-based sub-state units. The Ethiopian Constitution does not explicitly recognise local government as an order of government. It leaves all matters relating to the establishment of local government to the states, even though it implicitly requires each state to establish regular local government, and, depending on the ethnic diversity of a state, ethnic local governments.68 The latter are sub-state extensions of the ethnic federal configuration, allowing the territorial accommodation of intrastate ethnic minorities. The regular system of local government is established for the purpose of enhancing democratic participation and ensuring efficient service delivery, allowing each state to determine the tiers, numbers and size of local government units within its jurisdiction.69 In practice, there is a regular local government system, which is made up of some 800 rural woredas and 98 cities. Ethnic local governments, made up of liyu woredas (special districts) and nationality zones, have been established in five states.70 65 66 67 68

69 70

Ibid. National Electoral Board of Ethiopia (nebe), “Official Results of the 24th May 2015 Elections” (2015), http://www.electionethiopia.org/en/ (accessed 11 December 2015). Transparency International, “Corruption Perception Index 2014: Results” (2014), https:// www.transparency.org/cpi2014/results (accessed on 09 December 2015). Arts. 39(3) and 40(4), fdre Constitution. See further, Z. Ayele and Y. Fessha, “The Constitutional Status of Local Government in Federal Systems: The Case of Ethiopia”, Africa Today, 58 (2012) 89–109. See further, Ayele and Fessha, “The Constitutional Status of Local Government”, supra. Ibid.

Local Governments in African Federal and Devolved Systems

315

The fdre Constitution does not specify the functional competences of local government. In terms of the subnational constitutions liyu woredas and special zones have competences on cultural matters such as language. The subnational constitutions also authorise woredas, in general terms, to plan and implement developmental activities, without, however, clearly defining those developmental matters. In practice, woredas exercise functions relating to the delivery of basic “state services” such as education, water, agriculture and so on. Cities also exercise the above functional competences, and provide municipal services, such as sewerage, garbage collection and urban roads. 4.2 Local Government Revenue Since the fdre Constitution is silent on how local government is financed, it is determined by state legislation. A perusal of state constitutions and pieces of legislation shows that woredas have only delegated, as opposed to devolved, revenue-raising powers. Practice shows that the revenue they collect through these delegated powers is so negligible that they are almost entirely dependent on state transfers. Almost all state constitutions provide land-use fees, agricultural income taxes, and income taxes from employees of woredas as the main internal sources of revenue.71 These cannot be viewed as a woreda’s devolved internal sources of revenue since woredas are only authorised to collect the revenue from these sources on behalf of the states, which determine the rate of taxation. Moreover, as a rule, woredas are required to transfer to state treasuries the revenue they so collect. In practice, they retain the revenue, which is then “offset” against what would accrue to them in the form of state transfers.72 Some state proclamations authorise woredas to collect fees from users of local libraries, clinics, community halls, irrigation schemes and water wells, as well as fees for registrations of births, deaths, marriages and divorces.73 However, not much revenue can be collected from these sources since rural communities generally do not report births and deaths, and in most woredas one would be hard-pressed to find even so much as a single library.74 71 72 73 74

See Z. Ayele, Local Government in Ethiopia: Advancing Development and Accommodating Ethnic Minorities (Baden-Baden: Nomos Verlagsgesellschaft, 2014), 186–199. M. Garcia and A.S. Rajkumar Achieving Better Service Delivery through Decentralization in Ethiopia (Washington, dc: World Bank, 2008), 92. Art. 39, Tigray State (ts) Proclamation 99/ 2006; Benishangul-Gumuz State (bgs) Proclamation 86/2010. Z. Ayele, “Local Government and its Institutional Security within Ethiopia’s Federal System”, in A. Kefale and A. Fiseha (eds.), Federalism and Local Government in Ethiopia (Addis Ababa: Centre for Federal Studies, Addis Ababa University, 2015) 200–218, at 215.

316

Steytler and Ayele

Cities are authorised in state legislation to collect urban land-lease fees, land-use fees, municipal user charges, and fees for municipal services,75 which include: market fees; sanitation services; slaughterhouses; fire-brigade services; mortuary and burial services; registration of births, deaths and marriages; building plan approval; property registration and surveying; and the use of municipal equipment, transport or employees.76 Only the two federal cities (especially Addis Ababa, which covers 97 per cent of its expenditures from its own revenue) raise significant revenue from internal sources. However, even Addis Ababa’s revenue-raising autonomy seems to be guaranteed only so long as the eprdf is in control of the city; it might lose all its financial sources if an opposition party were to gain control of it. This was precisely what took place following the 2005 national elections, perhaps the most contested elections in Ethiopian history. The eprdf lost in Addis Ababa in a landslide, losing all but one seat in the city council to an opposition party. Being in control of the federal government, the ruling party immediately transferred the city’s most lucrative sources of revenue, in particular those related to public transport, to the federal government.77 Addis Ababa and Dire Dawa are authorised by the federal government to take loans from domestic sources.78 While the state constitutions are silent on borrowing by local government, some state proclamations provide that woredas and cities may borrow if so authorised by the relevant state government.79 Yet, the states themselves may borrow from internal sources if and when authorised to do so by the federal government (Art. 51.7 of the fdre Constitution). It is, therefore, unclear whether the states have the power to authorise local government to borrow, considering the fact that they could not borrow without the authorisation of the federal government. Even if the states could authorise local government to borrow, it is also unclear whether the authorisation of the state governments would suffice. Among others reasons, financial institutions are unwilling to extend loans to local government because of the legal uncertainty around lending to them.80 75 Ayele, Local Government in Ethiopia, supra. 76 S. Yilmaz and V. Venugopal, “Local Government Accountability and Discretion in Ethiopia”, International Studies Program, Working Paper 08-38, Andrew Young School of Policy Studies, (2008), at 21. 77 J. Abbink “Discomfiture of Democracy? The 2005 Election Crisis in Ethiopia and its Aftermath”, African Affairs, 105 (2006) 173–199, at 185. 78 Art. 54.1, fdre Proclamation 361/2003; art. 45.1, fdre Proclamation 416/2004. 79 See, for instance, art. 94, bgs Proclamation 86/2010; art. 42, ts Proclamation 99 /2006. 80 J. Werner and D. Nguyen-Thanh, “Municipal Infrastructure Delivery in Ethiopia: A Bottomless Pit or an Option to reach the Millenium Development Goals?” (2007), http:// www.ilpf.de/en/download/wp-01-2007.pdf (accessed 11 December 2015), at 29.

Local Governments in African Federal and Devolved Systems

317

Woredas receive unconditional block grants from the states. The decision to transfer such grants was more a policy direction rather than an explicit constitutional mandate. There are thus no hard and fast rules regarding what percentage of its budget a state is expected to transfer as block grants to the woredas within its jurisdiction. It is reported that every state sets aside a minimum of 50 per cent of its annual revenue for this purpose.81 The Southern Nations Nationalities and Peoples’ State (snnp), which reserves up to 80 per cent of its annual budget as local government block grants, makes the highest allocation.82 Woredas even take the lion’s share of the spending for some sectors, such as education and health, administering up to 80 per cent of the states’ recurrent spending for these sectors.83 The states have been testing various grant formulas for the purpose of horizontally dividing block grants among woredas with a view to ensuring equity in the distribution of revenue among the woredas. Since the policy decision to transfer a portion of state block grants to woredas was made in early 2000 by the federal government, it is an arrangement that could change depending on the party in power and its priorities. Moreover, the fact that woredas do not have a constitutional claim on state block grants means that a state government may refuse to transfer a block grant to a woreda under the control of a party other than the one in power in the state government. This is not merely a theoretical possibility, since this did actually transpire in the Sheko-Mejenger woreda of snnp. After the woreda came under the control of the Shako-Mejenger Democratic Unity Party (smdup), an ethnic-based opposition party, in the 2000 local elections, the Southern Ethiopia Democratic Movement (sepdm), the ruling party at the state level, ceased transferring block grants to that particular woreda, sending a clear message to the residents of the woreda that they had elected the “wrong” party.84 Block grants have also not kept pace with the growing expenditure needs of woredas, which spend over 90 per cent of their block grants on the salaries of their employees.85 They have very little left for capital investment. Local communities are thus often forced to contribute to cover the costs of building schools, roads, water tanks and the like.86 Cities are treated differently when it comes to block grants. In general, they do not get formula-based block grants, but rather grants are determined on an 81 82 83 84 85 86

Garcia and Rajkumar Achieving Better Service Delivery, supra, at 24. Ibid. Ibid, at 35. Yilmaz and Venugopal, “Local Government Accountability”, supra, at 8. Garcia and Rajkumar, Achieving Better Service Delivery, supra, at 36. Yilmaz and Venugopal, “Local Government Accountability”, supra, at 17–18.

318

Steytler and Ayele

ad hoc basis with a view to financing the recurring costs of their “state functions” such as health and education.87 They are also not allowed to use the grants to cover the financial costs of providing municipal services. They are expected to cover these costs from service fees and are required to administer and record revenue collected in the form of municipal service fees separately from their revenue for state functions.88 The federal government covers the costs of federal projects, such as foodsecurity and other safety-net programmes that states and local government execute.89 Woredas also receive from donors or the federal government, even if indirectly, conditional grants, also known as special-purpose grants (spgs), that are meant to finance such programmes.90 Addis Ababa and Dire Dawa receive further finances from the federal government for specific, nationally relevant projects.91 4.3 Expenditure The lack of revenue autonomy is equally evident on the expenditure side. Woredas and cities did not have the authority to approve their own budget until a policy was adopted by the federal government in 2000 to empower them to prepare and implement their own social and economic development plans consistent with the national and state development policies and plans.92 Along with this power came the power of woredas and cities to decide on their expenditure preferences as expressed in their plans and budgets. All state constitutions and relevant state and federal proclamations were amended to this effect. Now a woreda executive council or, in the case of cities, the mayoral committee prepares the budget, while the woreda or city council has the final say on the budget; the approval of the states is not, at least legally, required. The states do, however, have a variety of post-budget institutional mechanisms at their disposal for ensuring fiscal discipline. They, both in law and practice, exercise oversight over the financial management of local government; the state finance and economic development bureau is entrusted with the “primary responsibility” of monitoring woredas with respect to their financial

87 Garcia and Rajkumar, Achieving Better Service Delivery, supra, at 24. 88 Werner and Nguyen-Thanh, Municipal Infrastructure Delivery in Ethiopia, supra, at 16. 89 Garcia and Rajkumar, Achieving Better Service Delivery, supra, at 38. 90 Ibid, at 38. 91 Art. 46.1, fdre Proclamation 416/2004; art. 55.1, fdre Proclamation 321/2003. 92 Ayele Local Government in Ethiopia, supra, at 198.

Local Governments in African Federal and Devolved Systems

319

management.93 This office, which is in charge of distributing block grants to woredas, receives reports from woredas regarding their financial management and its officials’ conduct on sight inspection. The bureau may withhold block grants if it finds any financial mismanagement. In some states, a case in point being the snnp, the bureau enters into a “performance agreement” with woredas, which allows it to withhold block grants if the woreda mismanages the funds that have been transferred to it.94 There are constitutional and legal mechanisms that allow states to intervene in local governments if financial mismanagement is detected. For instance, proclamations from Tigray and Benishangul-Gumuz provide that the “misappropriation of funds or properties” is grounds for intervention.95 When “misappropriation of funds or properties” is detected, the state is authorised to suspend or dissolve the council of the improperly functioning woreda (Art. 74.4b of the Tigray State Constitution). The state executive council or chief administrator then has the authority to establish a transitional woreda administration and appoint provisional heads of executive bureaus until elections are held.96 The transitional administration assumes the executive powers of the woreda and also the power to adopt the woreda’s annual plan and budget (Art. 74.5a of the ts Constitution). There has been, however, no reported incident of state intervention in local government or the withholding by a state of funds. This is not, however, because no financial mismanagement takes place at the local level. The reason seems rather that the eprdf, which controls all local governments in the country, takes care of mismanagement through the party’s centralised structure and decision-making processes rather than through formal intergovernmental channels.97 4.4 Concluding Remarks Woredas and cities appear to enjoy a degree of spending autonomy even if they have restricted autonomy in terms of raising their own revenue. The initiative to financially empower woredas and cities, to the extent they are empowered 93 94 95 96 97

C. Heymans and M. Mussa, Intergovernmental Fiscal Reforms in Ethiopia: Trends and Issues (Washington dc: World Bank, 2004), 9. Garcia and Rajkumar, Achieving Better Service Delivery supra. Art. 58.1, ts Proclamation 99/2006; art. 103.1, Benishangul-Gumuz State (bgs) Proclamation 86/2010. Art. 103, bgs Proclamation 86/ 2010; art 74.4b, ts Constitution. For more on this, see A. Fiseha, Federalism and the Accommodation of Diversity in Ethiopia: A Comparative Study (Nijmegen: Wolf Legal Publishers, 2007).

320

Steytler and Ayele

to do so, did not, however, come from the states. Using the ambiguity of the federal Constitution concerning the place and status of local government, the states kept the latter under their political and financial thumb until the federal government came up with development policies that required the active involvement of local government for implementation. The eprdf was comfortable in devolving expenditure autonomy to local government, and faced no resistance from the states when doing so, only because it was in control of all levels and units of government. If any lesson is to be taken from the measures the party took with respect to Addis Ababa following the 2005 elections, it is that all this may change and local government may fall back under the complete financial control of the states (or the federal government in case of Addis Ababa and Dire Dawa) if an opposition party comes into power in any of its units. The political and financial position of local government therefore continues to be uncertain so long as the federal Constitution remains silent on the matter. 5

Kenya

In 2014, Kenya had an estimated population of 44.8 million people, 25 per cent of whom lived in urban areas but with a very high urbanisation rate of 4.34.98 With a per capita gdp of usd 1,290 (gdp of usd 60.94 billion), Kenya is considered a “lower-middle-income” country.99 However, due to skewed resource distribution, close to 46 per cent of its population lives under the national poverty line.100 Politically, the political dominance of the Kenya African National Union (kanu) since independence in 1963 has withered away, and the current president, Uhuru Kenyatta, won the presidency in 2013 by the narrowest of margins. He also won the first round of the 2017 presidential elections with small margin. However the Kenyan Supreme Court annulled the result of the election and ordered a re-run. Finally, with 98 per cent of the votes reportedly being cast in his favour, Kenyatta was declared a winner in the second round of the presidential elections, which Riala Odinga, Kenyatta's opponent, boycotted. Kenya is, like Nigeria, notorious for its high level of corruption and ranks 145th

98

The World Bank, “Data: Kenya” (2015), http://data.worldbank.org/country/kenya (accessed 11 December 2015). Also Central Intelligence Agency, “The World Factbook: Country Comparison to the World” (2015), https://www.cia.gov/library/publications/the -world-factbook/fields/2212.html (accessed 14 December 2015). 99 World Bank, “Data: Kenya”, supra. 100 Ibid.

Local Governments in African Federal and Devolved Systems

321

out of 175 countries, with a score of 25 out of 100, in the 2014 Transparency International’s Corruption Index.101 5.1 Counties in Kenya’s Bi-Level Government System Kenya’s 2010 Constitution introduced a devolved system of government with a national government at the centre and 47 counties as the second order of government. There is no third level of government, although the Constitution envisages that the counties may create sub-county administrative units in urban areas.102 Counties are thus envisaged to take up the functions of both a mesolevel government and a local government. Yet, when analysing their policy and taxing powers, they more closely resemble a local rather than a meso-level state structure. The list of competences in Part 2 of Schedule 4 of the Constitution contains both concurrent and exclusive powers.103 The concurrent powers­ include aspects of health services, agriculture, environment, energy, tourism, culture and sports. Exclusive competences include the implementation of some national policies (environment, energy), water and sanitation services, primary health care, electricity distribution, liquor licences and county planning, all of which are commonly viewed as belonging to local government.104 5.2 County Revenue The sources of county revenue are also typically those associated with local government. Counties have the constitutional authority to collect revenue by levying property rates, entertainment tax and charging fees for services (Art. 209.3 of the Constitution). These powers are to be exercised within a framework provided by national legislation. Revenue from property rates, the principal source of counties’ own revenue, is not sufficient to cover expenditures. Apart from the few cities, most land in Kenya is either not registered or is communally owned, making the levying of property rates very difficult.105 Even where land is registered, it is difficult to collect property rates; valuation rolls are not updated in time, and property owners often use inefficient service delivery as an excuse to withhold rates 101 Transparency International, “Nigeria’s Corruption Challenge”, Politics and Govern­ment, (2015), http://www.transparency.org/news/feature/nigerias_corruption_challenge (accessed 09 December 2015). 102 J. Cottrell Ghai, “Governing Metropolitan and Urban Areas in Kenya”, in Steytler and Ghai, Kenyan–South African Dialogue, supra, 278–301, at 285. 103 J. Mutakha Kangu, Constitutional Law of Kenya on Devolution (Nairobi: Strathmore University Press, 2015), 197–202. 104 Ibid, at 193. 105 N. Kirira, “Financing Counties in Kenya”, in Steytler and Ghai, Kenyan–South African Dialogue, supra, 227–250, at 231.

322

Steytler and Ayele

due.106 Entertainment taxes are unlikely to yield much since there is no clarity on what and how these taxes can be levied and collected. Although counties should be able to raise significant sums of revenue from the sale of electricity and water, these functions are being performed by national utilities. In terms of the Kenyan Constitution, counties are authorised to borrow funds provided that the national government guarantees the repayment of their debt (Art. 212 of the Constitution). This possible source of revenue is further strictly regulated by the Public Finance Management Act of 2012, which, among others things, provides that counties may borrow funds from both domestic and foreign sources (s. 142.2). This Act also provides that a county may raise short-term loans (repayable within a year) only for bridging purposes and long-term loans for infrastructure developments. The prescribed process for borrowing makes this source of revenue difficult to access. Section 58 of the Public Finance Management Act alone lists more than 10 conditions that have to be fulfilled before the national government can guarantee a county loan. In the first year since their establishment in March 2013, the counties made little effort to raise revenue from internal sources. In the 2013–2014 fiscal year, they achieved only 43 per cent of their target in terms of revenue collection. This was due, according to the World Bank, “to insufficient fiscal efforts by the counties”.107 They also collected less than the local authorities they replaced. This was no doubt due to establishment difficulties, but it may also have been encouraged by the entitlement to a share of the revenue raised nationally. The Constitution entitles counties to a minimum floor of 15 per cent of the revenue raised nationally (there is a county movement afoot, seeking to increase this to 45 per cent).108 The county share is then equitably divided among the 47 counties based on a formula recommended by the Commission on Revenue Allocation (cra), a constitutionally established independent advisory body, and then adopted by Parliament. The factors included in the formula are population (45 per cent), equal share (25 per cent), poverty gap (20 per cent), land mass (8 per cent) and fiscal effort (2 percent).109 The national government can also transfer additional grants—conditionally or unconditionally—to the counties. Practice shows that the counties are extremely dependent on national transfers. A World Bank report indicates that in the 2013–2014 fiscal year, the counties were responsible for 20 per cent of all public spending, but they ­covered 106 Ibid, at 231. 107 World Bank, Decision Time: Spend More or Spend Smart? Kenya Public Expenditure Review, vol 1 (Washington, dc: World Bank Group, 2014), 25. 108 See N. Steytler and Y.P. Ghai, “Devolution: What Can Kenya Learn from South Africa?” in Steytler and Ghai, Kenyan–South African Dialogue, supra, 442–483, at 457 note 1. 109 Kirira, “Financing Counties in Kenya”, surpa, at 235.

Local Governments in African Federal and Devolved Systems

323

less than 20 per cent of that from their own revenue.110 They have also made little effort to assert their autonomy by raising more revenue from their internal sources. 5.3 Expenditure As a fully fledged order of government, counties enjoy a considerable measure of spending autonomy. However, this autonomy is to be exercised with the disciplinary constraints of national legislation. For the constitutional and legislative framework, the Kenyan constitutional drafters borrowed liberally from the South African Constitution. The county assemblies have the freedom to adopt their own budgets (Art. 224 of the Constitution), but must do so within the regulatory framework of national legislation, the Public Finance Management Act. The budget must be balanced, as borrowing for a deficit on current expenditure is not permitted. The form of the budget is also prescribed by the Act. Some budget items may be shaped by external bodies. The Salaries and Remuneration Commission is entrusted with the responsibility of setting the remuneration of county assembly members and governors, and advising on the same for other county officials (Art. 230.4 of the Constitution). At the post-budget stage, financial discipline is exercised by a novel national institution, the Controller of the Budget (CoB): public funds are released only with the prior authorisation of the CoB after the latter has ascertained that the amount requested is in a county’s approved budget or is authorised by law.111 The CoB also makes quarterly budget execution assessments that it submits to county assemblies and governors so as to “help them make informed decisions on budget implementation”.112 The auditor-general is the other constitutional body with the responsibility of preparing annual audit reports on county governments for submission to both the county assemblies and the national Parliament (Art. 229 of the Constitution). The first year of county spending has not been a shining example of fiscal discipline. Profligate spending by county assemblies raised the hackles of governors, who refused to sign off on trips “to sports events in different parts of the country, or ‘benchmarking’ trips to overseas countries, including those without a devolved system of government”.113 The first year in office, counties also had difficulty spending available revenue: they closed their books for the 2013–2014 fiscal year having underspent by 38 per cent of their approved 110 111 112 113

World Bank, Decision Time, supra, at 27. Kirira, “Financing Counties in Kenya”, supra, at 243. Ibid. P. Wayande, “The Implementation of Kenya’s System of Devolved Government” in Steytler and Ghai, Kenyan-South African Dialogue, supra, 419–442, at 436.

324

Steytler and Ayele

budgets.114 Furthermore, counties tended to spend much of their revenue on administrative matters as opposed to development and service delivery. For instance, they spent a third of the overall county budget on the administrative expense of county assemblies and executives such as salaries and allowances.115 Only 22 per cent of their actual spending (38 per cent of the approved budget) went to developmental matters.116 A number of supervisory mechanisms are provided for in the Constitution. First, the Senate “exercises oversight over the national revenue allocated to county governments” (Art. 96.3 of the Constitution). It is not clear how such oversight can be effected, but Mutakha Kangu argues that it is as “soft” supervision, as no direct enforcement mechanisms are provided for.117 Clear enforcement measures are to be found in the stopping by the national government of transfers (including of the equitable share) to a county for the serious or persistent material breach of measures that ensure expenditure control and transparency (Art. 225 of the Constitution). The Constitution also envisages national legislation that enables the national government to interfere in the affairs of a county that does not operate a financial system in compliance with national legislation (Art. 190.3 of the Constitution). As is the case in Nigeria and South Africa, preventive and corrective measures may look good on paper, but if decisions are made on grounds of political expediency and corrupt motives, poor financial discipline will continue unabated. 5.4 Concluding Remarks In the absence of a democratically elected order of local government, counties perform typical local functions. Also being a meso-level government, their constitutional position is more secure than most local governments, even those in South Africa, although their revenue-raising powers are no more than those of South African municipalities. Very different from the self-sustaining South African city governments, counties rely on transfers that may on paper be untied, but in practice may be constrained in many respects by expenditure demands and national oversight, all at the expense of fiscal autonomy. Financial discipline is an important issue given the pervasive culture of corruption. It is also a vital requirement for devolution to be a vehicle for equitable development and not yet another carriage on the gravy train for local elites.

114 115 116 117

World Bank, Decision Time, supra, at 30. Ibid, at 31. Ibid, at 30. Mutakha Kangu, Constitutional Law of Kenya, supra, at 280.

Local Governments in African Federal and Devolved Systems

6

325

Some Comparative Remarks

The fact and manner of constitutional recognition of local government determine not only the level of political autonomy but also, and equally importantly, the measure of fiscal and financial autonomy, without which the former would be an empty shell. The full recognition of local government in South Africa and Kenya attest to this, while partial or no recognition in the case of Nigeria and Ethiopia, respectively, prove the opposite. The latter two countries provide yet another example of the perceived competition between states and local government, which is seen by states as a zero-sum game, with any local political and financial autonomy coming at the expense of the states. Where local government falls under the states’ control, the usual result is that local government is starved of both political and financial autonomy. Having constitutionally secured access to one’s own sources of revenue is a key ingredient of financial autonomy. However, because available sources are either limited or favour only urban governments, most local governments are reliant on national transfers within a system of equalisation. In all four countries, property taxes remain the mainstay of local taxing powers. Its value as a revenue source is almost entirely dependent on the location where it is levied. It is mainly urban governments that can reap its benefits, as in rural areas communal landholding, among other things, militates against this form of tax. South Africa seems to be the only country where municipalities are raising considerable revenue through their trading services in electricity, water, sanitation and refuse removal. These services in the other countries are provided by state enterprises that are owned and controlled by national governments, for instance, in Ethiopia and Kenya. Although local governments are theoretically able to borrow funds, the legal framework is often forbidding, and due to a lack of local governments’ own revenue, it is also only the main cities with access to their own revenue that can repay such loans. Even though it is constitutionally permissible for Nigerian local governments and Kenyan counties to borrow, this is rendered impractical through almost unattainable substantive and procedural legislative requirements. In the Ethiopian case, it is not even clear whether local government can legally borrow funds, except for the cities of Addis Ababa and Dire Dawa. This does not mean, however, that restricting local government borrowing is necessarily a bad idea. National governments have an interest in ensuring that local government functions within its means and does not upset national fiscal stability. Unrestrained local autonomy to borrow funds has often caused

326

Steytler and Ayele

economic crises when the national government is compelled to “bail out” errant subnational governments.118 The result of having limited access to own revenue is that most local governments are dependent on transfers from the central government, which inevitably constrains spending decisions, particularly when they come in the form of conditional grants. The case studies also show that whether local governments are constitutionally entitled to national transfers and whether they have direct access to them also have an impact on their financial autonomy. In South Africa and Kenya, municipalities and counties, respectively, are constitutionally entitled to unconditional transfers that are paid directly to them by the national treasury. In Nigeria, even though local governments receive transfers as a constitutional entitlement, the fact that such funds then fall under the control of the states undermines their financial autonomy. In Ethiopia, transferring block grants is merely a policy decision, which is dependent on the political interest of the political party in charge of the state government. Easy access to money through transfers also takes away the need for local governments to undertake the arduous task of raising their own revenue by which they can enhance their own political and financial autonomy. In this sense, they contribute to their dependence and subjection to senior levels of government. The case studies also show the dire state of many municipalities in respect of financial management, characterised by maladministration and corruption in spending decisions. There is also an emerging perception that local government is often geared to serve the benefit of incumbent politicians and officials rather than residents. The need for financial discipline imposed by “senior” levels of government is thus evident, which takes the form of legal regulation, monitoring, support and intervention. These disciplinary measures inevitably have an impact on the autonomy of local government. In the case of South Africa and Kenya, supervision lies primarily in the form of a legal framework with some corrective measures in case of non-compliance. While recognising the need for imposing strict controls, a proper balance between it and financial autonomy is crucial if innovative and accountable local government (as opposed to mere administrations) is to be fostered. In Nigeria, supervision also takes the form of prior approval of the budget, while in Kenya the Controller of Budget must sign off on expenditure decisions. Oversight by senior levels of government to ensure financial prudence and the effective use of public money is hardly likely to be effective if they themselves 118 World Bank, World Development Report 1999/2000: Entering the 21st Century (Oxford: Oxford University Press, 1999), 117.

Local Governments in African Federal and Devolved Systems

327

are in the grip of a culture of maladministration and corruption, a situation that is prevalent, to varying degrees, in the four countries studied. Although local government is portrayed as involving uniform institutions, they range from highly effective metropolitan governments to impoverished rural local authorities. It is likely that only the main cities will be able to exercise a measure of policy discretion, backed up by their own skills and capacity to raise most of their revenue. As they are also the focal point of economic development, political control over them is therefore highly prized. However, should they also become the home base of opposition parties, as has frequently happened, a dominant political culture of intolerance of political opponents may lead to the curtailment of city autonomy. When opposition parties captured Addis Ababa in Ethiopia, Kampala in Uganda and Harare in Zimbabwe, the governance of those cities was effectively nationalised.119 This again brings us back to the initial point: the key to enhancing effective and innovative local government is the need for constitutionally protected space allowing for political and financial autonomy. Such autonomy must, however, be balanced with the necessary fiscal and financial discipline to protect macroeconomic stability, as well as citizens from predatory local governments. Such discipline should not, however, render local authorities mere administrations on behalf of the centre, but should allow them to be local governments that are also accountable to their electorates for the collecting and spending of their revenue.

119 J. de Visser, “The Right to (Be) the City: The Place of Cities in the Institutional and Constitutional Context of Federal Decentralised Systems in Africa”, in S. Shastri (ed.), Equality and Unity in Diversity for Development: 5th International Conference on Federalism, 2010 (Addis Ababa: Shama Books, 2012) 22–29.

chapter 13

Fiscal Sovereignty in a Globalised World: The Pressure of European Economic Governance on Domestic Public Finance Jan Schnellenbach 1

Introduction

Economic governance in Europe has changed significantly since the financial and sovereign debt crisis. Before the crisis, the eu relied on a set of relatively simple numerical fiscal rules that were not, however, reliably enforced. The crisis has been widely interpreted as a failure of the original framework. Consequently, far-reaching reform packages have been implemented starting in 2011. These reform packages strengthen vertical mechanisms of control of fiscal policies and of their coordination. One could argue that this implies a loss of national-level fiscal sovereignty. However, proponents of these measures also argue that they are necessary in order to deal with the tensions that occur when a monetary union and decentralised fiscal policymaking co-exist. On the other hand, one could also argue that, instead of relying on political measures, market discipline would be a preferable option. Financial markets that correctly price country-specific sovereign risks send efficient price signals to political decision-makers. The price signals in turn serve as incentives. Ideally, national-level policymakers are guided towards fiscally sustainable choices through risk premiums determined by the markets. The core problem is that this framework, which preserves national-level fiscal sovereignty to the ­greatest extent, depends crucially on a credible no-bailout clause and, unfortunately, restoration of such credibility is difficult starting from the current status quo. These difficulties will not become smaller if the United Kingdom indeed leaves the European Union, as a slim majority of the electorate has demanded in the legally non-binding referendum of July 2016. So far, the country has been one of the less centralisation-friendly member states, and after a completed Brexit, bargaining power would likely shift towards countries that are more inclined to support bail-outs as well as eu-level interventions into national budgetary processes.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_015

Fiscal Sovereignty in a Globalised World

329

The discussion will proceed as follows: Section 2 discusses the rationale for having fiscal governance at the European level at all. Section 3 sums up how the Stability and Growth Pact has failed and how it was adjusted after the crisis. Section 4 discusses the effectiveness of these adjustments, as well as alternatives that are, in principle, still available. Finally, Section 5 offers some conclusions. 2

The Rationale for Fiscal Governance at the eu Level

The debate surrounding the introduction of a common framework for fiscal policy in the Eurozone has been present in academic and policy circles ever since the first steps were taken in negotiations towards a European Monetary Union (emu). As early as 1989, the Delors Report called for “macroeconomic policy coordination, including binding rules for budgetary policies”1 as a necessary part of the emu. Consequently, rules that were supposed to limit the scope for national deficit spending have been a part of the emu institutional framework ever since the determination of the so-called convergence criteria in the Maastricht Treaty,2 which originally served as benchmarks to specify the eligibility criteria for an eu member country to join the common currency. There are a number of economic arguments that lead to the presumption that there should be limits to nation-state fiscal sovereignty in a monetary union. One central issue is that membership in a monetary union implies that a country forfeits its ability to finance deficits by printing its own money.3 The situation is structurally similar to developing countries that issue debt in foreign currencies, with the implication that a higher risk of formal default is accepted in exchange for eliminating any ability of the bond-issuing country to devalue its currency on its own in order to ease the real burden of servicing its debt. As such, a decision in favour of monetary self-restraint does not necessarily require the imposition of formal limits on fiscal policy. Price ­signals ­supplied 1 J. Delors, Report on Economic and Monetary Union in the European Community (Brussels: ­European Commission, 1989), 16. 2 The criteria concerning fiscal policy are a public-deficit-to-GDP ratio of no more than 3% and a public-debt-to-GDP ratio of no more than 60%. In addition, a country’s long-term interest rate on 10-year government bonds is required not to exceed by more than 2 percentage points the average rate of the three member countries with the lowest inflation rates. 3 See O. Sievert, “Geld, das man nicht herstellen kann—ein ordnungspolitisches Plädoyer für die Europäische Währungsunion”, in P. Bofinger et al. (eds.), Währungsunion oder Währungschaos? (Wiesbaden: Gabler, 1996) 13–24.

330

Schnellenbach

by financial markets, such as interest rates increasing with the perceived risk of sovereign default, can be instrumental in guiding national policymakers. However, a problem that specifically occurs in monetary unions, and also in federal countries, is that a default or an acute risk of default in one member country or jurisdiction may have repercussions on other members of the union. 2.1 Fiscal Bailouts and Soft Budget Constraints The most prominent issue related to fiscal policy in a monetary union is the possibility that other members of the union will have to fiscally bail out an insolvent member country. The first possible direction of causation here is that markets may create the expectation that such a bailout will occur. As a result, market discipline imposed on highly indebted countries by means of interest-rate differentials and credit rationing will be weak.4 The relevance of this problem has been doubted in earlier contributions to the literature,5 since well-functioning credit markets should be capable of identifying ­country-idiosyncratic risks. This, however, presupposes the credibility of the announcement of a strict no-bailout policy within the monetary union. The less credible such an announcement is, the more credit markets will tend to evaluate not individual-country risks but the risks for the monetary union as a whole. Indeed, there was a dramatic convergence of yields on government bonds issued by emu countries in the pre-crisis period. There are different reasons that can explain this observation. It may be that markets expected the institutional framework of the emu to foster financial sustainability of individual member countries, leading to a significantly smaller perceived default risk for each of them. It may, however, also be that implicit expectations of bailouts were present. Interestingly, empirical evidence suggests that financial markets, even in the pre-crisis period of general convergence, still imposed risk premiums on countries that had higher deficits than Germany.6 But since these risk premiums were relatively small in absolute terms,7 they failed to enforce fiscal discipline in some EMU countries.

4 See W. Buiter et al., “Excessive Deficits: Sense and Nonsense in the Treaty of Maastricht”, Economic Policy, 8 (1993) 58–99, at 78–80. 5 Ibid. 6 See K. Reinboth et al., “Sovereign Risk Premiums in the European Government Bond Market”, Journal of International Money and Finance, 31 (2012) 975–995. 7 See European Economic Advisory Group, Report on the European Economy 2011 (Munich: ­CESifo, 2011), 72.

Fiscal Sovereignty in a Globalised World

331

Market expectations of fiscal bailouts in a monetary union establish incentives to increase national debt to inefficient levels because interest rates do not reflect the true country risks. If decision-makers in member countries themselves believe that they will be bailed out, then a second direction of causation comes into play, working through a soft budget constraint. The relevant incentive here is the expectation that some costs of domestic public goods and services can be externalised to taxpayers in those countries that will eventually finance the fiscal bailout.8 This is in line with empirical observations of an increased level of deficits after joining the emu, specifically for those countries that have traditionally shown a strong preference for deficit finance.9 Market and government expectations of fiscal bailouts thus established incentives to increase national debt levels, and implicitly also the risk of default at the national level. This problem may be exacerbated because the risk of any single country defaulting on its sovereign debt increases in a monetary union, as it becomes more vulnerable to speculative attacks when it is not allowed to print its own money.10 The blind spot of these arguments is that the lack of credibility of a no-bailout clause, as it has existed in the emu from its very beginning, needs to be substantiated. In other words, the incentives of countries that themselves have sustainable public finances to organise a bailout need to be identified. Spillover Effects of a High Deficit and Debt Levels within a Monetary Union A sovereign default within a monetary union can have immediate negative ­effects on other members. An example of such a negative spillover could be a general loss of trust in government.11 This might follow from a severe breach of contract by the government itself, whose institutional purpose is to legally enforce contracts. Costly and time-consuming bargaining over the details of default and the distribution of its burden can lead to uncertainty,12 possibly

2.2

8

9 10 11

12

In the political-economy context, the argument was originally made for federations. See T. Goodspeed, “Bailout in a Federation”, International Tax and Public Finance, 9 (2002) 409–421. See T. Baskaran and Z. Hessami, “Monetary Integration, Soft Budget Constraints and the emu Sovereign Debt Crisis”, Discussion Paper 2013-03, University of Konstanz (2013). See, for example, B. Eichengreen and C. Wyplosz, “The Stability Pact: More Than a Minor Nuisance?” Economic Policy, 13 (1998) 65–113. See W. Buiter, “The ‘Sense and Nonsense of Maastricht’ Revisited: What Have We Learnt About Stabilization in emu?” Journal of Common Market Studies, 44 (2006) 687–710, at 694–695. Ibid.

332

Schnellenbach

with negative macroeconomic side effects. If the perceived relevant polity is not the defaulting country, but the union as a whole, then negative spillovers of this type will occur. A more serious problem exists when a highly indebted country with a corresponding high risk of default threatens the financial stability of the monetary union as a whole. In this case, it can take other countries as financial hostages and use its own central role in order to extort bailout payments.13 This is not a merely theoretical possibility. The example of Greece shows how a country that is small in absolute and relative terms can nevertheless trigger bailout transfers because, at the time of acute crisis, the effects of an uncontrolled Greek default were uncertain. This concerned both the effects on European banks and the possibility of a political chain reaction being set in motion, where further countries ponder their exit from the emu. The example of Greece, as a small country, also shows that it is not even necessary to resort to “too big to fail”arguments to generate a willingness to bail out an ailing member country.14 Even with lesser levels of debt and deficits that do not immediately question sustainability, negative fiscal spillovers can occur. This argument rests on the so-called fiscal theory of the price level.15 This theory states that under some conditions, the price level may adjust in order to balance a government’s intertemporal budget constraint. In other words, the price level is influenced by the fiscal policy parameters chosen by governments. In this case, a reckless fiscal policy by one member state can have not only fiscal externalities but also monetary policy externalities. Again, the problem is rooted in an externality. A single country that chooses an expansionary fiscal policy takes into account the domestic costs of inflation or monetary adjustment but not those costs that occur in other member countries of the monetary union. The result is an inefficiently expansionary fiscal policy that may be overcome through institutions like an effective Stability and Growth Pact.16 There is even the danger that 13

14

15

16

See T. Mayer, “What More Do European Governments Need to Do to Save the Eurozone in the Medium Run?” in R. Baldwin et al. (eds.), Completing the Eurozone Rescue: What More Needs to Be Done? (London: cepr, 2010) 49–54. For this argument in the context of federations, see, originally, R. McKinnon, “MarketPreserving Federalism in the American Monetary Union”, in M. Bleijer and T. TerMenassian (eds.), Macroeconomic Dimensions of Public Finance (London: Routledge, 1997) 73–93. For the relationship between debt and inflation, see in particular J. Cochrane, “Long-Term Debt and Optimal Policy in the Fiscal Theory of the Price Level”, Econometrica, 69 (2001) 69–116. For one influential model of this type, see R. Beetsma and H. Uhlig, “An Analysis of the Stability and Growth Pact”, Economic Journal, 109 (1999) 546–571.

Fiscal Sovereignty in a Globalised World

333

the advantages of monetary commitment to low inflation levels are entirely negated by unconstrained, discretionary fiscal policy.17 Another rationale for limiting the fiscal scope of member countries is to protect the central bank from political pressures to provide an inflationary bailout, i.e., to reduce the real burden of debt by raising the price level.18 The assumption of pressures of this kind is not arbitrary. Even though the European Central Bank (ecb) is an entirely independent institution, its decision-making bodies are still influenced by the broader political situation and sentiments. Current events show how extraordinary circumstances lead to extraordinary measures taken by the ecb that would have been unthinkable before the recent crisis. Should there be pressures on the ecb to accept inflation in favour of relieving the burden of debt, it is likely that fears of contagion would once again play a major role.19 2.3 Institutional Consequences The problems discussed above seem to have straightforward institutional consequences: they appear to call for restraints on national fiscal policies in order to deal with fiscal and monetary policy externalities, avoid the necessity of fiscal bailouts and reduce inflationary pressures on the central bank. Such an institutional framework reduces the discretionary leeway of national policymakers, and it also reduces the leeway for existing preferences of citizens to be transformed into policies. In this sense, the sovereignty of national governments and also of citizens is prima facie impaired. However, this can be interpreted differently from a constitutional economic perspective.20 In economic terms, the central issue is the debate on having rules versus discretion,21 which originally dealt with problems of time inconsistency. These problems surface whenever a decision-maker is confronted with short-term incentives that lead them to deviate from behaviour that is optimal in the long 17

18 19 20 21

See A.K. Dixit and L. Lambertini, “Monetary-Fiscal Policy Interactions and Commitment versus Discretion in a Monetary Union”, European Economic Review, 45 (2001) 977–987. In a different theoretical framework, the externality problem emerges if the monetary authority cannot fully commit to pursuing a low-inflation policy. See V.V. Chari and P. ­Kehow, “On the Need for Fiscal Constraints in a Monetary Union”, Journal of Monetary Economics, 54 (2007) 2399–2408. See Eichengreen and Wyplosz, “The Stability Pact”, supra, at 71. Ibid., at 72. See V. Vanberg, “Market and State: The Perspective of Constitutional Political Economy”, Journal of Institutional Economics, 1 (2005) 23–49. See F.E. Kydland and E.C. Prescott, “Rules Rather than Discretion: The Inconsistency of Optimal Plans”, Journal of Political Economy, 85 (1977) 473–491.

334

Schnellenbach

term. Binding rules that constrain her short-term choices can help her to overcome this problem of time inconsistency. Similarly, appropriate rules that limit the scope of admissible fiscal policies in the short run can help reach a more favourable equilibrium in the face of the externality problems described above. If that is the case, a contractarian foundation for rules that limit ­short-term fiscal sovereignty can be constructed. They help citizens to pursue a common constitutional interest of enforcing policies that steer the economy towards a favourable, rather than unfavourable, equilibrium.22 Proper rules in this sense secure citizen sovereignty, somewhat paradoxically by limiting discretionary policymaking. The interesting question then is not whether there should be an institutional framework for national fiscal policies in a monetary union, but rather which kind of framework is effectively instrumental in ruling out an inefficient equilibrium. 3

The Failure and Adjustment of the Stability and Growth Pact

Assuming that the purpose of the Stability and Growth Pact (sgp) is to limit public deficits and the growth of public debt relative to gdp in emu member countries, the sgp can be unambiguously called a failure. As early as 2008, the ecb pointed out serious problems with the enforcement of the deficit rule in particular, with repeated and persistent breaches of the reference value at 3 per cent of gdp.23 In particular, the ecb was worried about lax enforcement of the formal rules of the sgp, and stated: “[I]t is the responsibility not only of the Member States, but also of the European Commission and the eu Council, to strictly apply the rules and procedures of the Pact and to exercise adequate pressure with regard to the pursuit of prudent fiscal policies”.24 Not only were there widespread excessive deficits and persistent debt-to-gdp ratios above the benchmark, but there was also a factual breach of the Maastricht Treaty’s no-bailout clause when the European Financial Stability Facility (efsf), the European Financial Stabilisation Mechanism (efsm) and the European Stability Mechanism (esm) were set up. One can thus legitimately argue that both pillars of the sgp—the credible commitment to shun bailouts in order to enforce market discipline on debtor countries and the fiscal rules designed in order to enforce additional political discipline—failed within less than 15 years after the Maastricht Treaty entered into force. Furthermore, the ecb 22

For a summary of the basic approach of constitutional economics, see G. Brennan and J.M. Buchanan, The Reason of Rules (Cambridge: Cambridge University Press, 1985). 23 See ecb, Monthly Bulletin October 2008 (Frankfurt: ecb, 2008), 53–66. 24 Ibid., at 65.

Fiscal Sovereignty in a Globalised World

335

has, albeit reluctantly at first, de facto assumed the lender-of-last-resort role for emu member countries with its Outright Monetary Transactions (omt) programme since 2012.25 The risk that a reckless fiscal policy eventually will lead to monetary repercussions became more salient with this step, and the discipline exerted by credit rationing in bond markets was further reduced.26 The provisions in the original sgp that were supposed to exert political control over the fiscal policies of member countries consisted of a corrective arm, i.e., the limits on deficit and debt ratios and the initiation of an excessive deficit procedure in case of non-compliance, and a preventive arm, i.e., the monitoring of fiscal policies in order to prevent excessive deficits from occurring at all. Both of these arms have been revised and adjusted, at first in 2005 and then in response to the recent crisis. The first revision was interpreted by most economists as a weakening of the original sgp.27 In the preventive arm, a medium-term objective (mto) was to be defined for every member country as its medium-term sustainable deficit level. Since the mto was defined in structural terms, there was some room for interpretation and discretion. In addition to this, time limits for the imposition of sanctions were prolonged in the corrective arm.28 As a whole, the 2005 reforms can be seen as an accommodation of rules to the reality that many emu countries, including Germany, failed to abide by the rules during the recession that had occurred beforehand. Rather than receiving a severe response to their excessive deficits from the European Commission (ec), emu member countries learned that the ec was more than reluctant to impose major sanctions. The subsequent European debt crisis can be seen as a result of this laxity in fiscal governance.29 25 26

27

28 29

See, for example, P. De Grauwe and Y. Ji, “How Much Fiscal Discipline in the European Monetary Union?” Journal of Macroeconomics, 39 (2014) Part B: 348–360. This danger is particularly salient among German policy-makers and Bundesbank officials. Jens Weidmann, the Bundesbank president, has voiced his opposition against omt on the grounds that it leads to close ties between monetary and fiscal policy, and thus endangers ecb independence. A group of German members of parliament has also ­attempted to stop omt by bringing the case to the Constitutional Court of Germany, arguing that omt exceeds the competencies granted to the ecb in eu contracts and is therefore unconstitutional under German law. The Court has, however, not followed this line of argument. See, for example, L. Calmfors, “The Roles of Fiscal Rules, Fiscal Councils, and Fiscal Union in European Integration” in H. Badinger and V. Nitsch (eds.), Handbook of the Economics of European Integration (London: Routledge, 2016) 157–169. Ibid. See M.S. Feldstein, “The Euro and European Economic Conditions”, nber Working Paper, 17617 (2011) and C. Keuschnigg, “Should Europe Become a Fiscal Union?” CESifo Forum, 11 (2012) 35–43.

336

Schnellenbach

In response to the events of the euro crisis, attempts have been made to reform the institutional structure of the emu in the exact opposite direction, with the aim of imposing stricter fiscal discipline among member countries. This is a logical step given that market discipline became less of a factor with the further erosion of any credibility of the no-bailout clause, both through fiscal transfers such as the direct bail-outs of fiscally troubled member countries and monetary interventions like the omt. Among the steps taken in the postcrisis reforms, a reversed qualified majority requirement appears to be particularly important.30 In the 2005 version of the sgp, an excessive deficit procedure could only be enforced if the steps proposed by the ec received a qualified majority in the EcoFin Council. Now, a qualified majority would be required to stop the ec from proceeding. In addition to this, the so-called Sixpack reform package of 2011 also introduced more precise requirements for convergence towards the 60 per cent debt limit for those countries that are not in compli­ ance with it. Similarly, the Sixpack includes quantitative benchmarks for a deviation from an mto to be considered as significant. Transcending fiscal policy in the narrower sense, there is also a macroeconomic imbalance procedure that was introduced with the Sixpack legislation. It is based primarily on a scoreboard containing 11 indicators that cover, for example, a country’s balance of payments, its unit labour costs, its export shares, private-sector debt, equity price levels and unemployment. The aim is to comprehensively understand a country’s macroeconomic situation, to give an early warning if risks are foreseeable and to supply the countries concerned with policy proposals to reduce existing macroeconomic imbalances. In addition to this preventive arm, there is also an excessive imbalance procedure that can lead to financial sanctions of up to 0.1 per cent of gdp if a country is diagnosed with excessive imbalances and fails to implement policies that reduce the imbalances. Somewhat paradoxically, the corrective arm of European fiscal governance may have become more effective with the Sixpack by allowing for more graduated sanctioning of countries that do not comply with the rules.31 Before the reforms, the severity of possible sanctions is one factor that led the ec and the EcoFin Council to only very reluctantly use this instrument. If, on the other hand, sanctions can start at relatively low levels and be gradually increased 30

31

For an overview of post-crisis reforms, see Calmfors, “The Roles of Fiscal Rules”, supra; L. Eyraud and T. Wu, “Playing by the Rules: Reforming Fiscal Governance in Europe”, imf Working Paper, 15/67 (2015); M. Andrle et al., “Reforming Fiscal Governance in the European Union”, imf Staff Discussion Note, 15/09 (2015). See Calmfors, “The Roles of Fiscal Rules”, supra.

Fiscal Sovereignty in a Globalised World

337

during the process, the likelihood that some sanctions are actually imposed increases. In addition to this, fines can also be imposed on a country that does not truthfully report its national-level data in reports submitted to the ec. Already in 2010 and leading up to the Sixpack, the preventive arm was strengthened by the introduction of the so-called European Semester. Its purpose was to align the timing of budgetary decision-making in eu member countries and to facilitate policy coordination both between the countries and between the ec and national decision-makers. In its currently applied version, the European Semester begins in November with the ec publishing reports that outline both the eu-level priorities for economic and fiscal policies, and the need to further investigate possible imbalances in specific countries. The ec provides recommendations and its opinion on the budget plans submitted earlier by member countries. This is followed by a period of bilateral meetings between the ec and member countries, as well as fact-finding missions by ec representatives. Also in this period, budget plans are adopted by member countries, ideally taking ec suggestions into consideration. ­Between February and April, country reports are issued by the ec and discussed with the member country governments, who in turn respond with policy agendas on economic and fiscal policies. In May, the ec draws up country-specific ­recommendations, which ought to be taken into consideration during the budget planning period for the next fiscal year. By 15 October, member governments should have their budgetary plans drafted,32 and the European Semester starts anew. The Fiscal Compact, which came into effect in 2013, is formally an intergovernmental contract and not a part of eu legislation. It reinforces the deficit rule—national deficits should be less than 3 per cent of gdp—and the medium-term budgetary objective concerning the structural deficit. Signatory countries committed themselves to implement the debt brake rule of the sixpack legislation (i.e., the rule for convergence of excessive debt levels to 60 per cent of gdp) into national law. Similarly, signatory countries need to implement automatic correction mechanisms for excessive deficits in national law. In order to provide surveillance for such corrections, independent fiscal councils are to be created in all signatory countries. In addition to this, the Fiscal Compact also contains several provisions for coordination of economic and fiscal policies between member countries and the ec.

32

In the so-called Two Pack legislation, also decided upon in 2011, the submission date for the national budget plans was set to 15 October each year. The Two Pack legislation also contains provisions for closer monitoring of countries with excessive deficits.

338

Schnellenbach

The post-crisis reforms outlined here have made the system of European fiscal governance significantly more complex than before.33 One major aim has been to take the broader macroeconomic framework into account more extensively, and at the same time to increase the likelihood of making the framework of rules more binding than before. Thus, the rules and the procedures for their application had to be made state-dependent, which contributed significantly to the increase in complexity. Overall, the different constraints specified in eu legislation now target six different fiscal aggregates:34 the level of nominal fiscal balance, net expenditure growth, the level of the structural fiscal balance, the change of the same balance, and the level and change of public debt. While formally retaining full national sovereignty over the budgetary process, the post-crisis reforms embed national-level decision-making of fiscal policies within a broad framework of supranational control and consultation that clearly implies a material loss of fiscal sovereignty. Given the problems of national-level fiscal policy in a monetary union that have already been discussed, this may be inevitable. However, the jury is still out as to when the effectiveness of the new system of fiscal governance is to be judged. This is due to practical reasons—there is simply too little experience with the new rules so far. Nevertheless, the main components of fiscal governance at the eu level can be evaluated on a theoretical level, and by referring to empirical evidence from other cases. 4

Institutional Approaches to European Fiscal Governance: An Economic Evaluation

4.1 On the Effectiveness of Fiscal Rules The fiscal rules that are at the core of the emu framework can be categorised as numerical fiscal rules. They define more or less unambiguous benchmarks that make it possible to distinguish legitimate from illegitimate national fiscal policies simply by looking at a few indicator values. An important point of criticism with regard to these fiscal rules is that there is a risk that important threats to fiscal discipline are overlooked. An example is Ireland, which allowed the accumulation of huge amounts of private debt before the financial crisis, and at 33

34

See Eyraud and Wu, “Playing by the Rules”, supra, at 14. It has also been argued in the literature that the complexity itself reduces the likelihood of success. See M. Dolls et al., Reconciling Insurance with Market Discipline: A Blueprint for a European Fiscal Union (Munich: CESifo, 2016). Ibid., at 15.

Fiscal Sovereignty in a Globalised World

339

the same time sharply reduced its public debt. The sudden transformation of private into public debt in an attempt to stabilise its banking system led to a public debt crisis in Ireland practically overnight.35 Numerical fiscal rules were in no sense instrumental in preventing or even predicting the crisis. In this respect, the emu’s current approach is helpful because it complements fiscal rules with a macroeconomic imbalance procedure and early-warning mechanisms that observe a broader set of indicators. It is important to note that fiscal rules and debt brakes are a widespread phenomenon. Even without the specific institutional setting of a monetary union, many countries worldwide have voluntarily implemented some kind of numerical fiscal rule.36 In many of these cases, the aim is to deal with purely domestic deficit biases and to control domestic, intergenerational externalities that occur when public consumption to the benefit of current generations is financed through public debt. However, there are general problems with fiscal rules as such that also apply to the specific European situation. For example, fiscal rules are seldom successful if they are not backed by a political will to implement sound fiscal policies, because they can either be modified by the current political majority at any point in time or can simply be ignored and transgressed.37 This is exactly what happened in the emu prior to and during the 2005 reform of the sgp, but it is also an experience that has occurred in other countries.38 A prominent example is Germany that had constitutionally limited deficit financing before the modern debt brake that is currently being phased in. However, under the old set of rules, a simple majority in parliament could declare the existence of a macroeconomic imbalance, which would then result in a suspension of this constitutional limit. Consequently, declarations of macroeconomic imbalances have occurred frequently in ­Germany, and debt levels have increased accordingly. The post-crisis reforms in the emu have also attempted to address this problem by embedding national policymaking in a more formally structured procedural framework, by increasing the likelihood of sanctions by allowing smaller sanctions to be imposed and by a reduction of countercyclical effects. Nevertheless, it remains to be seen whether the ec will be willing to impose 35

36 37 38

See C. Wyplosz, “Fiscal Rules: Theoretical Issues and Pracitical Experiences”, in A. Alesina­ and F. Giavazzi (eds.), Fiscal Policy after the Financial Crisis (Chicago: University of Chicago Press) 495–524, at 499. For a broad survey, see International Monetary Fund, Fiscal Rules—Anchoring Expectations for Sustainable Public Finances (Washington: imf, 2009). Ibid., at 15. See Wyplosz, “Fiscal Rules”, supra, at 505.

340

Schnellenbach

sanctions even in politically difficult environments, and whether there will be attempts in the EcoFin Council to build coalitions that can stop excessive deficit procedures with a qualified majority. The imposition of sanctions in case of fiscal deficits is still far from automatic or foolproof. Examples of successful fiscal rules that contain stronger automatic components can be found on the cantonal level in Switzerland.39 One example is the cantonal debt brake in St. Gallen, which requires automatic tax increases if deficits become too large and limits the scope of tax reductions to situations where clearly defined benchmarks for sound public finances are met. Another mechanism that may help fiscal rules to become effective is that they serve as a political benchmark for fiscal policy even if they are not always complied with.40 Empirical evidence for 11 eu countries with their own fiscal rules for the 1994–2012 period shows that these rules were not complied with in about 50 per cent of all instances in the sample. However, even after an instance of non-compliance, fiscal policy often tended to revert back towards the benchmark established by the rule.41 In this sense, there can be a stabilising medium- and long-term effect of rules even if short-term deviations are commonplace. A final interesting point is whether financial markets themselves value fiscal rules and reward countries implementing them with lower risk premiums on government bonds. Given the short period of time since their implementation, there is no evidence for the reformed emu set of fiscal rules so far, but some conclusions can be drawn from other international evidence. In general, the picture is mixed but points more in the direction that fiscal rules are rewarded by, albeit often very small, reductions of risk premiums.42 But exploiting the natural laboratory of 26 autonomous Swiss cantons, it is shown that strong fiscal rules appear to be evaluated as particularly credible by financial markets. A fiscal rule is strong if there are: (i) a clear numerical rule; (ii) a direct link between budget planning and budget execution and (iii) fully automatic sanctions in the form of automatic tax increases.43 Unfortunately, the European 39

40 41 42 43

See L.P. Feld and G. Kirchgässner, “On the Effectiveness of Debt Brakes: The Swiss Experience” in R. Neck and J.E. Sturm (eds.), Sustainability of Public Debt (Cambridge, ma: mit Press, 2008) 223–255. For this argument, see W.H. Reuter, “National Numerical Fiscal Rules: Not Complied With, But Still Effective?” European Journal of Political Economy, 39 (2015) 67–81. Ibid. For a survey, see L.P. Feld et al., “Sovereign Bond Market Reactions to Fiscal Rules and NoBailout-Clauses – The Swiss Experience”, zew Discussion Paper, 13-034 (2013). Ibid., at 11.

Fiscal Sovereignty in a Globalised World

341

fiscal rules are still weak with regard to the third requirement, which is good reason to question their credibility even after the post-crisis reforms. 4.2 Delegation of Fiscal Policy Competencies A different type of mechanism to deal with the fiscal common pool problem in a monetary union would be the delegation of political decision-making authority, which can come in different forms. An extreme form would be the creation of a fiscal policy body that is completely independent of political pressures, akin to an independent central bank.44 This would be theoretically conceivable if a clear-cut objective for this body, such as fiscal sustainability, were defined in a similar way as price stability is the single overarching goal of most independent central banks. However, contrary to monetary policy, all decisions on fiscal policy have immediate and visible redistributive effects that require democratic legitimacy.45 It is highly unlikely that citizens value fiscal discipline so dearly that they could consent to surrendering democratic control over redistributive fiscal policies. However, it is conceivable that only limited parts of the budgetary process can be delegated to bureaucratic, rather than political, organisations. For example, there is evidence that governments often use biased, overoptimistic tax projections to legitimise high levels of spending.46 This in turn leads to the proposal to delegate data collection and projection to independent agencies.47 An example of this is the British Office for Budget Responsibility, whose purpose is forecasting, as well as the evaluation of government policies in terms of long-term fiscal sustainability and compliance with fiscal rules. It is allowed to act only on the grounds of official government policy, i.e., it has no competence to evaluate alternative policies.48 Such a very limited definition of the competencies of an independent body is clearly compatible with democratic standards, but it is potentially capable of improving the quality of budgetary decision-making. The post-crisis emu reforms took up this thread by introducing the requirement for member countries to establish independent fiscal councils. These are 44 45 46 47 48

See Wyplosz, “Fiscal Rules”, supra, at 505. Ibid. See I. Bischoff and W. Gohout, “The Political Economy of Tax Projections”, International Tax and Public Finance, 17 (2010) 133–150. See, for example, L. Jonung and M. Larch, “Improving Fiscal Policy in the eu: The Case for Independent Forecasts”, Economic Policy, 21 (2006) 492–534. See S. Wren-Lewis, “Fiscal Councils: The uk Office for Budget Responsibility”, CESifo Dice Report, 3 (2011) 50–53.

342

Schnellenbach

to evaluate the consolidation progress of their respective country in general and to evaluate whether a given economic environment qualifies as an emergency situation. The national governments should adopt a “comply or explain” approach in interacting with their fiscal council, i.e., they should either follow the policy proposals given by the council or explain why they choose not to.49 This type of fiscal council implicitly reduces the leeway of national governments by making independent, evidence-based policy proposals and by shifting the burden of justification to the government whenever it wants to deviate from these proposals. Especially in countries where trust in the willingness of governments to pursue a sound fiscal policy is low, the independent opinion of a council of experts can either increase trust, if it concurs with the government’s plans, or provide a benchmark for a critical public debate. In this way, citizen sovereignty can be strengthened as the level of reliable information available to citizens increases.50 From the perspective of the government, the costs of pursuing a reckless fiscal policy increase, at least in terms of potentially lost reputation. In specialising on giving advice rather than making decisions, fiscal councils can also strengthen fiscal rules by acting as credible watchdogs vis-à-vis the government.51 In the past, there have nevertheless been incidents where the advice of fiscal councils has been ignored, or where fiscal councils that were critical of their governments have simply been abolished.52 With the ec supervising the independence of fiscal councils, however, at least such extreme measures have become unlikely. A more extreme approach to the delegation of fiscal authority within the eu is associated with proposals to establish a fiscal union to complement the monetary union. This step would require that the eu level be provided with its own tax source, and that decision-making authority over tax rates and spending be delegated to the ec and the European Parliament. The eu would essentially be transformed into a federal structure. Although there are federations where the no-bailout rule for lower-tier governments by the central government is highly credible, such as Switzerland53 and the United States of America, it is very likely in the current situation that expectations would arise 49 50 51 52 53

See H. Burret and J. Schnellenbach, Implementation of the Fiscal Compact in the Euro Area Member States (Wiesbaden: Council of Economic Advisors, 2014), 6. See also Calmfors, “The Roles of Fiscal Rules”, supra. See L. Calmfors and S. Wren-Lewis, “What Should Fiscal Councils Do?” Economic Policy, 26 (2011) 649–695. See more recently R. Beetsma and X. Debrun, “Fiscal Councils: Rationale and Effectiveness”, cepr Discussion Paper, 11140 (2016). See Feld et al., “Sovereign Bond Market Reactions”, supra.

Fiscal Sovereignty in a Globalised World

343

that eu-level revenue would be used primarily to finance transfers to member countries. This is particularly the case because internalising spillovers in the fiscal common pool would require that decision-making powers be delegated to alter national policies to the supranational level, while states’ own tax bases would not be necessary to accomplish this task. Furthermore, historical experience shows that federations exhibit a tendency towards increasing centralisation. The reason is that lower-level governments often face tighter fiscal constraints than higher-level governments. This makes them susceptible for political trades where sub-central governments transfer autonomy and responsibility for tax policies to the central level and receive fiscal transfers in return.54 In extreme cases such as Germany, the fiscal constitution can become entirely unbalanced, leaving public debt as the only flexible source of revenue to the state level.55 In any case, the moral hazard that arises with a central eu budget available for vertical transfers calls for more extensive and immediate institutional controls on national-level deficits, possibly even a centralisation of bond emission at the eu level.56 If the aim is to preserve fiscal sovereignty at the national level, it is difficult to see how the establishment of a fiscal union could be useful. Once the transfer of competencies to the central level starts, this tendency is difficult to control in the long run. 4.3 Fiscal Institutions One could argue that the numerical fiscal rules discussed above are at best a medicine that can control deficits as symptoms of a deeper underlying problem but not solve the problem itself. From the institutional point of view, the core of the problem is a common pool issue, but not the common pool at the emu level. Instead, a common pool within the national-level process of policymaking is scrutinised together with the institutions that can help solve this problem.57 Whenever a special-interest group lobbies for targeted spending, it expects that its own members’ taxes will finance only a very small share of required expenditures, while the bulk of the burden can be externalised to other taxpayers. When every interest group acts accordingly, and policymakers are 54

55 56 57

See T. Döring and J. Schnellenbach, “A Tale of Two Federalisms: Germany, the United States and the Ubiquity of Centralization“, Constitutional Political Economy, 22 (2011) 83–102. See H. Seitz, Subnational Government Bailouts in Germany (Bonn: zei, 1999). See Calmfors, “The Roles of Fiscal Rules”, supra. See M. Hallerberg et al., Fiscal Governance in Europe (Cambridge: Cambridge University Press, 2010).

344

Schnellenbach

responsive to the demands of these groups, spending, deficits and debt levels increase.58 When the burden can be shifted to future taxpayers instead of increasing somebody’s current tax burden, the effect is reinforced. If the deficit bias that is created by these incentives can be controlled at the domestic level, the problem of excessive debt at the emu level would also be controlled to a large extent. There are two general solutions to this problem. One is to negotiate target levels for policy variables such as aggregate spending and the deficit, and then to negotiate how the cake should be divided among competing interest groups. The other mechanism is to delegate ­decision-making powers to a strong minister of finance, who oversees the fiscal externalities implied by the demands presented by interest groups, and ensures that spending will not be excessive.59 Which type of institutional solution will be more effective depends on the type of government that can be expected. A strong finance minister is more effective with single-party governments or those with a small number of wellaligned coalition parties, where succumbing to a strong finance minister from a particular party is not a major issue. In more diverse multi-party governments, it is more effective to resort to the negotiation approach.60 In this framework, (numerical) fiscal rules can arise endogenously, e.g., to signify benchmarks for indicators that have been agreed upon in negotiations. An interesting implication of the focus on the institutions of the budgetary decision-making process is that it may not be necessary, or useful, to impose strict numerical rules on all member states of a federation or a monetary union. Instead, it may be sufficient to encourage the implementation of domestic fiscal institutions that steer the budgetary process towards sustainable fiscal policies. Other than that, discretionary fiscal sovereignty at the national level could be preserved to a large extent. The major caveat here, however, is that in order to keep union-level moral hazard in check, the credibility of the no-bailout clause would need to be established again. 4.4 Resurrecting the No-Bailout Rule As discussed above, the credibility of the no-bailout rule was severely diminished during the crisis. Even when policy proposals to establish a fiscal union 58

59

60

See A. Velasco, “A Model of Endogenous Fiscal Deficits and Delayed Fiscal Reform”, in J. Poterba et al. (eds.), Fiscal Institutions and Fiscal Performance (Chicago: University of Chicago Press, 1999) 37–57. See M. Hallerberg and M. von Hagen, “Electoral Institutions, Cabinet Negotiations and Budget Deficits within the European Union”, in J. Poterba and J. von Hagen (eds.), Fiscal Institutions and Fiscal Performance (Chicago: University of Chicago Press, 1999) 209–232. Ibid.

Fiscal Sovereignty in a Globalised World

345

in order to implement a formal European transfer system appear not to be politically viable at present, implicit expectations that member countries with unsustainable public debt will at least partially be bailed out are present. The core question is then whether credibility can be resurrected starting from the current, highly problematic status quo. Proposals to achieve this often, and prima facie paradoxically, start with a mechanism to deal with the high levels of legacy debt that are currently burdening emu member countries.61 The reason is that high debt levels reduce growth prospects and threaten to overburden the ecb, in particular with expectations to act as a lender of last resort and provide a monetary bailout should everything else fail. Furthermore, it has been argued that financial markets are currently in a binary “risk on/risk off” mode and are unable to correctly price marginal variations in sovereign risk.62 If all this is the case, then the slate needs to be wiped clean and a solution needs to be found for legacy debt in order to restore the proper functioning of markets and emu institutions. One proposal is to set up a stability fund that would buy back legacy debt and that could be financed out of fixed shares of the tax revenue of participating countries.63 The mechanism to cope with legacy debt should be complemented with a credible mechanism for lending in future crises that would replace the current fiscal lending facilities. A core element of such a mechanism would be provisions that make it politically expensive to use these crisis lending mechanisms in order to minimise incentives to purposefully steer a country into a situation where it would become eligible to receive crisis lending. Finally, a credible insolvency procedure would be necessary. The possibility for member countries to become insolvent in an orderly way is essential for financial markets to perceive the no-bailout threat as credible and price sovereign risk accordingly. An example is the Viable Insolvency Procedure for Sovereigns (vips).64 The vips is triggered when a country seeks assistance from the esm. It is then placed in a quarantine period of up to three years where it receives conditional loans from the esm. This esm shelter period is intended to gauge whether a country is indeed insolvent or is only suffering from temporary illiquidity. After three years, the country can voluntarily return to financing itself on the market or it can enter an insolvency procedure, which can last another year. 61 62 63 64

See G. Corsetti et al., A New Start for the Eurozone: Dealing with Debt (London: cepr Press, 2015). Ibid., at 10. Ibid., at 18. See C. Fuest et al., “A Viable Insolvency Procedure for Sovereigns in the Euro Area”, zew Discussion Paper, 14-053 (2014).

346

Schnellenbach

During this year, restructuring of the insolvent country’s debt is negotiated, and the country is supported with liquidity by the esm. After restructuring and reaching an agreement with creditors, the country returns to the market. At present, institutional innovations that could help restore the credibility of the no-bailout clause are being discussed in academic circles, but they are not on the political agenda. This is particularly unfortunate because a credible commitment of this kind could be the silver bullet that enforces market discipline on emu member countries, renders extensive vertical mechanisms of fiscal oversight superfluous and thereby preserves national-level fiscal sovereignty to the greatest extent. 5

Conclusions

The interaction between a monetary union and fiscal sovereignty produces a tension that needs to be addressed through an effective institutional framework. The main result of the literature surveyed in this paper is that there are two broad ways to achieve this. One way could be called a hierarchical approach. It relies on extensive vertical control mechanisms, and correspondingly on detailed restrictions of national-level fiscal sovereignty. There are, of course, many different possibilities for constructing such an institutional framework. The variations range from a fully fledged fiscal union to vertically imposed fiscal rules, and to formal coordination mechanisms such as the ­European ­Semester. At present, it is too early to empirically assess the ­effectiveness of the post-crisis institutions at the European level. What is for certain, ­however, is the fact that the emu has embarked on an institutional pathway that r­elies to a great extent on hierarchical mechanisms of control and coordination. The alternative would be a credible resurrection of the no-bailout clause. It is somewhat questionable whether a full resurrection could in fact be achieved or whether market participants and governments would implicitly always expect some form of bailout. There are, however, federations that have solved this problem in a credible and sustained fashion, such as Switzerland and the United States. If the emu followed this lead, then financial markets that efficiently recognise and price idiosyncratic sovereign risks could provide sufficient incentives for emu member countries to ensure sustainability of their public finances, in whatever way they prefer. Fiscal sovereignty would be preserved to the greatest possible extent, and complex vertical political mechanisms would not be necessary.

chapter 14

Fiscal Federalism in Times of Crisis: An Iron Law of Centralisation? Karl Kössler and Martina Trettel 1

Introduction

While the current global crisis is undoubtedly a multidimensional phenomenon that also encompasses social and political dimensions, the latter were triggered by the initial economic and financial crisis. After all, this core problem exacerbated societal front lines and instigated an increasingly fierce political struggle for dwindling resources. In this light, this chapter aims to assess, from a comparative perspective, whether and to what extent the economic and financial crisis has given rise to increasing fiscal centralisation. Thus, we take centralisation to mean the degree to which national governments have gained dominance relative to subnational entities by limiting their financial autonomy concerning revenue raising and spending and by making unilateral decisions about their respective responses to the crisis. Since the revenue and spending side of subnational financial autonomy are inextricably linked, it seems imperative to jointly analyse them. The canonical view of fiscal federalism literature holds that, from a normative perspective, “finance should follow function”.1 This approach ultimately suggests the allocation of revenue-raising powers to be determined by the prior distribution of legislative and executive competences and, as a consequence, of spending.2 In practice, however, federal systems are typically characterised by a vertical fiscal gap,3 that is, the own revenues of subnational entities fall short of their ­expenditure needs. The corresponding predominance of national ­governments * While this chapter was discussed jointly by both authors, Sections 1, 2 and 4 were written by

Karl Kössler and Section 3 by Martina Trettel. 1 A. Shah, “Introduction: Principles of Fiscal Federalism”, in A. Shah (ed.), The Practice of Fiscal Federalism: Comparative Perspectives (Montreal: McGill-Queen’s University Press, 2007) 3–42, at 9. 2 See Chapter 4 of this volume, “The Principles of Separation and Correspondence, the Comparative Method and the Problem of Semantic Change”, by M. Nicolini. 3 For a differentiation between vertical fiscal gap and vertical fiscal imbalance, see A Shah, “Introduction: Principles of Fiscal Federalism”, supra, at 28.

© koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_016

348

Kössler and Trettel

in generating revenues is commonly justified by public economists with their responsibility for macroeconomic stabilisation and redistribution, while subnational governments should be in charge of providing tailor-made public services at the local level.4 As a result of this predominance, the insufficient own revenues of subnational entities are supplemented, to a greater or lesser extent, by fiscal transfers from the national government, which are often subject to a variety of conditions.5 It is evident that conditional grants may be just as detrimental to the spending side of subnational financial autonomy as insufficient powers to raise own revenues. The revenue and spending sides of subnational financial autonomy are thus inherently interlinked. Times of economic and financial crisis seem to affect both these dimensions of autonomy appear to be in particular ways, which might favour or reinforce trends towards fiscal centralisation. With regard to revenues, national governments may be interested in augmenting their relative share of taxation to fulfil their above-mentioned responsibilities of macroeconomic stabilisation and redistribution. These tasks are unquestionably even more challenging and financially demanding during crises than in normal times. According to public economists, achieving such stabilisation requires that the national government be in control not only of monetary policy, but also of costly countercyclical fiscal measures because the latter are, if adopted at the national level, more effective in stimulating a depressed economy.6 Efforts towards stabilisation may likewise prompt the national government to restrict the ability of subnational entities to acquire funds through borrowing. As for redistribution, the typically differential effects of economic crises across a country are often regarded as a strong argument for centralising this function. Rather than through horizontal transfers, these effects appear to be smoothed out more effectively through vertical transfers from the national level to subnational entities and individuals.7 As far as payments to individuals are concerned, large social security schemes like family allowances or unemployment insurance usually incur particularly high costs in periods of crisis. As a rule, these schemes are—with Belgium being a 4 See, among many others, R.A. Musgrave, “Devolution, Grants and Fiscal Competition”, Journal of Economic Perspectives, 11 (1997) 65–72. 5 For a comprehensive overview, see R. Boadway and A. Shah (eds.), Intergovernmental Fiscal Transfers (Washington, dc: World Bank, 2007). 6 See W.E. Oates, “Assignment of Responsibilities and Fiscal Federalism”, in R. Blindenbacher and A. Koller (eds.), Federalism in a Changing World: Learning from Each Other (Montreal: McGill-Queen’s University Press, 2002) 39–50, at 40. 7 See. T. Persson and G. Tabellini, “Federal Fiscal Constitutions: Risk Sharing and Moral Hazard”, Journal of Political Economy, 104 (1996) 979–1009.

Fiscal Federalism in Times of Crisis

349

­notable exception8—the exclusive responsibility of the national government, which thus has to cover these increased spending needs. In analysing how subnational autonomy on the revenue side has been affected by the current crisis, we will focus in particular on powers of taxation and borrowing. Nonetheless, we are, of course, aware of other revenue sources such as proceeds stemming from natural resources and fees for the provision of public services. Natural resources are, however, only a major source of revenue in some of the cases covered in this chapter, such as the Canadian provinces of Alberta, Saskatchewan and Newfoundland,9 but largely irrelevant in others. Fees are, in comparison to taxes, typically much less lucrative and less contentious in times of crisis than borrowing. Likewise, on the spending side of subnational autonomy, there are several reasons to expect a crisis-induced penchant towards centralisation. From a historical point of view, exceptional situations have usually tended to entail an expanded role for the national government regarding expenditures.10 Moreover, the issue of the fiscal discipline of subnational entities and how to ensure it is unquestionably more central during crises than in normal times. Several observers regard the alleged lack of such discipline in general as a matter of concern in federal systems. According to them, the fiscal prudence of subnational entities is influenced in particular by opportunities for a federal bailout,11 by obscured accountability through large transfers in relation to subnational entities’ scarce own revenues and the degree of their access to borrowing.12 The problems of the disparity between subnational entities’ spending autonomy, on the one hand, and their allegedly lacking fiscal responsibility, on the other,

8

9

10

11 12

In response to Flemish pressure, the 2011 state reform in the aftermath of the Belgian government crisis assigned to the communities for the first time jurisdiction over parts of the large social security schemes, i.e., certain family allowances. See P. Popelier and B. Cantillon, “Bipolar Federalism and the Social Welfare State: A Case for Shared Competences”, Publius, 43 (2013) 626–647, at 627. See, specifically on this, G. Brosio, “The Assignment of Revenue from Natural Resources”, in E. Ahmad and G. Brosio (eds.), Handbook of Fiscal Federalism (Cheltenham: Edward Elgar, 2006) 431–458. With reference to various cases, Anwar Shah mentions in this regard times of war, secessionist threats, combating terrorism and, of course, imperatives of public debt management and fiscal discipline. See, A. Shah, “Comparative Conclusions on Fiscal Federalism”, in Shah, The Practice of Fiscal Federalism, supra, 370–393. Ibid., at 385–386. See J. Rodden, “The Dilemma of Fiscal Federalism: Grants and Fiscal Performance around the World”, American Journal of Political Science, 46 (2002) 670–687.

350

Kössler and Trettel

have surfaced in particular during the current crisis.13 It is thus plausible to expect that instruments like conditional grants, debt restrictions and balancedbudget requirements may become more widespread in times of crisis. In order to address the question of fiscal centralisation in times of crisis, we will draw on the experiences of two classical federations in North America (the United States and Canada) in Section 2 and then two devolutionary regional states in Europe (Italy and Spain) in Section 3. In both cases, it seemed imperative to us to first provide specific contexts, which partly shaped the impact of the current crisis on financial relations or at least functioned as significant backgrounds. For the United States and Canada, this is the historical context of experiences of crisis-related centralisation during times of war and the Great Depression. In the case of Italy and Spain, which, due to their regionalisation only after such kinds of historical crises, have not had similar experiences, the indispensable context is different, namely that of the European Union. After all, the centralisation of much of the crisis response at the supranational level had enormous repercussions on the eu member states, and both their national governments and subnational entities. Section 4 provides concluding remarks. 2

United States and Canada

The Historical Context: Centralisation in Times of War and the Great Depression The fact that the current crisis has come to be termed, especially in the us context, as the “Great Recession”14 immediately evokes images of the Great Depression of the 1930s. As intuition suggests, the above-mentioned economic crises effectively triggered processes of centralisation in both cases, albeit of different intensities and durations. In the us case, this occurred against the background of a constitutional framework that places few restrictions on the taxation powers of both the Congress (Article 1.8) and the states.15 As a result, both levels of government are bound to compete for the same sources of 2.1

13

14 15

See P.E. Peterson and D.J. Nadler, “Federalism’s Emerging Fiscal Crisis”, in P.E. Peterson and D.J. Nadler (eds.), The Global Debt Crisis: Haunting u.s. and European Federalism (Washington, dc: Brookings Institution Press, 2014) 3–14. See J. Jonas, “Great Recession and Fiscal Squeeze at u.s. Subnational Government Level”, imf Working Papers, 12/184 (2012) 1–39. However, a rather significant limit for the states, which is not expressly mentioned in the Constitution, arises from the dormant commerce clause doctrine. According to the Supreme Court states are prohibited by the commerce clause (Article i, Section 8, Clause 3) from passing any laws, also those introducing taxes, which would discriminate against

Fiscal Federalism in Times of Crisis

351

revenue. In 1913, however, the adoption of the 16th Amendment established the constitutional basis for the fiscal predominance of the national government. On the eve World War i, this amendment enabled the introduction of a federal tax on income, which was not subject to the constitutional obligation to apportion the revenue from direct taxes among the states on the basis of their respective populations (Article 1.2.3). During the war, the Congress made extensive use of this newly acquired competence by broadening the coverage of the income tax and by increasing the very low pre-war rates, which did not exceed 6 per cent, to up to 65 per cent.16 A landmark Supreme Court case in 1923 then enabled the national government to effectively utilise centralisation on the revenue side by augmenting its leeway on the expenditure side. At a time when the court still used to outlaw many national initiatives by invoking the residuary power of the states under the 10th Amendment, it explicitly allowed the practice of conditional grants in Massachusetts v. Mellon.17 Soon afterwards, the national government started to make ample use of this new prerogative during the Great Depression of the 1930s, above all, of course, to finance the country’s first comprehensive, and thus expensive, social welfare programmes introduced by the New Deal legislation. The Supreme Court was asked to review much of this legislation and upheld most of it. Even though Franklin D. Roosevelt’s infamous plan to alter the composition of the court in his favour eventually fell victim to widespread opposition,18 the judges nonetheless became more sympathetic to his New Deal and finally affirmed the constitutionality of its centrepiece, the 1935 Social Security Act.19 Whereas the aftermath of World War i had still witnessed a partial reversion of fiscal centralisation, long-standing federal predominance arguably began only in the context of the Great Depression and was maintained, thereafter, during World War ii and the Cold War.20

16

17 18

19 20

interstate commerce. See, W. Hellerstein et al, “Commerce Clause Restraints on State Taxation after Jefferson Lines”, Tax Law Review, 51 (1995) 47–114, at 50. See B.A. Wallin, “Forces Behind Centralization and Decentralization in the United States”, in Commission on Fiscal Imbalance (ed.), Texts Submitted for the International Symposium on Fiscal Imbalance (Quebec: Commission on Fiscal Imbalance, 2002) 25–36, at 26. Massachusetts v. Mellon, 262 u.s. 447 (1923). For an account of Roosevelt’s infamous Judicial Procedures Reform Bill (the “Court Packing Plan”) proposed in 1937 see W.E. Leuchtenburg, The Supreme Court Reborn: The Constitutional Revolution in the Age of Roosevelt (New York: Oxford University Press, 1995). Steward Machine Company v. Davis, 301 u.s. 548 (1937); Helvering v. Davis, 301 u.s. 619 (1937). See W. Fox, “United States of America”, in Shah, The Practice of Fiscal Federalism, supra, 345–369, at 348.

352

Kössler and Trettel

Canada differs from the us case in several regards, even if the two countries share a high degree of overlap between national and subnational taxation powers. Apart from certain specific revenue sources of the provinces,21 the latter are granted by the constitution the power of “direct taxation within the province” (Section 92(2) Constitution Act 1867, hereinafter ca 1867). The federal government, by contrast, was endowed with exclusive power over customs and excise taxes (Section 122 ca 1867) and, moreover, was given the power to raise “money by any mode or system of taxation” (Section 91(3) ca 1867). While the first two above-mentioned revenue sources of the national government accounted in the early days of federal Canada for roughly 80 per cent of the overall tax yield,22 the latter provision, vesting the national government with a general power of taxation, proved in the long run much more significant. After all, it made the most important revenues source of today, namely income taxes, which did not even exist in 1867,23 an area of federal-provincial concurrency. The national government could raise income taxes by virtue of its general power entrenched in Section 91(3) ca 1867, while the provinces could do so under Section 92(2) ca 1867, empowering them to direct taxation.24 Whereas the Canadian constitutional framework is in terms of the considerable overlap of revenue sources quite similar to the United States, crisis-related centralisation trends, shaped essentially by constitutional jurisprudence and intergovernmental agreements on fiscal relations, have been rather different. First, the Judicial Committee of the Privy Council (jcpc)25 was in comparison to the us Supreme Court clearly more reluctant to accept the Great Depression as justifying or even requiring a stronger national government. In times of crisis, the jcpc largely continued to maintain its tradition of interpreting powers, against the centralist intentions of the drafters of the constitution, in a

21 22 23

24 25

Examples are “licenses” (Section 92(9) Constitution Act 1867) and “taxation of natural resources” (Section 92A (4)). See P.W. Hogg, Constitutional Law of Canada (5th ed., Scarborough: Carswell, 2000), 6–1. See R. Knopff and A. Sayers, “Canada”, in J. Kincaid and G.A. Tarr (eds.), Constitutional Origins, Structure, and Change in Federal Countries (Montreal: McGill-Queen’s University Press, 2005) 103–142, at 124. For the issues of tax coordination and harmonization, see Section 2.1. Even though the Supreme Court of Canada was established in 1875, it was not really supreme in constitutional matters until 1949, when the appeal to the jcpc was finally abolished. Until then, the judicial activism of the members of the jcpc, a committee of the British House of Lords, left such a strong imprint on Canadian federalism that they have been referred to as “the lawmakers”, J. Saywell, The Lawmakers: Judicial Power and the Shaping of Canadian Federalism (Toronto: University of Toronto Press, 2002), 111.

Fiscal Federalism in Times of Crisis

353

deliberately province-friendly manner.26 This revisionist attitude earned them the characterisation as the “wicked stepfathers of confederation”27 and was reflected in a very extensive interpretation of the provincial jurisdiction over “property and civil rights in the province” (Section 92(13) ca 1867)28 and, more generally, the doctrine that the competences of each government level are “watertight compartments”.29 Not surprisingly, this insistence on dual federalism and a restrictive interpretation of the federal authority to make “laws for the Peace, Order and good Government of Canada” (Section 91 ca 1867), that is, its general residuary power, even during the Great Depression, sparked much discontent among federal politicians and some scholars. For example, one prominent lawyer commented: “the federal ‘general power’ is gone with the wind. It can be relied upon at best when the nation is intoxicated with alcohol, at worst when the nation is intoxicated with war; but in times of sober poverty, sober financial chaos, sober unemployment, sober exploitation, it cannot be used, for these, though in fact national in the totality of their incidents, must not be allowed to leave their watertight compartments”.30 The fact that the “Canadian New Deal” found less approval in constitutional jurisprudence is just one aspect in which the development of ­national-subnational relations in times of crisis differed from that in the United States. Another is the widespread use of implicit or explicit intergovernmental agreements on fiscal issues, which shaped the temporary trends of centralisation, decentralisation or recentralisation. During World War i, for instance, the levy of personal and corporate income taxes by the national government was broadly viewed as a transitional solution to finance increasing expenditures. In other words, there was still, at that time, intergovernmental consensus “that the field of direct taxation should be left to the provinces; and indeed federal rates were substantially reduced in the period between the First and the Second World Wars”.31 During World War ii, a negotiated compromise was 26

It was the declared mission of the jcpc to protect the provinces against Ottawa’s centralism by “adapting” the distribution of powers on the model of the United States and Australia. See Saywell, The Lawmakers, supra, at 121 and Hogg, Constitutional Law, supra, at 5–17. 27 Eugene Forsey, quoted in Hogg, Constitutional Law, supra, at 5–18. 28 This trend already started with the first ever jcpc case dealing with the Canadian Constitution in 1881. See Citizen Insurance Company v. Parsons, [1881] 7 a.c. 96. 29 Canada (ag) v. Ontario (ag), [1937] a.c. 326, 354. 30 Gilbert Kennedy, quoted in A. Linden, “Flexible Federalism: The Canadian Way”, in J. Fedtke and B.S. Markesinis (eds.), Patterns of Regionalism and Federalism: Lessons for the uk (Oxford: Hart Publishing, 2006) 17–60, at 29. 31 Hogg, Constitutional Law, supra, at 6–3.

354

Kössler and Trettel

once again reached that foresaw the temporary abandonment of provincial income and inheritance taxes in exchange for unconditional transfers from the national government. In the early post-war period, however, the Canadian government argued that such major challenges as the creation, unlike in the United States, of a comprehensive welfare state would equally necessitate fiscal centralisation as in wartime.32 Eventually, they succeeded in persuading the provinces to enter, in 1947, into “tax rental agreements”. According to this settlement, they would not relinquish their constitutional power to raise income and inheritance taxes but would “rent” them for compensation through federal transfer payments.33 However, this fiscal centralisation did not turn out to be permanent, as the provinces, above all Quebec and Ontario, struggled— somewhat successfully—for a greater share of revenues and more autonomy in setting tax rates.34 This eventually gave rise to the tax collection agreements of 1962. 2.2 Impact of the Current Crisis Before assessing whether and, if so, how the current crisis has tilted the financial relations between the national and subnational levels of government towards centralisation, it seems imperative to first examine the impact of the economic downturn on fiscal capacity. After all, potential centralisation is largely a legal and political reaction to precisely this impact. With regard to Canada, it is important to put the current crisis into perspective, as its effects concerning gdp decline and fiscal capacity were, compared to recessions in 1982–1983 and 1991–1993, less profound and also followed by a quicker recovery.35 In the United States, by contrast, the financial impact was significantly more profound. Tax collection in the states experienced a dramatic decline, including, in 2009, the largest drop since 1963.36 Even though the impact on revenues was diverse due to differences between the states regarding tax s­ ystems 32

33 34

35 36

See K.G. Banting, “Canada: Nation-building in a Federal Welfare State”, in H. Obinger et al (eds.), Federalism and the Welfare State: New World and European Experiences (Cambridge: Cambridge University Press, 2005) 89–137, at 95. See Hogg, Constitutional Law, supra, at 6–22. See D.M. Brown, “Fiscal Federalism: Searching for Balance”, in H. Bakvis and G. Skogstad (eds.), Canadian Federalism: Performance, Effectiveness and Legitimacy (Oxford: Oxford University Press, 2008) 62–87. See R. Simeon et al, “The Resilience of Canadian Federalism”, in Peterson and Nadler, The Global Debt Crisis, supra, 201–222, at 212. See D.J. Boyd and L. Dadayan, “State Tax Decline in Early 2009 Was the Sharpest on Record”, The Nelson A. Rockefeller Institute of Government State Revenue Report, 76 (2009) 1–19.

Fiscal Federalism in Times of Crisis

355

and the ability of their economic structure to overcome the crisis, significant budget gaps became a widespread problem and increased overall from usd 12.8 billion in 2008 to a peak of usd 174.1 billion in 2010. Immense gaps emerged in some of the most important states, such as California and New York.37 The fact that so many states struggled to meet their expenditure requirements originated in part also from the constitutional framework. Whereas the us Constitution is practically silent on financial matters, these are typically dealt with in great detail by the state constitutions.38 As a rule, the latter give “priority to taxpayers over service recipients” and “make it more difficult for states and localities to raise funds to finance public services”.39 Quite often, the revenues from certain taxes cannot be spent by policymakers for whatever purposes they deem fit, but are earmarked for specific purposes, e.g. ensuring small classes in public schools in California. This naturally limits the margin of fiscal flexibility in times of crisis. Other common restrictive provisions require qualified majorities in the parliamentary process for new taxes or even tax increases or even their ratification through a referendum. The deeply rooted distrust of “big government” is also reflected in regulations on borrowing, which often establish cumbersome special procedures and/or set a ceiling for the level of debt, although this is applied only to certain categories of debt.40 Against this constitutional background, in terms of own revenue sources, the states have mostly relied on sales tax and personal income tax for more than half a century, with the latter gaining in relative importance from 9 per cent of total state revenues in 1950 to 34 per cent in 2009.41 While revenues from income tax rise with income growth and those from sales tax with increases in consumption, the opposite is true in times of economic downturns. It is true that the crisis vulnerability of state sales taxes is aggravated by the fact that they tend to exclude such basic items as food from the tax base.42 37 38

39 40 41 42

For an overview see G.A. Tarr, “The Global Financial Crisis: A View from the American States”, L’Europe en Formation, 4/358 (2010) 33–49, at 33–34. See R. Briffault, “State and Local Finance”, in: G.A. Tarr and R.F. Williams (eds.), State Constitutions for the Twenty-first Century: The Agenda of State Constitutional Reform (Albany: State University of New York Press, 2006) 211–240. R. Briffault, “The Disfavored Constitution: State Fiscal Limits and State Constitutional Law”, Rutgers Law Journal, 34 (2003) 907–958, at 909. See G.A. Tarr, Understanding State Constitutions (Princeton: Princeton University Press, 2000), 109–113. See D. Brunori, “Introduction” in D. Brunori (ed.), The Future of State Taxation (Washington, dc: Urban Institute Press, 1998) 1–14, at 1. See D.A. Super, “Rethinking Fiscal Federalism”, Harvard Law Review, 118 (2005) 2576–2577, at 2631.

356

Kössler and Trettel

On the other hand, the federal government, being dependent on income and payroll taxes with even higher cyclical volatility, has experienced during the current crisis an even sharper decline in tax revenues. In light of decreasing fiscal capacity, states’ main crisis reactions have been spending cuts and, in many cases, making use of so-called rainy-day funds, that is, funds accumulated in economically prosperous periods for worse times. However, these funds are, for political reasons, rarely sufficiently endowed, and the states have proved, beyond certain procedural limitations, rather reluctant to exploit them fully.43 This is not to say that the states have absolutely refrained from introducing new taxes and raising tax rates. But these measures rather concern very specific taxes, such as Maine’s famous taxes on candy, haircuts and movie tickets. It does not concern, due to political motives and/or the above-mentioned constitutional restrictions on taxation, those generating significant revenue.44 In the Canadian case, the latter limits on provinces generating their own revenues are completely absent. As mentioned above, the provinces have, in comparison to most other subnational governments, extraordinary fiscal autonomy. As a result, own revenues as a share of total revenues clearly increased after the above-mentioned transitional periods of centralisation through coordinated federal tax reductions and provincial increases from 43 per cent in the early 1960s to 60 per cent in 2009.45 It has been argued that during the current crisis “the relative fiscal independence of Canada’s subnational governments was, indeed, a boon to the Canadian economy”.46 This relative independence has had not only the immediate effect of enabling the provinces to generate revenues from taxation. More indirectly, it was also one of three key reasons for the favourable conditions that they are usually offered when borrowing money. The other reasons are Canada’s aaa credit rating and the implicit federal guarantee on provincial debt. While the national government still had allowed Alberta to default in the 1930s, today’s fiscal equalisation system includes a Stabilization Agreement that foresees payments to provinces with declining revenues.47 The combined effect of these three factors is that 43

See S.D. Gold, “Lessons for the Future”, in S.D. Gold (eds.), The Fiscal Crisis of the States. Lessons for the Future (Washington, dc: Georgetown University Press, 1995) 367–382, at 370. 44 See Jonas, “Great Recession”, supra, at 21. 45 See J.-F. Tremblay, “Fiscal Problems, Taxation Solutions: Options for Reforming Canada’s Tax and Transfer System”, Mowat Centre’s Fiscal Transfers Series, 4 (2012) 1–25, at 5. 46 Simeon et al, “The Resilience”, supra, at 214. 47 See R.M. Bird and A. Tassonyi, “Constraining Subnational Fiscal Behaviour in Canada: Different Approaches, Similar Results?” in J.A. Rodden et al (eds.), Fiscal Decentralization and the Challenge of Hard Budget Constraints (Cambridge, ma: mit Press, 2003) 85–132.

Fiscal Federalism in Times of Crisis

357

the provinces’ default risk has been generally judged as negligible. As a result, they have enjoyed, unlike the us states and despite differences regarding their creditworthiness, quite similar bond returns.48 This has obviously facilitated borrowing and prepared the provinces, together with their extensive tax autonomy, to weather periods of crisis. As a matter of fact, their cumulative net deficit as a percentage of gdp has not exceeded 2 per cent in any year of the current crisis.49 The subnational entities in Canada and the United States differ from one another not only regarding the generation of their own revenues but also concerning financial transfers from the national government. Tackling the fiscal problems of the states was, besides stimulating the economy, the main rationale behind the 2009 American Recovery and Reinvestment Act (arra). Funding from the national government in this manner covered approximately 40 per cent of the budget gaps of the states and made transfers in 2009 the largest single revenue source for the first time in history.50 Much of the arra payments to the states served the purpose of limiting education cuts through a State Fiscal Stabilization Fund (sfsf) and of sustaining Medicaid by increasing federal contributions to this shared-cost healthcare programme for people with low income. Whereas most states welcomed the transfers from the national government, they were rejected by some as undue centralisation. Most prominently, the self-styled “governors against state bailouts”,51 Mark Sanford of South Carolina and Rick Perry of Texas, spoke out against arra funds because they would create a mentality of relying on federal funds among the states. As Sanford refused to accept funding for education, the us Congress adopted an amendment to the arra authorising the state legislature with a concurrent resolution of both houses to override the governor’s decision. Finally, he was ordered by the South Carolina State Supreme Court to accept arra payments.52 Whether arra is more than a transitional infusion of funds 48

49 50 51

52

See V. Galvani and A. Behnamian, “A Comparative Analysis of the Returns on Federal and Provincial Bonds”, Alberta University’s Department of Economics Working Papers, 7 (2009) 1–24. See K. Hanniman, “Fiscal Federalism and Provincial Credit Risk”, Mowat Centre’s Research Papers, 104 (2015), 1–30, at 6–10. See A. Greenblatt, “Where the Crisis Began, Governments Pass Creative Law and Desperate Measures”, Federations, 8 (2009), 5–7, at 5. R. Perry and M. Sanford, “Governors against State Bailouts”, Wall Street Journal, (2 December 2008), http://online.wsj.com/article/SB122818170073571049.html (accessed 28 October 2015). See J. Kincaid, “The Global Financial Crisis: Continuity in u.s. Federalism”, L’Europe en Formation, 358 (2010), 15–32, at 23–24.

358

Kössler and Trettel

in the sense of having a long-term and lasting (centralising) impact on federalstate relations remains doubtful.53 In comparison to Canada, the crisis-induced ad hoc character of fiscal relief for the us states is striking. To be sure, the Canadian government also took singular emergency measures, most notably a bailout of the automobile industry by the governments of Canada and Ontario, as well as the Economic Action Plan of 2009. Importantly, the adoption of the latter clearly involved more intergovernmental coordination than the arra in the United States, with numerous First Ministers’ Meetings (fmm) and gatherings at lower levels.54 This intense intergovernmental collaboration was highly unusual for the Canadian government of Stephen Harper, which was otherwise pursuant to its idea of “Open Federalism” averse to cooperation with the provinces. Notably, it convened only one fmm before the crisis, but four during the crisis. These meetings were closely intertwined with international crisis coordination taking place at G-20 summits. Still, these ad hoc emergency measures are supplemented in Canada, unlike in the United States, by continuous financial flows through a system of fiscal equalisation. That the “parliament and government of Canada are committed to the principle of making equalization payments to ensure that provincial governments have sufficient revenues to provide reasonably comparable levels of public services at reasonably comparable levels of taxation” is even enshrined in Section 36(2) of the Constitution Act 1982, even though the justiciability of this provision remains controversial.55 In any event, one impact of the crisis and the resulting emphasis on efficiency was the reinforcement of an older debate about the sustainability of equalisation, especially in light of Alberta’s fiscal capacity, which was by far the highest in Canada due to the province’s natural resources, and the fact that Ontario changed in 2009 to a province receiving federal equalisation payments. While all parties from Ontario, which has been suffering from a decline in its manufacturing industry, have demanded a “fair share” of federal spending through equalisation and other transfers,56 the crisis response has made the claim of disadvantage less forceful. After all, the above-mentioned auto 53 54 55

56

See Tarr, “The Global Financial Crisis”, supra, at 47. See C. Johns et al, “Formal and Informal Dimensions of Intergovernmental Relations in Canada”, Canadian Public Administration, 50 (2008) 21–41. In favour of enforceability in court, A. Nader, “Providing Essential Services: Canada’s Constitutional Commitment under Section 36”, Dalhousie Law Journal, 19 (1996) 306–372, at 349 and rather against it, Hogg, Constitutional Law, supra, at 6–10 and 33–3. See J. Hjartarson et al, “A Report Card on Canada’s Fiscal Arrangements”, Mowat Centre Report, 1 (2010) 1–26.

Fiscal Federalism in Times of Crisis

359

bailout ­benefitted that ­province above all. As to equalisation more generally, the image of Canada as a “giant mutual insurance company”57 still seems to enjoy—despite renewed debates—considerable support in both richer and poorer parts of the country.58 On the expenditure side, most subnational governments in both the United States and Canada share a tradition of fiscal discipline.59 In the Canadian case, this is combined with considerable intergovernmental coordination of expenditures, which to some extent follows from the constitutional framework. Since the Supreme Court allows, without an explicit provision in the constitution, both the federal government60 and the provinces61 to make expenditures beyond their respective fields of legislative competence, a minimum of coordination is required to avoid overlaps. This case law has enabled the federal government to use—in view of its greater financial resources—conditional grants for intrusions into provincial jurisdictions. In the words of the Supreme Court, “the simple withholding of federal money which had previously been granted to fund a matter within provincial jurisdiction does not amount to the regulation of that matter”.62 But, long before the crisis, federal grants relating to provincial jurisdiction underwent a development from unilateralism towards less stringent conditions and more fiscal coordination with the provinces during the budgetary process.63 During the crisis, this coordination worked quite well, which was arguably facilitated by developments that took place in the 1990s. Back then, mounting debt prompted the federal government to significantly reduce transfers to the provinces, above all for social spending, and to make a shift towards fiscal restraint. The broad ideological consensus since that time,

57 58 59 60

61

62 63

Former Saskatchewan Premier Allan Blakeney, quoted in Simeon et al, “The Resilience”, supra, at 211. Ibid., at 210. C. Sancak et al, “Canada: A Success Story”, in P. Mauro (ed.), Chipping Away at Public Debt: Sources of Failure and Keys to Success in Fiscal Adjustment (Hoboken: Wiley, 2011) 1–30. The court thereby usually refers to Section 91 (3) ca 1867 (“The raising of money by any mode or system of taxation”) and its power to provide appropriations guaranteed in Section 106 ca 1867. See Hogg, Constitutional Law, supra, at 6–17. The extensive interpretation of Section 92 (2) ca 1867 (“raising of a revenue for provincial purposes”) was championed in Ontario v. Board of Transport Commissioners, [1968] s.c.r. 118. Reference Re Canada Assistance Plan (bc), [1991] 2 s.c.r. 525, 567. See H. Lazar, “In Search of a New Mission Statement for Canadian Fiscal Federalism”, in: H. Lazar (ed.), Canada: The State of the Federation 1999/2000 (Montreal: McGill-Queen’s University Press, 2000) 3–39, at 29.

360

Kössler and Trettel

which attaches great importance to balanced budgets, has arguably facilitated coordination during the current crisis.64 Interestingly, for us states, fiscal discipline is not, or not only, a matter of political choice, but one of legal obligation. Every state, with the exception of Vermont, is subject to requirements to have balanced operating budgets.65 Interestingly, these restrictions were not imposed by the national government, as has been the case in other countries, but were entrenched by the states themselves in their constitutions and/or ordinary statutes. More specifically, balanced-budget requirements were already introduced in the wake of the state debt crisis of the 1840s after several states had defaulted on their debt incurred by infrastructure investments that failed to generate expected revenues.66 More recently, these historical limits were complemented in some states by additional spending restrictions, which, for example, restrict the rate of spending increase to the rate of growth of the state economy. It should be noted, however, that states’ balanced-budget requirements differ considerably in terms of their stringency, and typically only concern the operating budget, but not the capital budget, which includes long-term investments. Moreover, in periods of economic crisis, states have not always abided by these rules. A case in point is California, which has incurred deficits in several years since the late 1980s.67 While, in many states, these legal rules are actually followed, it has been argued that this often results rather from the underlying political tradition of fiscal discipline than from the restrictive force of the provisions themselves.68 Regardless of the forces behind states’ balanced-budget policies in times of crisis, it is obvious that these policies counteract and partially offset federal efforts of Keynesian countercyclical economic stimulation, as undertaken with the arra.69 If states have to raise taxes and/or cut spending for the sake of fiscal discipline and at the same time account for roughly 40 per cent of total government spending in the United States, such an effect is inevitable.

64 65 66

67 68 69

See T. Lewis, In the Long Run We’re All Dead: The Canadian Turn to Fiscal Restraint (Vancouver: University of British Columbia Press, 2003). See R. Briffault, Balancing Acts: The Reality behind State Balanced Budget Requirements (New York: Twentieth Century Fund, 1996). See A. Grinath et al, “Debt, Default and Revenue Structure: The American State Debt Crisis in the Early 1840s”, National Bureau of Economic Research Historical Paper, 97 (1997) 1–55. See J.I. Chapman, “California: The Enduring Crisis” in Gold, The Fiscal Crisis. supra, 104–140. See Briffault, Balancing Acts, supra, at 60. See Super, “Rethinking Fiscal Federalism”, supra, at 2609–2610.

Fiscal Federalism in Times of Crisis

361

As subnational governments in both countries typically aim to achieve balanced budgets, a crucial question is whether they do this primarily by increasing taxes or by decreasing spending. As we have seen above, tax increases in the us case were minimal, so that balanced budgets had to be achieved through measures on the expenditure side. Because the states had certain mandatory spending items—above all obligations to retirees that were tied to the stock market and were underfunded due to the crisis70—they typically implemented massive cuts of discretionary spending.71 This meant that funding shortages also concerned several countercyclical programmes like subsidies for the unemployed or people on low incomes, which would naturally require through enrolment growth precisely in times of crisis rather increased funding.72 3

Italy and Spain

3.1 The eu Context: Centralisation at the Supranational Level The current economic and financial crisis is having a significant impact on European countries. From a constitutional perspective, the crisis has been altering the equilibrium of national constitutional systems.73 This has happened, in particular, through a reduction of (financial) national sovereignty in favour of a “new European sovereignty” that has been justified by the need to develop fiscal and financial policies directed at preserving the economic stability of the Union as a whole and, moreover, as the sum of its 28 member states.74 In fact, since the start of the economic crisis in 2008, the European Union has started to develop regulatory measures in order to control and coordinate financial and deficit-making behaviours through the reduction of the public debt and—more generally—through mandatory budget rules. Consequently, the (financial) sovereignty of the member states has been, to some extent, put at stake, involving even their respective subnational entities.75 70 71 72 73

74 75

On this problem see J. Rodden, “Can Market Discipline Survive in the u.s. Federation?” in Peterson and Nadler, The Global Debt Crisis, supra, 40–61. See Jonas, “Great Recession”, supra, at 17–18. See Tarr, “The Global Financial Crisis”, supra, at 45. I. Ciolli, “The Constitutional Consequences of Financial Crisis and the Use of Emergency Powers: Flexibility and Emergency Sources”, Rivista dell’Associazione Italiana dei Costituzionalisti, 1 (2015) 1–22, at 5. Ciolli speaks about a “wound in the sovereignty”. Ibid., at 9. D. Braun and P. Trein, “Federal Dynamics in Times of Economic and Financial Crisis”, European Journal of Political Research, 53 (2014) 803–821, at 815.

362

Kössler and Trettel

It is worth considering if the new form of European economic governance, which is the sum of a long series of anti-crisis measures adopted at the supranational level, has had a direct impact on the decentralisation structure of the member states by imposing a centralising force on the evolution of territorial subnational structures. In order to be able to answer this question, it is necessary to briefly consider the essential legal measures specifically adopted to cope with the economic and financial crisis. Fiscal and financial rules have been part of the Eurozone governance structure since the very establishment of the European Monetary Union (emu) itself. In fact, the Maastricht Treaty (1992) created the first legal framework for setting up the Eurozone. This, as modified in 1997 by the Stability and Growth Pact (sgp), established limits on debt and deficit levels, specifically stating that: a) the annual deficits could not exceed 3% of gdp; b) public debt should remain under 60% of gdp; and c) no bailouts are allowed.76 While no effective instruments had been introduced specifically to guarantee the implementation of such a governance model, the rules of this framework, though violated by several states already before, definitively collapsed in 2008. At that time, Greece announced that its increasing cost of borrowing was leading the country towards a debt default.77 Given the fact that under the Maastricht economic regime a default was the only possible way out, the markets and the other eu member states, especially Germany, reacted to these statements by affirming that a bailout for Greece would be necessary in order to ensure the stability of the European Union.78 As a result, a new impetus was given to the search for true fiscal coordination at the eu level. In 2010, the eu elaborated three new tools to impose fiscal harmonisation: the so-called European Semester, the Euro Plus Pact and the Sixpack.79 The latter consisted of five regulations and one directive,80 which were used to conduct further reforms of the sgp that focused on improving compliance. These reforms did not change any of the conditions already imposed by the sgp but were aimed at enforcing greater budgetary discipline among the member states of the Eurozone by stipulating that sanctions would 76 77 78 79 80

R.D. Kelemen, “Law, Fiscal Federalism and Austerity”, Indiana Journal of Global Legal Studies, 22 (2015) 379–400, at 385. D. Ziblatt, “Between Centralization and Federalism in the European Union”, in Peterson and Nadler, The Global Debt Crisis, supra, 113–133, at 126. See Kelemen, “Law, Fiscal Federalism, and Austerity”, supra, at 387. Ziblatt, “Between Centralization and Federalism”, supra, at 127. Regulation (eu) 1175/2011; Regulation (eu) 1177/2011; Regulation (eu) 1176/2011; Regulation (eu) 1174/2011; Regulation (1173/2011); Directive 2011/85/EU.

Fiscal Federalism in Times of Crisis

363

have to come into force earlier and more consistently. Furthermore, the Sixpack, which is related to the Macroeconomic Imbalance Procedure, is an early-­warning system and correction mechanism for excessive macroeconomic imbalances.81 In 2011, eu leaders agreed on a “permanent architecture for bailouts” through the establishment of a fund called the European Stability Mechanism (esm) that would be at the disposal of the member states under severe conditionality limits.82 This completely new economic strategy pursued financial stability through the adoption of legal measures based on more strict fiscal discipline (e.g. balanced budget rules), which forced some states to adopt austerity measures.83 These provisions failed to calm the markets and so the heads of state of 25 member states84 agreed in January 2012 to adopt another anti-crisis measure, the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (tscg), commonly known as the European Fiscal Compact. As to its legal nature, it represents an international intergovernmental agreement that goes beyond the scope of eu law.85 The treaty includes certain major modifications regarding financial obligations that member states have to observe. The most important is the so-called “golden rule”, which requires the budgets of national governments to be balanced or in surplus. According to the treaty, this principle needs to be incorporated in the member states’ ­constitutions (or norms that are comparable in terms of rigidity). In this regard, scholars talk about “judicial enforcement of austerity”.86 If the states do not comply with this imposition: (1) they will not get financial assistance if 81

82 83 84 85

86

J. Martinez Sierra and C. Ferrer, “emu Governance and Decision Making Impact on Member States and eu Constitutional Systems”, Conference Paper, Constitutional ­Challenges: Global and Local, University of Oslo, (2014), http://www.jus.uio.no/english/ research/news-and-events/events/conferences/2014/wccl-cmdc/wccl/papers/ws12/w12 -sierra&ferrer.pdf (accessed 12 October 2015). Kelemen, “Law, Fiscal Federalism, and Austerity”, supra, at 388. Ibid., at 390. All but the United Kingdom and the Czech Republic. For more on this issue, see L. Pierdominici, “Constitutional Change through Euro Crisis Law: A Multi-level Legal Analysis”, Report of the European University Institute, (2014), http://eurocrisislaw.eui.eu/country/italy/topic/fiscal-compact/ (accessed 30 November 2015) and A. Kocharov, “Another Legal Monster? An eui Debate on the Fiscal Compact Treaty”, eui Working Papers, Law, 2012/09, (2012), http://cadmus.eui.eu/bitstream/ handle/1814/21496/LAW_2012_09_Kocharov_ed.pdf?sequence=1 (accessed 30 November 2015). See V. Ferreres Comella, “Amending the National Constitutions to Save the Euro: Is It the Right Strategy?” Texas International Law Journal, 48 (2013) 223–240.

364

Kössler and Trettel

needed;87 and (2) the European Court of Justice (ecj) will have the authority to sanction them if sued by another member state.88 As Ferreres Comella has argued, forcing the member states to constitutionalise specific financial obligations that were already part of the European economic governance system only resulted in demonstrating the profound weakness of the European institutional design. In fact, eu law prevails over national law in those fields in which competences have been transferred to the supranational level. That said, the use made by European institutions of national constitutions in order to make the latter comply with eu provisions has only had the effect of “[publicising] its own institutional limitations”.89 According to Colino and del Pino, all these measures are characterised by a “double upward shifting of authority”,90 first of the member states’ central governments towards the European Union and, second, of subnational entities towards central governments. This double centralisation effect has been forced by the anti-crisis measures, which, on the one hand, oblige member states to decrease their public deficits, balance their national budgets and maintain sound public finances and, on the other hand, as a waterfall effect, they indirectly impose, through the national level, severe restrictions over subnational budgetary policies.91 Indeed, “the stability treaty not only requires […] constitutional changes in each of the signatory states, but also raises significant questions about its relationship with eu law and the extent of the discretion left to member states to make fundamental decisions about taxation and spending”.92 It is to this extent that member states’ sovereignty over financial issues has been greatly reduced, thereby affecting, on the one hand, national levels of welfare and, on the other hand, the degree of autonomy granted to subnational entities by national “economic constitutions”.93

87 88 89 90

91 92 93

See Kelemen who discusses the bailout as a quid pro quo for financial rescue in “Law, Fiscal Federalism, and Austerity”, supra, at 390. As the Commission was not empowered to initiate the process. See Ferreres Comella, “Amending the National Constitutions”, supra, at 227. Ibid., at 231. C. Colino and E. del Pino, “Spanish Federalism in Times of Crisis”, in P. Peterson and D. Nadler (eds), Fiscal Crisis and the Centralization of Political Power (Washington: Brookings Institution Press, 2014) 159–178, at 161. Martinez Sierra and Ferrer, “emu Governance and Decision Making”, supra, at 8. S. Peers, “The Stability Treaty: Permanent Austerity or Gesture Politics?” European Constitutional Law Review, 8 (2012) 404–441, at 404. Martinez Sierra and Ferrer, “emu Governance and Decision Making”, supra, at 15.

Fiscal Federalism in Times of Crisis

365

3.2 Impact of the Current Crisis “Any federal financial constitution oscillates between solidarity and autonomy”.94 In times of economic stability, this sentence perfectly describes how fiscal federalism arrangements work with regard to the relationship between subnational entities and national governments. Despite this, it could be argued that this statement does not currently fit European federal (or quasifederal) states, where the outbreak of the financial crisis in 2008 forced them to introduce austerity plans and anti-crisis measures that had a strong influence on the relationship between central and subnational governments. Moreover, as was already mentioned in Section 3.1, most of the Eurozone countries had to accept a far-reaching loss of autonomy and discretion regarding fiscal and financial policies especially in the form of the imposition of regulatory measures decided first at the eu level and only at a second stage at the national level.95 Thus, it is worth wondering to what extent the current financial crisis has distorted the way in which federal (financial) arrangements work in the two largest and quasi-federal Southern European states, Italy and Spain.96 In these two countries, subnational governments have been forced by the central governments to accept, first, cuts in their budgets and, second, imposed conditions in order to receive extra funding to guarantee public services to their citizens.97 All this has happened through, on the one hand, the explicit revision of financial structures and, on the other hand, an implicit stretch of the constitutional and legal order. In fact, in Italy and Spain, the constitutional decentralisation of competences has not prevented central governments from taking decisions over economic and fiscal matters even in policy fields devolved to subnational governments.98 Constitutional courts have obviously played a pivotal role in the process of defining the competence-sharing conflicts that have arisen between the regions and the central states. 94

95 96 97

98

U. Hufeld, “Between Emergency Aid and Rütli Schwur: the Reconstructing of the Economic and Monetary Union in Times of Economic Crisis”, L’Europe en Formation, 361 (2011) 53–72, at 58; on that see thoroughly the chapter of Cheryl Saunders in this volume. Ibid. Braun and Trein, “Federal Dynamics”, supra. E. del Pino and E. Pavolini, “Decentralization in a Time of Harsh Austerity: Multilevel Governance and the Welfare State in Spain and Italy Facing the Crisis”, European Journal of Social Security, 2 (2015) 246–270, at 267. M. Corretja, “The Impact of the Financial and Sovereign Crises and the Budget Stability Principle on Fiscal Decentralization in Spain”, in A. Lütgenau (ed.), Fiscal Federalism and Fiscal Decentralization in Europe: Comparative Case Studies on Spain, Austria, the United Kingdom and Italy (Innsbruck: Studienverlag, 2014) 175–202, at 197.

366

Kössler and Trettel

In order to understand how the crisis has affected the territorial structure of Italy and Spain, we should highlight that, until 2007, a clear trajectory towards decentralisation in the form of a clear shift of powers and responsibilities from central governments to subnational ones prevailed in both countries. This led to an increase of regional welfare state expenditure, particularly in Spain. On average, the real yearly increase from 2000 to 2007 was equal to +13.7 per cent in Spain and to +4.1 per cent in Italy. This situation changed drastically following the onset of the crisis when the economy collapsed, bringing, in 2009, Spain’s and Italy’s deficits to, respectively, −11.4 per cent and −5.3 per cent. The austerity policies, mostly imposed by the eu (see Section 3.1), resulted in cuts especially at the subnational government level, which greatly reduced all expenditures, and especially those related to welfare (−7 per cent in Spain and −4 per cent in Italy).99 The way in which these two states identified responses to the financial crisis can be traced back to the institutional structure of financial arrangements, which are shaped by two essential and common elements. First, both systems are characterised by structures that lead towards what scholars have called “recurring fiscal indiscipline”100 to the extent that they foresee for subnational entities, on the one hand, small amounts of taxation and revenue powers and, on the other hand, vast spending powers and responsibilities in the sense of obligatory expenditures. The kind of structures in which revenue powers are assigned to a government level that does not detain the corresponding spending powers raise questions in terms of accountability and responsibility. In fact, it has been argued that they induce irresponsible financial management.101 To this extent, the Spanish variant of fiscal federalism relies on revenue and tax sharing with narrow autonomous taxation powers for the Autonomous Communities (acs), while they manage more than a third of Spain’s expenditures mostly in the areas of education, health and social services in which the acs enjoy even legislative and administrative powers. The system is integrated by an unconditional equalisation system that affects the regions in different ways depending on their fiscal capacity: the richest ones should be considered as “net payers” and the poorest as “net receivers”. This system has given rise to an over-compensation effect: by applying the rules of the equalisation funds, some acs that have higher fiscal capacity than the

99

This data is elaborated by del Pino and Pavolini, “Decentralization in a Time”, supra, at 251–253. 100 Colino and del Pino, “Spanish Federalism”, supra, at 160. 101 Corretja “The Impact”, supra, at 194.

Fiscal Federalism in Times of Crisis

367

mean end up below this mean, while other acs, with less fiscal capacity, end up above the mean.102 In Italy regions enjoy a lower degree of revenue autonomy if compared to Spain, while they are vested with a large degree of spending autonomy linked to the regional legislative and administrative powers as described in Articles 117 and 118 of the Constitution.103 The Italian system of intergovernmental financial relations is designed by Law 42/2009, which started to implement the financial part of the devolutionary project specified in Constitutional Law 3/2001. One very blurry aspect of the whole system concerns the equalisation mechanisms, which have neither been implemented nor completely designed. This is partly due to the impact of the crisis that resulted in a profound slowdown in the implementation of Law 42/2009. In fact, the issue of fiscal decentralisation and fiscal federalism has nearly disappeared from the political agenda in recent years.104 The second element that describes the Spanish and the Italian decentralised structures is the high degree of de jure asymmetries between regions, which also touches upon financial relations. In Italy, five regions105 enjoy—for historical, social and/or political reasons—a different system of autonomy that was bilaterally negotiated with the central government and granted in special regional statutes that have constitutional rank.106 The statutes provide specific arrangements even regarding financial relations. Compared to ordinary regions, the special regions get higher portions of the tax revenues collected on their territories that they can spend autonomously in accordance with the legislative and administrative competences foreseen in the special statutes. For example, Sicily keeps 100 per cent of the taxes generated in the region; the provinces of Trento and Bolzano are granted 90 per cent of their revenues,

102 Ibid., at 160. 103 See A. Valdesalici, “Features and Trajectories of Fiscal Federalism in Italy”, in A. Lütgenau (ed.), Fiscal Federalism and Fiscal Decentralization in Europe: Comparative Case Studies on Spain, Austria, the United Kingdom and Italy (Innsbruck: Studienverlag, 2014) 73–101 and S. Piperno, “Implementing Fiscal Decentralization in Italy between Crisis and Austerity: Challenges ahead”, Perspectives on Federalism, 4 (2012) 98–124. 104 Piperno, “Implementing Fiscal Decentralization”, supra, at 121. 105 Trentino-South Tyrol, Friuli Venezia Giulia, Aosta Valley, Sicily and Sardinia. 106 On the special status of regions in Italy, see P. Bilancia et al., “The European Fitness of Italian Regions”, Perspectives on Federalism, 2 (2010) 122–174 and F. Palermo, “Federalismo fiscale e Regioni a statuto speciale. Vecchi nodi vengono al pettine”, Le Istituzioni del Federalismo, 1 (2012) 9–26.

368

Kössler and Trettel

while Friuli Venezia Giulia keeps 70 per cent of its taxes.107 Similarly in Spain, the financial relations of two acs, namely Navarra and the Basque Country, are ruled, for historical reasons, by the so-called foral system entrenched in the Spanish Constitution. According to this system, they conclude bilateral agreements with the central government through which they directly collect all revenues generated on their territories and pay a fee to the central government for the services the latter offers inside their borders.108 These characteristics have had some kind of influence on the way in which the financial crisis has been faced. In Italy, since the very beginning of the crisis, the political and economic strategy has been to contain the crisis through a return to the central government as the principal decision-making actor while reducing the space for the decision-making powers of other levels of government, even if granted in the Constitution.109 The small amount of subnational revenue autonomy and the consequent reliance of constituencies on central governments in order to obtain resources made it possible to reduce the constitutional portions of self-government powers especially in terms of financial autonomy. These behaviours have been justified by the central governments using both the arguments of the emergency caused by the crisis and of the austerity measures imposed by the eu. It can still be argued that the fiscal federalism structure, as conceived in the constitution, has not been able, in times of scarce resources, to defend the autonomy of subnational entities.110 In fact, the above-mentioned problematic combination of little revenue powers with vast spending responsibilities has only aggravated the fiscal distress of the regions. The austerity measures imposed by the central government on the regions were hardly discussed and negotiated with subnational governments and therefore have been frequently challenged in front of the Italian Constitutional Court as violating regional financial autonomy. In this regard, Law Decree 138/2011, Additional Urgent Measures for Financial Stabilisation and for Development111 (converted into law by Act No 148/2011) provided, in its Article 114, that the number of members of the regional councils had to be reduced and introduced measures to make regions comply with this specific provision. 107 See F. Palermo, “Autonomy and Asymmetry in the Italian Legal System: The Case of the Autonomous Province of Bolzano/Bozen”, in G. Pola (ed.), Principles and Practices of Fiscal Federalism: Experiences, Debates and Prospects (Farmham: Ashgate, 2015) 227–241. 108 The other 15 acs are governed by the common regime in which the central government collects all tax revenues and then distributes them among the acs and the central government in specific percentages. See Corretja “The Impact”, supra, at 180. 109 del Pino and Pavolini, “Decentralization in a Time”, supra, at 256. 110 As argued by Corretja “The Impact”, supra, at 201–202. 111 On that see Ciolli, “The Constitutional Consequences”, supra and A. Carmona-Contreras, “El decreto-ley en tiempos de crisis”, Revista Catalana de Dret Pùblic, 47 (2013) 1–20.

Fiscal Federalism in Times of Crisis

369

Given that the competence over the internal organisation of regional councils is exclusively regional, the Constitutional Court declared this norm partly unconstitutional, especially with regard to autonomous regions.112 Many other examples regarding the intrusion of the central government into regional competences could be mentioned, especially with regard to imposed cuts on expenditures, which have been challenged in the Constitutional Court. Mostly, these measures have been safeguarded, due to the scarcity of resources, which affects all levels of governments and the subsequent financial emergency seen as a justifying tool to the centralist reinterpretation of the constitutional division of competences.113 On the other hand, in Spain, the first phase of the crisis (2008–2010) was characterised by a reform of the regional financial system, which, opposed to what happened in Italy, intensified fiscal decentralisation, encouraged the implementation at the subnational level of countercyclical measures and authorised debt operations on the part of the acs.114 The central government made use of its spending powers to stimulate spending on the part of the regional authorities through conditional and matching grants and to subsidise regional industries. Therefore, the acs continued to increase their spending since they did not have many incentives for cutting expenses or limiting borrowing behaviours. This changed in 2009, when the excessive deficit procedure activated by the eu signalled that there were some problems.115 Thus, the second phase of the crisis (from 2010 onwards) has been faced differently due to the stronger influence of eu requirements that had a deep impact on central governments’ financial behaviour and consequently on the financial autonomy of the acs. As a matter of fact, the central state imposed cuts on wages and salaries of acs’ public employees, procyclical policies on all acs, and limits on regional borrowing, monitored and enforced compliance and excluded the acs from the negotiation process regarding the establishment of budget consolidation and debt limits. It was even decided to set an additional margin for the adjustment burden conceded by the eu in favour of the central government’s consolidation target.116

112 G. Martinico and L. Pierdominici, “Crisis, Emergency and Subnational Constitutionalism in the Italian Context”, Perspectives on Federalism, 6 (2014) 116–140, at 124. 113 T. Groppi, “Lo stato regionale italiano nel xxi secolo. Tra globalizzazione e crisi economica”, Revista d’Estudis Autonomics i Federals, 21 (2015) 35–72, at 58–61. 114 Corretja “The Impact”, supra, at 175. 115 Colino and del Pino, “Spanish Federalism”, supra, at 170–171. 116 Ibid., at 172–173; on that also, C. Viver Pi-Sunyer, “Impact of the Global Economic Crisis on the Political Decentralization in Spain”, L’Europe en Formation, 358 (2010) 61–90.

370

Kössler and Trettel

As in Italy, the Spanish Constitutional Court played a central role in defining the legitimate measure of the central government’s intrusion in acs’ financial policymaking. In this regard, Judgment No 134/2011 declared that the provisions of some internal laws on stability were unconstitutional for violating acs financial autonomy, while, at the same time, pointing out the implicit existence of a balanced-budget principle in the Spanish legal order even without a specific constitutional provision.117 This statement did not stop the central government from implementing the balanced-budget principle (commonly known as the “golden rule”) in the constitutional order as required by the European Fiscal Compact.118 Therefore, in 2011, for the second time since its adoption, the Spanish Constitution was modified. Its new Article 135 states: “The Autonomous Communities in accordance with their respective laws and within the limits referred to in this article, shall take the appropriate procedures for effective implementation of the principle of stability in their rules and budgetary decisions”. Further, it specifies: “the state and the self-governing communities must be authorised by an act in order to issue public debt bonds or to contract loans”. These measures served as an “accelerator” for the ongoing process of fiscal recentralisation and as a constitutional guarantee of its consequences.119 As pointed out by Martinico, “this limitation of the possibility of contracting loans for the acs represents a major paradigm shift compared to the past, when financial autonomy and the increase in regional tax power had been the center of the reformist agenda of the Spanish Autonomous State”.120 Even in Italy, in 2012, the constitutional budget rules were modified in order to comply with the European Fiscal Compact, introducing in Article 81 a balanced-budget rule.121 As far as subnational autonomy is concerned, it is ­important to underline that Article 117, devoted to the division of legislative powers between the state and the regions, was amended to grant the state exclusive legislative power over the “harmonisation of public budgets”, when it had previously been a shared competence between the state and the 117 V. Ruiz Almendral, “Curbing the Deficit in Spain and its Autonomous Communities: a Constitutional Conundrum”, tvcr, 1 (2013) 68–77. 118 Ferreres Comella, “Amending the National Constitutions”, supra, at 234. 119 G. Martinico, “The Impact of the Treaty On Stability, Coordination and Governance on the National Constitutional Structure: The Regional Example”, Michigan Journal Of International Law, (2013) 101–113, at 110–111. 120 Ibid., at 111. 121 Legge Costituzionale n. 1/2012, http://www.normattiva.it/uri-res/N2Ls?urn:nir:stato:legge .costituzionale:2012-04-20;1!vig= (accessed 30 November 2015).

Fiscal Federalism in Times of Crisis

371

­regions.122 Even Article 119 was modified by placing limits and constraints on the possible recourse to borrowing for regions and other territorial entities, and it now specifies the obligation to comply with “the concomitant adoption of amortisation plans and subject to the condition that budget balance is ensured for all authorities of each regions, taken as a whole”.123 While the Spanish and the Italian central governments were—and still are—“imposing” a centralising force on subnational entities, the regions— especially­the richest ones—are counteracting and pushing for more ­self-government powers, especially in the financial and fiscal field. They use as one of their arguments the existence of subnational entities that already enjoy special autonomous powers, creating a strong financial asymmetry throughout the national territory. The tensions with the central government contributed to the emergence of subnational requests (from regions such as Catalonia in Spain and Veneto in Italy) for self-determination and secession. At the same time, these demands can be read as an attempt to “de-solidarise” the financial dynamics by trying to force the central state to establish limits on the solidarity principle in exchange for putting an end to secessionist claims.124 In this regard, Catalonia, after Judgment 31/2010,125 which frustrated the possible improvement of legislative and financial powers, started to demand that the central government adopt a finance model similar to the one used in Navarra and the Basque Country. The denial of these requests can be considered one of the factors that led the Catalan government to follow the secessionist path.126 4

Conclusions

“If the President is satisfied that a situation has arisen whereby the financial stability or credit of India or of any part of the territory thereof is threatened”, he may declare by virtue of Article 360 of the Indian Constitution a state of emergency. Such a declaration, among other things, subjects the money bills of subnational legislatures to the approval of the Indian president and ­empowers him to give directives to the states in financial matters. To be sure, none of 122 Martinico and Pierdominici, “Crisis, Emergency and Subnational Constitutionalism”, supra, at 123. 123 Ibid. at 124. 124 Corretja “The Impact”, supra, at 178. 125 stc 31/2010 adopted on June 28th 2010, on the 2006 Catalan Statute. 126 See X. Lastra-Anadòn, “Regional Identity and Fiscal Constraints in Spanish Federalism”, in Peterson and Nadler, The Global Debt Crisis, supra, 179–200.

372

Kössler and Trettel

the four countries analysed have a similar constitutional provision that would authorise the central government in the case of a national or merely regional financial crisis to intervene as directly and forcefully. Yet, the current crisis has had a significant impact on all of them, albeit to different degrees. On the revenue side, it is obvious that subnational entities in both European countries have less tax autonomy than the us states, notwithstanding their extensive self-restrictions in state constitutions, and even much less than the Canadian provinces. But unlike in Italy, where the crisis contributed to further protraction of the reform of regional financing through Law 42/2009, Spain effectively implemented a reform to increase the autonomy of its Autonomous Communities with regard to both taxation and borrowing. Interestingly, this occurred when the crisis was already under way and was only at a later stage reversed in light of eu measures for being damaging to efforts aimed at countrywide fiscal consolidation. In Canada, by contrast, the fiscal autonomy of the provinces has come to be valued, precisely in the crisis context, as a cornerstone of their resilience. Their extensive taxation powers have not only bolstered the ability of the provinces to raise funds from their citizens, but also turned out to be one of the key factors in being offered favourable borrowing conditions. Compared to Canada, the subnational governments of the other countries have generally been much more reliant in a time of crisis on having their more scarce own revenues supplemented by financial transfers from the national government. Quite remarkably, even the United States, which otherwise largely adheres to Alexander Hamilton’s vision of strictly dual fiscal federalism,127 witnessed a situation where transfers became, at one point of the crisis, the states’ largest single source of revenue. While certain states regarded the arra as an act of undue centralisation, even if most likely without a long-lasting impact, it has not been generally met with widespread opposition at the subnational level. This is quite remarkable because intergovernmental financing has been traditionally in the United States, due to the Hamiltonian paradigm of dual fiscal federalism, much less prevalent than in the other three cases. Even though Canada, Italy and Spain also implemented specific ad hoc crisis responses, these have been complemented, unlike in the United States, by continuous funding from schemes of fiscal equalisation. Not surprisingly, older debates about the sustainability and efficiency of these schemes resurfaced during and due to the crisis. This holds true, for instance, for Canada and, even 127 Hamilton underlines that “the individual States should possess an independent and uncontrollable authority to raise their own revenues for support of their own wants”. (The Federalist Papers, No. 32).

Fiscal Federalism in Times of Crisis

373

more so, for Spain, whose specific, and complex, equalisation model had been criticised long before for providing perverse incentives to the Autonomous Communities. Moreover, in Spain with Catalonia and, to a lesser degree in Italy with Veneto, issues of financial redistribution more broadly have been increasingly linked in the context of the crisis with claims for secession. As far as subnational spending is concerned, there is a noticeable gap between the two North American cases and the two European ones. Most of the Canadian provinces and us states uphold a deeply rooted tradition of fiscal restraint, which is epitomised in the latter case by self-imposed balancedbudget rules enacted in state constitutions as early as in the mid-19th century. By contrast, the finding that the constitutional framework favours “recurring fiscal indiscipline”,128 as it grants subnational entities little autonomy regarding their own revenues, but burdens them with vast spending responsibilities, applies both to Spain and Italy. Typically, such an excessive disproportion between transfers, on the one hand, and subnational entities’ own revenues, on the other, tends to obscure accountability and impair fiscal prudence.129 In both countries, the obligation to comply with eu anti-crisis measures finally provided the context in which constitutional changes were made, albeit like in other member states,130 towards balanced-budget rules rather than towards more autonomy concerning subnational entities’ own revenues (Article 135 of the Spanish Constitution, as well as Articles 81 and 119 of the Italian Constitution). Paramount among these eu measures is, of course, the 2012 “European Fiscal Compact”, which not only requires “constitutional changes in each of the signatory states, but also raises significant questions about its relationship with eu law and the extent of the discretion left to member states to make fundamental decisions about taxation and spending”.131 While the abovementioned amendments were undoubtedly induced by the crisis and the eu’s response to it, they must also be seen as being embedded in a broader trend since the 1990s towards the adoption of formal legal rules mandating fiscal responsibility.132 Such provisions are generally deemed significant even if the fiscal discipline of subnational entities is achieved in some instances without 128 Colino and del Pino, “Spanish Federalism”, supra, at 160. 129 See Rodden, “The Dilemma of Fiscal Federalism”, supra, at 670–687. 130 New provisions of a similar nature include Article 13(2) of the Austrian Constitution (Bundes-Verfassungsgesetz), as well as Articles 109 and 110 of the German Basic Law. 131 S. Peers, “The Stability Treaty: Permanent Austerity or Gesture Politics?” European Constitutional Law Review, 8 (2012) 404–441, at 404. 132 See L. Liu and S.B. Webb, “Laws for Fiscal Responsibility for Subnational Discipline: International Experience”, World Bank Policy Research Working Paper, 1 (2011) 1–56.

374

Kössler and Trettel

them and in others is absent in spite of them.133 That policymakers may indeed have in certain contexts strong political incentives to violate rules is, of course, one of the main themes of so-called second-generation fiscal federalism.134 If such provisions are in fact applied, however, they inevitably thwart to some extent the usual countercyclical policies of national governments for economic stimulation. Overall, the fiscal response to the crisis in Canada, took the form of national-­ subnational coordination, which was clearly more consensus-driven and, in general, appears to be one of the crucial factors for multilevel systems to endure times of recession.135 Strikingly, the formidable challenge posed by the crisis made even a Canadian government otherwise averse to intergovernmental cooperation remindful of the country’s long-standing tradition of what has been famously dubbed “federal-provincial diplomacy”.136 In Italy and Spain, this coordination between government levels has been less extensive. As a consequence, crisis responses in these two countries, decided more or less unilaterally, were more prone to be challenged by subnational entities before the respective constitutional courts. These courts have demonstrated, similar to the us Supreme Court during the Great Depression, on the whole a certain penchant for upholding in light of the crisis change towards centralisation. In doing so, they have frequently echoed the classical arguments of public economists, already outlined in the introduction, that in a situation of financial emergency, it is desirable to have a strong national government for both territorial redistribution of financial resources and macroeconomic stabilisation. Accomplishing the latter task concerns not only fiscal policy, of course, but also monetary policy, which reminds us that fiscal federalism is only part of the broader challenge for multilevel systems to weather a crisis. Monetary policy, of course, is typically beyond the subnational sphere of influence, for it is shaped at the national level and, within the Eurozone, the supranational level. Even if the imperatives of redistribution and stabilisation may indeed justify fiscal centralisation as emergency measures, the crucial question is whether they turn out after the crisis to be merely provisional or remain in place.137 133 See Shah, “Comparative Conclusions”, supra, at 385–386. 134 See W.E. Oates, “Toward a Second-Generation Theory of Fiscal Federalism”, International Tax and Public Finance, 12 (2005) 349–373. 135 See M. Molnár, “Fiscal Consolidation: What Factors Determine the Success of Consolidation Efforts?” oecd Journal: Economic Studies, 1 (2012) 123–149, at 143–147. 136 R. Simeon, Federal-Provincial Diplomacy (Toronto: University of Toronto Press, 1972). 137 N. Bosch and A. Solé-Ollé, “Editorial: Decentralisation and the Crisis / The Crisis and Recentralisation”, Report on Fiscal Federalism 2012, Institute d’Economia de Barcelona (ieb) 7–11, at 9.

Fiscal Federalism in Times of Crisis

375

Indeed, crisis-related strengthening of the national government may be used to justify, like for some time in post-war Canada, more sustained centralisation. Arguably, this seems more likely in younger multilevel systems like Italy and Spain, where national-subnational fiscal relations are more in flux and, beyond that, further complicated by the eu as an additional level. Whether longer-term fiscal centralisation will occur in any of the countries analysed, it is still too early to judge.

chapter 15

Comparative Research and Fiscal Federalism Ronald L. Watts 1

The Limitations and Value of Comparative Analysis

In a collection on devolved financial arrangements and fiscal federalism in a variety of countries, it is important to begin by recognizing the limitations of comparative analysis. There is no single model of federalism or of federal financial arrangements that is universally applicable everywhere. The basic notion of federalism involves the combination of shared rule for some purposes and constituent unit self-rule for other purposes within a single political system so that neither order of government is constitutionally subordinate to the other. This basic idea has been applied extensively in many countries, but in a variety of different ways to fit different circumstances. One cannot, therefore, just pick institutional models off a shelf. Even where almost identical institutions have been adopted, different circumstances may make them operate differently. A classic illustration of this is the operation of similar formal constitutional amendment procedures in Switzerland and ­Australia during the past century which produced nearly 100 amendments in the former and only eight in the latter. But as long as the cautions about the transferability of institutions are kept in mind, there is a genuine value in undertaking comparative analyses. Many of the basic problems of federal finance are common not only to virtually all federations but also to most systems of multi-level governance, including such decentralised unions as Italy and Japan. Examples include such issues as: • the correction of vertical financial imbalances between expenditure responsibilities and financial resources; • pressures for equalisation to correct horizontal imbalances; • the need for coordination of taxing powers; • the need for regular adjustment of financial arrangements • and the relation of all of these to the political setting, institutions and processes of the country concerned. * Originally published as R.L. Watts, “Introduction: Comparative Research and Fiscal Federalism”, Regional & Federal Studies, 13 (2007), 1–6. Reprinted by permission of the publisher Taylor & Francis Ltd, www. tandfonline.com. © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_017

Comparative Research and Fiscal Federalism

377

Furthermore, comparative studies may help to identify options that might otherwise be overlooked and to identify consequences that might otherwise not be foreseen. In addition, comparisons may provide both positive and negative lessons, enabling us to learn not only from the successes but also the failures of arrangements and processes in other countries. 2

The Importance of Political Setting

In the study of federal and intergovernmental financial relations there is an important body of literature, largely written by economists, which covers much useful ground in the comparison of intergovernmental financial arrangements in analytical and normative terms. Authors such as Richard Musgrave, Wallace Oates, Charles McLure, James Buchanan, Richard Bird, Robin Boadway and many others have written about allocations of taxing powers, expenditures, correcting imbalances, and the use of conditional and unconditional transfers in terms of such concepts as economic efficiency and equity. Useful as these are, I would argue that in considering the dynamics of federal and intergovernmental relations, it is also important to consider the broader social and political context within which these financial relations operate.1 In virtually all federal and intergovernmental systems, financial arrangements have invariably constituted an important, indeed crucial, aspect of their political operation. The importance of financial arrangements derives from the fact that financial resources a) b) c)

play a large part in determining the relative political and economic roles and influence of the different governments within the polity; are a major means for facilitating flexibility and adjustment; and shape public attitudes about the costs and benefits of the activities of different governments.

This political significance places financial relations between central and constituent-unit governments at the heart of the processes of intergovernmental relations. Intergovernmental financial arrangements are therefore not simply technical adjustments but inevitably the result of political compromises. Thus, understanding intergovernmental financial relations requires an 1 R.L. Watts, “Federal Financial Relations: A Comparative Perspective”, in H. Lazar (ed.), Canada: The State of the Federation 1999/2000: Towards a New Mission Statement for Canadian Fiscal Federalism (Kingston, Ontario: Institute of Intergovernmental Relations, Queen’s University, 2000) 371–388.

378

Watts

­ nderstanding of the political context within which they occur. That is why u it is entirely appropriate that this collection includes discussions of both the principles and politics of devolved financial arrangements. An illustration of how distinctive institutional forms can affect intergovernmental financial arrangements is provided by the impact in Canada of the parliamentary form of its federal institutions and the distinctive character of its constitutional distribution of powers. Other parliamentary federations such as Australia, India and Germany share with Canada second chambers that are weaker than the popular chamber and a tendency for executive predominance leading to “executive federalism” as the characteristic form of intergovernmental relations, including those concerning intergovernmental financial ­arrangements. In the constitutional distribution of powers, while most other federations have emphasised interdependence more and have included large areas of concurrent jurisdiction, the Canadian constitution has emphasised the exclusive jurisdiction of each order of government. Such differences not only affect media interpretations and public attitudes about intergovernmental relations, which in Canada have taken on a competitive and zero-sum character, but have important implications for appropriate financial arrangements. 3

Decentralisation and Non-Decentralisation

Whether in the developed oecd countries, the developing countries of Latin­ America, Asia or Africa or the transitional economies of Eastern Europe, decentralisation and devolution are currently being widely discussed or advocated as possible cures for many of the ills afflicting the countries in question.2 But in speaking of the processes of decentralisation, we need to be clear about what we mean by “decentralisation”. Two important sets of distinctions need to be noted. The first is that between the processes by which decentralisation is achieved, i.e., the process of devolution involving a transfer within a political system whereby regional or local legislative, administrative or financial responsibilities are progressively increased, and the processes by which non-centralisation is maintained and secured to ensure that regional and local autonomous policy-making is not encroached upon. These two different senses of decentralisation may indeed require quite different processes and dynamics.

2 R. Bird and T. Stauffer (eds.), Intergovernmental Fiscal Relations in Fragmented Societies (Bale: Helbing and Lichtenhahn, 2001).

Comparative Research and Fiscal Federalism

379

The second distinction is that between decentralisation and autonomy. The former may refer to the level at which activities are carried out, whereas the latter relates to the degree to which these activities are free from control or regulation by other levels of government. This distinction in broad terms parallels that between decentralised unitary political systems and the autonomy that characterises units within federal political systems. It should be noted in passing that in this respect constitutionally decentralised unions may constitute a hybrid between decentralised unitary systems and federations. The distinction between decentralisation and autonomy is significant, for instance, in the discussion of whether intergovernmental financial transfers should be conditional or unconditional in character and whether regional powers to levy their own taxes may be preferable to guaranteed shares of centrally levied taxes. The degree of relative financial autonomy may be an important factor in accommodating the political concerns of different groups within a fragmented society wishing to secure their distinctiveness. In considering decentralisation, non-centralisation and autonomy, it is also important to recognise the number of levels of multi-level governance. Both regional and local governments have become increasingly important. In some federations and quasi-federations the constitution explicitly recognises the roles, responsibilities and financial resources not only of the regional units of government but also of local governments. Notable examples are Switzerland, Germany, India and South Africa. Yet a further tier of multi-level governance occurs in those federations or quasi­federations which are members of the European Union, thus extending intergovernmental relations over four tiers. The interrelation of these multiple tiers has been particularly well illustrated in recent years by the developments within the German federation. 4

Common Elements of Intergovernmental Financial Arrangements

For comparative research we may identify for consideration a number of common elements affecting intergovernmental financial arrangements. These include: • the appropriate allocation of taxation powers and resources to different orders of government • the allocation of expenditure responsibilities to different levels of government • the size and nature of vertical and horizontal imbalances and the role of transfers adjusting these

380

Watts

• the scope and form of equalisation transfers • and the political processes and institutions established for adjustment of these. Appropriately, it is these issues upon which the contributions to this collection focus. A common characteristic of the allocation of taxing powers in nearly all federations and intergovernmental systems, although there is considerable variation, is that the majority of major revenue sources have usually been assigned to the central government in the interests of economic efficiency and minimizing conflicts in taxation policies. On the other hand, although again there is much variation, there has usually also been greater decentralisation of expenditure responsibilities on grounds that this contributes to efficiency. Given this situation, virtually every federal or decentralised political system has had to correct the resulting vertical imbalance between revenue allocations and expenditure responsibilities. The desire to rectify imbalance has typically raised the following issues: the need for intergovernmental transfers; the form they should take—whether in the form of transfers of taxing powers, shares of specific central tax proceeds, or conditional or unconditional grants; and the balancing of requirements for financial responsibility and accountability against the desirability of autonomy.3 Furthermore, given the corrosive effect upon political cohesion of disparities among regions in revenue capacities and needs, some form of equalisation transfers to correct horizontal imbalances has been typical of most intergovernmental financial arrangements, although here again the actual extent and form of these transfers has varied enormously. Since the values of revenue resources and the costs of expenditure responsibilities inevitably change over time, most intergovernmental financial regimes have also had to develop processes and institutions for the periodic adjustment of the financial arrangements in a way that is not dependent solely upon the unilateral decisions of the central government.4 5

Some Questions for Consideration

In undertaking comparative research on intergovernmental financial relations there are a number of questions that are appropriately considered in this collection: 3 Watts, “Federal Financial Relations”, supra, at 372–382. 4 Ibid., 382–384.

Comparative Research and Fiscal Federalism

381

• To what extent have differences in the degrees of internal social diversity, and hence in degrees of constitutional and political non­centralisation and emphasis on provincial autonomy, been responsible for the variations in financial arrangements? • To what extent has the acceptance of the unavoidable interpenetration of the functions of different orders of government in contemporary federal and intergovernmental systems made necessary considerable intergovernmental financial interdependence, regular adjustments of financial arrangements and the use of a federal spending power in areas of constituent unit jurisdiction in order to facilitate flexibility?5 • The period 1950–1980 saw most federations and unions undergo a pattern of decentralisation of both government revenues and expenditures.6 To what extent has that trend continued? • To what extent have the dynamics of intergovernmental financial relations been dominated by processes of intergovernmental political bargaining? • Taking into account experience in different countries, to what extent has such political bargaining been truly “intergovernmental” or involved unilateral elements, and what factors have been responsible for these? • To what extent have the different patterns of intergovernmental financial relations reflected not only the economy but also the character of the society, political culture, values and political institutions of each federation or decentralised union, rather than being derived from normative economic theory? What factors for example have led federations like Canada being the most decentralised in terms of the allocation of pre-transfer revenues and post-transfer expenditures and to the greater emphasis upon the relative financial autonomy of their constituent units? • Finally, in assessing various intergovernmental financial arrangements, what relative weight is to be put upon various criteria? These criteria ­include: economic efficiency; equity objectives for citizens in terms of equality of opportunity, economic security and equality in the provision of public services; autonomy to ensure genuine self-rule in the constituent units; transparency and accountability in the interests of democratic representative government; political stability and minimizing internal conflict; and adaptability to changing circumstances over time. 5 R.L. Watts, The Spending Power in Federal Systems: A Comparative Study (Kingston: Institute of Intergovernmental Relations, Queen’s University, 1999). 6 R. Bird, Federal Finance in Comparative Perspective (Toronto: Canadian Tax Foundation, 1986), at 17–18.

chapter 16

A Post Scriptum to Ron Watts: The Trajectory of Fiscal Federalism Francesco Palermo 1

Rigid Rules, Less Autonomy, More Asymmetry

When this book was first conceived, we asked Ron Watts to write the conclusions. Before the work really started, however, he passed away, and the community of federal scholars lost one of its main sources of inspiration. Fortunately for us, Professor Watts condensed much of his wisdom in several remarkable writings, including the paper that is republished here, from 2007. The content of the paper still applies today. In fact, in these past 10 years, little has changed with regard to the main achievements of comparative analysis of fiscal intergovernmental arrangements, or perhaps, more precisely, the trajectory of the evolution of financial relations, brilliantly sketched by Ron Watts a decade ago, has materialised in nearly every respect. There is only one aspect where Ron raised a question rather than providing an answer. This concerned the pattern of decentralisation of both government revenues and expenditures and of subnational financial autonomy more generally that was noted in most federal and regional countries before the 1980s. This trend continued throughout the 1990s before declining at the beginning of this century, when Watts wrote the piece above. Shortly thereafter—in the aftermath of the financial and economic crisis that hit the Western world beginning in 2008 and that, for several countries, has not yet ended—his open question got a rather clear answer: although there is neither a direct nor an indirect link between the dramatic financial and economic crisis and subnational financial autonomy per se this autonomy was one of the first victims of the measures and reforms adopted in the wake of the crisis. In Europe, new financial regulations were introduced in 2011 primarily by the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (the so-called European Fiscal Compact) and subsequently by national measures to implement and clarify it. Its primary objective is the prevention of future debt-related crises by imposing on the signatory states a duty to implement budget rules into their national legislation. While the Fiscal Compact specifies certain cornerstones of the budget rule, the precise legal © koninklijke brill nv, leiden, ���8 | doi 10.1163/9789004340954_018

A Post Scriptum to Ron Watts

383

wording is left to the states themselves. However, the Treaty recommends that members states amend their constitution or laws by introducing debt-brake rules and balanced-budget provisions, which subnational entities also have to respect. The leading example is Germany, where the constitutional reform of 2009 introduced a constitutional obligation to balance both Federation and Länder budgets (known as “debt brake”) and a subsequent reform in 2017 strongly centralized financial relations. The fact that Germany anticipated this reform that was subsequently “imposed” on all other member states caused a degree of disappointment in several of those countries. The Treaty introduces two main principles that are modelled after the ­German example. First, public expenditure is limited by a (constitutional) structural deficit rule that binds all governmental authorities, whereby the subnational level faces more stringent limits as compared to the national level. Second, a supervisory mechanism is established for an annual review of budgetary management and of the situation at all levels of government. Nearly all other member states have amended their constitutions to introduce similar mechanisms. While following different procedures, all of the federal/regional countries (Austria, Belgium, Italy and Spain) introduced strong financial constraints on subnational units in order to limit overall state deficits and debts. This inevitably reduced the margin of subnational autonomy in all involved countries. A similar dynamic can be observed overseas as well. In the usa, federal aid to state and local governments has been constantly increasing (in 2016, 6.8 per cent more compared to 2015, with similar growth expected in 2017). In 2016, Congress established a Fiscal Control Board over Puerto Rico, which is facing a dramatic financial crisis after 20 years of borrowing to cover its budget shortfalls. The Board is composed of seven unelected businesspeople, who now have total control over Puerto Rico’s economic and fiscal decisions. It is true that some asymmetric financial regulations could be implemented in certain regions in order to increase financial autonomy and responsibility. Examples include South Tyrol and the new financial agreement negotiated with the Italian government in 2014, and Scotland, by means of the Scotland Act 2016, which devolved to Edinburgh the ability to set income tax rates and bands on non-savings and non-dividend income, as well as the right to receive half of the vat raised in Scotland. These are, however, individual and politically motivated exceptions, the result of which is ultimately an increase in the overall asymmetry among regional governance, which has definitely expanded over the last 10 years. These changes have produced at least three main trends in financial relations over the past decade that mark the current stage of the trajectory of fiscal federalism, especially from a constitutional point of view.

384

Palermo

The first one is the much lesser degree, on average, of financial (and, as a consequence, overall) subnational autonomy. Financial constraints and rules to fight sovereign debt apply much more strictly to subnational units than to national levels of government, thus allowing the latter to expand their control over subnational units by supervising their expenditures and controlling their political choices through grant instruments. The German constitutional reform 2017 is a noteworthy example in this respect. The second trend is the increase in overall asymmetry of financial regimes. In a few cases, subnational financial autonomy has increased for individual entities and for specific political reasons, which led to a stronger differentiation in financial autonomy among different countries and, above all, within the same countries. All of this has prompted ever greater need for equalisation mechanisms, which create some (significant) political tension among territories within the same country, as asymmetry tends to produce tensions especially at a time of scarce resources. Third, in strictly legal terms, the regulation of financial relations has become more formalised and entrenched than it was before. The enhancement of constitutional norms—especially due to the European Fiscal Compact—has made such relations more rigid and thus more difficult to negotiate and change. This again means a reduction in the room to manoeuvre for subnational entities. 2

Future Trajectories

Against this background, it is fair to say that, while the 1990s were marked by (excessive) enthusiasm for the regional dimension (especially, but not exclusively, in Europe), the new millennium has witnessed a dramatic slowdown of federal and regional ideas and practices. Many countries have reformed or at least tried to reform their intergovernmental financial relations in the last decade by means of varying and sometimes even conflicting measures. These include the redrafting of regional (and local) spending and taxing powers; stronger intergovernmental transfer and equalisation systems; “tightening deficit, debt or spending rules at all levels of government; reforming local and regional tax systems; merging municipalities, abolishing counties or creating regions”.1 Although these reforms were meant to make intergovernmental financial relations more efficient, more equitable and more stable, many of them faced political resistance, and several 1 H. Blöchliger and C. Vammalle, Reforming Fiscal Federalism and Local Government: Beyond the Zero-Sum Game (Paris: oecd Publishing, 2012), at 1, http://dx.doi.org/10.1787 /9789264119970 -en (accessed 15 April 2017).

A Post Scriptum to Ron Watts

385

were ­watered down, postponed or even abandoned. Negotiations concerning financial regimes have generally become procedurally more cumbersome and politically more uncertain. In addition, the institutional counterpart of such reforms has been facing tremendous difficulties. Very little has changed in the (legislative framework of) existing territorial regimes (federal and regional systems) during the last 15 years. Several reforms have been started, although some did not materialise, while others ended up creating new difficulties in the division of powers, and the few successful ones were those that were limited and rather technical in their outcomes, without changing any structural aspects of the institutional system (such as the federal reforms of 2006, 2009 and 2017 in Germany). In short, institutional issues have remained largely unaddressed during the past decade. At the same time, financial rules have changed in every eu country, often more than once, normally not only at the legislative but also at the constitutional level. Furthermore, these reforms considerably reduced the margins of autonomy of subnational entities overall (with few exceptions), and at times they even sparked secessionist movements. Such asymmetry between frequent and pervasive reforms in the area of ­fiscal federalism and extremely limited, and often boycotted, institutional changes, not only put federal and regional regimes in a permanent state of flux without providing the structural instruments necessary for the balanced regulation of intergovernmental relations. It has also created a further paradox in terms of how to better approach the study of federal and regional systems. In fact, attention was traditionally paid first to the institutional regimes and only then to their financial consequences: the distribution of financial resources, taxation powers, rules on expenditure, financial grants and conditionality, fiscal equalisation among territories and the funding of local governments between the centre and subnational units. In other words, financial regulations had to be consistent with the institutional settings. At present, this ratio is being reversed. Intergovernmental financial relations are becoming centralised, and the question has arisen as to whether institutional settings are consistent with them. As this is less and less the case, institutional reforms are being initiated but very rarely are they producing tangible results, and when this happens, it is because of some technical, rather than structural, reform. Furthermore, and probably also as a consequence of the discrepancy between financial rules and institutional arrangements, federalism has become much less popular in political and academic discourse as compared to just 10 years ago. For many, the very existence of a vertical distribution of powers is responsible for excessive expenditures, for unnecessary complications and in some cases for triggering secession. Surveys indicate that even in consolidated

386

Palermo

federal countries the idea of federalism is less popular than in the past. In the usa, “the absence of federalism on the campaign trail is consistent with the post-1980 recession of federalism as a principle from the political arena. Although state-federal relations remain the vital sinews of the operation of the American federal system of government, intergovernmental relations remain invisible to voters and often misunderstood by elected officials”.2 This trend can be noted nearly everywhere. Does all this mean a gloomy future for financial relations and for fiscal federalism more generally? Drawing such a conclusion would be superficial and probably wrong. In fact, the essential elements of a compound financial arrangement as described by Watts remain the same; what has changed is the distribution of the weight among them. In other words, this is simply a time of growing centralisation of intergovernmental financial relations, which does not mean that such relations have ceased to exist. It is likely that such a ­centralizing moment will be followed by a new, decentralizing one once the deficits of strongly centralised financial regimes become more evident. In the end, intergovernmental financial relations are all about finding the right balance between autonomy and coordination, and, not unlike any other competence issue in a compound system, between self-rule and shared rule. They simply tend to be on the front line of federal dynamics and thus the first to change.

2 J. Kincaid, “State-Federal Relations: Lost on the Campaign Trail”, in, The Book of the States (Lexington: Council of State Governments, 2016), at 30.

Index Africa 241, 261, ch12 Anti-crisis measures 362ff, 373 Assignment Problem Expenditure 120 Revenue 4, 119 Tax 17, 18n29, 102 Asymmetry De facto 277 De jure 271, 277, 367 Financial 275, 371 Preconditions of 276f, 282, 284 Australia 27, 35, 55, 59, 61, 88f, 94f, 153f, 167, 170, 177ff, 187ff, 191, 193, 215, 223f, 230, 237ff, 272, 376 Commonwealth Grants Commission 177, 239 Council for the Australian Federation 240 Intergovernmental Agreement on Federal Financial Relations 238 Loan Council 240f Parliamentary Budget Office 263f Austria 30, 80, 90, 98, 100, 105, 108, 123f, 126, 128, 137, 139, 144, 146ff, 154, 160, 383 Bailout 330ff, 344ff, 362 Federal 72, 123, 251f, 267, 342, 349, 357ff Supranational 328, 330ff, 344ff, 362f Balanced budget 24f, 46, 104f, 114, 232, 252, 259, 304, 323, 350, 360f, 363, 370, 373, 383 Barnett formula 27, 31, 63, 257f, 288f Basque Country 69, 79, 202f, 255, 272, 282ff, 368, 371 Belgium 25, 30, 80f, 97, 99, 108, 123, 137, 139, 144, 148, 165, 202, 275, 284f, 293, 348, 383 Berlin 251f Block grants 31, 62ff, 75, 231, 237, 256ff, 317, 319, 326 Brazil 105, 108, 192, 299 Budgetary autonomy 53f, 56, 73, 135, 249, 318 forecast 63f, 259ff, 264f, 268ff, 341 Bundestreue 37n46, 248

Calman Commission 258 Canada 13, 18, 28, 34ff, 61, 86, 88f, 94, 109, 124, 126, 137, 139, 144, 148, 153f, 162, 167, 170, 174ff, 179, 192f, 195, 200, 202, 205ff, 215, 217, 234ff, 262f, 270ff, 275, 290ff, 295, 350, 352ff, 372, 374f, 378, 381 Parliamentary Budget Officer 262f Revenue Agency 206 Social Union Framework Agreement 237 Cantons 24, 89, 104, 153f, 160f, 340 Catalonia 30, 79, 189, 205, 283f, 371, 373 Centralisation Administrative 170, 209, 213 Fiscal 58, 90, 111, 152, 167, 170, 175, 213, ch14, 386 Chile 187, 189, 192 China 170, 173, 178, 181f, 186f, 189, 195ff, 199ff, 205, 216 Commonwealth of Australia, See Australia Competition, See Tax competition Connection Administrative 97 Principle of 96 Convergence Criteria 329 Correction mechanism 147, 337, 363 Correspondence between expenditures and responsibilities 85, 89, 95f legal principle of 100, 104f principle of 83ff, 88ff, 93ff, 97ff, 107ff rule 84, 93ff, 97ff, 107, 114 Crisis Budget 251, 266 Global financial 60, 224, 260, 273, ch14 Sovereign debt 328ff, 335f, 360 Financial 11, 28, 105f, 111, 113, 305, 338, 382f, See also Anti-crisis measures Debt brake 252f, 337, 339f, 383 Default Risk 72f, 309, 329ff, 357, 362 Delors Report 329 Discipline, See Fiscal Discipline Distribution of finance 91, 99ff, 107 Dualism, See Fiscal Dualism

388 Economic crisis, See crisis Ethiopia 299f, 313ff, 325ff Shako-Mejenger Democratic Unity Party 317 Southern Ethiopia Democratic Movement 317 European economic governance 11, 261, ch13, 362ff Financial Stability Facility (efsf) 334 Financial Stabilisation Mechanism (efsm) 334 Monetary Union (emu) 112, 155, 329ff, 362 Stability Mechanism (esm) 334, 345f, 363, 374 Eurozone 16, 22, 60, 73, 224, 265, 329, 362 Federal-provincial taxation agreements, See intergovernmental ( financial) agreement Finance commission 272 Financial Constitution 2, 19, 21, 23ff, 30, 35ff, 91, 102, 149, 249, 275, 290, 294, 365 Equalisation Law 30, 99, 252 equivalence 114 Finanzausgleichsgesetz, See Financial Equalisation Law Fiscal Common Pool 341, 343 Compact 105, 265, 337, 363, 370, 373, 382, 384 discipline 72f, 105, 122, 224, 235, 260, ch12, 330, 336, 338, 341, 349, 359f, 363, 366, 373 dualism 41, 89, 94f Equalisation Law, See Financial Equalisation Law Equivalence, principle of 93, 96f, 102ff, 111 gap, See Imbalances watchdogs 223f, 260f, 265f, 342 Germany 24, 28, 37, 42, 51ff, 78, 80f, 95, 87f, 100, 105, 108, 112, 123, 126, 137, 142, 144, 153f, 159f, 163, 167, 179, 181, 192, 197ff, 224, 234, 249ff, 266, 270, 272f, 299, 330, 335, 362, 378f, 383, 385 Goods and Services Tax 153f, 162, 167, 177f, 180, 191f, 206ff, 238, 245

Index Grants-in-aid programmes 231, 233, 242f Great Depression 174, 232, 350ff, 374 Imbalances Horizontal 93, 136, 188, 226, 376, 379f Vertical 93, 119, 128, 130, 143, 148, 174, 176, 238, 347, 376, 379f, See also macroeconomic imbalance procedure Independent fiscal institutions 224, 260, 265, 273 India 59, 100, 227, 230, 241ff, 270ff, 299, 371, 378f Finance Commission 58, 242, 272 “Need of assistance” 227, 242 niti Aayog 244f Instrument independence 60f Intergovernmental (financial) agreement 19, 25, 35, 237f, 291, 352f Italy 25, 28, 32, 36, 38, 80f, 99, 105, 108, 112, 123, 126, 128, 130, 139, 144f, 148, 153, 159, 170, 181ff, 187ff, 275, 279ff, 293f, 350, 361, 365ff, 376, 383 Kenya 299f, 320ff African National Union 320 Commission on Revenue Allocation 322 Legacy debt 345 Maastricht criteria, See Convergence criteria Macroeconomic imbalance procedure 336, 339, 363 Medium-term objective (mto) 335 “Mixed rule” 89 Negotiations Bilateral 257, 272, 279, 281 Political 34ff, 237, 239, 259, 271, 282, 293 Nigeria 299, 306ff, 324ff Office of the State Auditor-General 312 Revenue Mobilisation Allocation and Fiscal Commission 310 Northern Ireland 224, 256ff, 269, 271, 288, 290 Good Friday Agreement 259, 288 Stormont House Agreement 259

389

Index Parliamentary Budget Office 261 Property Tax (property rates) 45, 54, 109, 137, 139, 142, 145ff, 152, 154, 159, 166, 193, 197, 209ff, 302ff, 308, 321, 325 Régimen Foral 29, 38, 202f, 255, 282ff, 368 Relations Bilateral 235, 271, 294 Intergovernmental (financial) 18, 21, 25f, 28, 37, 83f, 86, 89, 93ff, 107, 113, 150, 169, ch10, 276, 290, 377ff, 385 Revenue-raising autonomy 308, 316 power 24, 87f, 97ff, 241, 315, 324, 347 “Robin Hood” models 179f redistribution 179f Scotland 27, 31, 38, 61ff, 69f, 77, 147, 224, 256ff, 269, 271, 288ff, 383 Sewel Convention 31 Smith Commission 258 Separation, principle of ch5 Sixpack (legislation) 336f, 362f “Social federalism” 89, 93, 104 Solidarity principle 24, 51, 282, 371 South Africa 86f, 97, 99, 108, 110, 218, 230, 236, 246ff, 265, 269ff, 299ff, 323ff, 379 African National Congress Party 249 Finance and Fiscal Commission 246, 303 Intergovernmental Relations Framework Act 248 Municipal Property Rates Act 302 Parliamentary Budget Office 265 Spain 13, 25, 29, 38, 69, 79ff, 90, 99, 105, 108, 112, 126, 128, 137, 153, 195, 200, 202ff, 224, 253ff, 267ff, 275, 282ff, 293fm, 299, 350, 361, 365ff, 383 Independent Authority for Fiscal Responsibility 267 Special purpose transfers 172f, 318 Spillover effects 61, 65ff, 71, 151, 331f, 343 Stability and Growth Pact 268, 329, 332, 334ff, 362

Statute law 27, 31 Supremacy clause 49 Constitutional 106, 301 Federal Supremacy/ of federal law 44f, 50 Switzerland 18, 24, 28, 78, 89, 95, 100, 104, 126, 137, 142, 144, 148, 153f, 160, 191, 215, 299, 340, 342, 346, 376, 379 Target independence 60f Tax administration agreement 207 assignment problem, See assignment problem base-sharing 18, 131ff, 139, 141, 145 collection agreement 18, 207n56, 237, 354 competition 46, 67ff77, 89, 99, 121f, 133ff, 149, 151ff, 159f, 163ff, 204, 212, 234 rental agreement 90, 291f, 354 raising power 78, 146, 259, 289 Transfer dependency ratio 128 Treaty on Stability, Coordination and Governance in the Economic and Monetary Union, See Fiscal Compact United Kingdom 27, 31, 139, 147, 179, 182, 189, 202, 224, 236, 254ff, 275, 287ff, 328 Office for Budget Responsibility 259, 264, 268, 341 United States (of America) 13, 20, 26, 43ff, 55ff, 95, 104ff, 168, 173, 191ff, 197, 231ff, 342, 346, 350, 352ff, 357ff Advisory Commission on Intergovernmental Relations 234 Congressional Budget and Impoundment Control Act 1974 262 Wales 147, 224, 256ff, 269ff, 288ff Holtham Commission 258 Silk Commission 258f Welsh Revenue Authority 259