International Tax Coordination: An Interdisciplinary Perspective on Virtues and Pitfalls

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

International Tax Coordination

International taxation is a major research topic, and for a field of research at the intersection of so many disciplines there has been surprisingly little done across disciplinary boundaries. This book fills the gap by combining teams from business, economics, information science, law and political science to offer a unique and innovative approach to the issue of international tax coordination. All the chapters are written in collaboration between at least two authors from two different disciplines. This approach offers a rich and nuanced understanding of the many issues of international tax coordination. The book has seven chapters, each one a valuable contribution in itself, beginning with current problems of international taxation and finishing with potential solutions. The essays explore current EU legislation, tax avoidance and tax fraud, as well as double tax agreements, dividend repatriation and hybrid finance and tax planning. Providing methodological answers to the question of how to conduct interdisciplinary research, the book also gives an accessible introduction into research questions and answers that are important in related disciplines for scholars in various areas. This book will be of interest to postgraduates and researchers in the fields of economics, business, informational science, law and political science, as well as to professional accountants and tax lawyers. Martin Zagler is Associate Professor of Economics at WU (Vienna University of Economics and Business), Austria.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Routledge international studies in money and banking

  1 Private Banking in Europe Lynn Bicker   2 Bank Deregulation and Monetary Order George Selgin   3 Money in Islam A study in Islamic political economy Masudul Alam Choudhury   4 The Future of European Financial Centres Kirsten Bindemann   5 Payment Systems in Global Perspective Maxwell J. Fry, Isaak Kilato, Sandra Roger, Krzysztof Senderowicz, David Sheppard, Francisco Solis and John Trundle   6 What Is Money? John Smithin   7 Finance A characteristics approach Edited by David Blake   8 Organisational Change and Retail Finance An ethnographic perspective Richard Harper, Dave Randall and Mark Rouncefield

  9 The History of the Bundesbank Lessons for the European Central Bank Jakob de Haan 10 The Euro A challenge and opportunity for financial markets Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF ) Edited by Michael Artis, Axel Weber and Elizabeth Hennessy 11 Central Banking in Eastern Europe Edited by Nigel Healey and Barry Harrison 12 Money, Credit and Prices Stability Paul Dalziel 13 Monetary Policy, Capital Flows and Exchange Rates Essays in memory of Maxwell Fry Edited by William Allen and David Dickinson 14 Adapting to Financial Globalisation Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF ) Edited by Morten Balling, Eduard H. Hochreiter and Elizabeth Hennessy

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

15 Monetary Macroeconomics A new approach Alvaro Cencini

24 Financial Market Risk Measurement and analysis Cornelis A. Los

16 Monetary Stability in Europe Stefan Collignon

25 Financial Geography A banker’s view Risto Laulajainen

17 Technology and Finance Challenges for financial markets, business strategies and policy makers Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF ) Edited by Morten Balling, Frank Lierman and Andrew Mullineux 18 Monetary Unions Theory, history, public choice Edited by Forrest H. Capie and Geoffrey E. Wood 19 HRM and Occupational Health and Safety Carol Boyd 20 Central Banking Systems Compared The ECB, the pre-­Euro Bundesbank and the Federal Reserve System Emmanuel Apel 21 A History of Monetary Unions John Chown 22 Dollarization Lessons from Europe and the Americas Edited by Louis-­Philippe Rochon and Mario Seccareccia 23 Islamic Economics and Finance: A Glossary, 2nd edition Muhammad Akram Khan

26 Money Doctors The experience of international financial advising, 1850–2000 Edited by Marc Flandreau 27 Exchange Rate Dynamics A new open economy macroeconomics perspective Edited by Jean-­Oliver Hairault and Thepthida Sopraseuth 28 Fixing Financial Crises in the Twenty-­first Century Edited by Andrew G. Haldane 29 Monetary Policy and Unemployment The U.S., Euro-­area and Japan Edited by Willi Semmler 30 Exchange Rates, Capital Flows and Policy Edited by Peter Sinclair, Rebecca Driver and Christoph Thoenissen 31 Great Architects of International Finance The Bretton Woods era Anthony M. Endres 32 The Means to Prosperity Fiscal policy reconsidered Edited by Per Gunnar Berglund and Matias Vernengo

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

33 Competition and Profitability in European Financial Services Strategic, systemic and policy issues Edited by Morten Balling, Frank Lierman and Andy Mullineux 34 Tax Systems and Tax Reforms in South and East Asia Edited by Luigi Bernardi, Angela Fraschini and Parthasarathi Shome 35 Institutional Change in the Payments System and Monetary Policy Edited by Stefan W. Schmitz and Geoffrey E. Wood 36 The Lender of Last Resort Edited by F. H. Capie and G. E. Wood 37 The Structure of Financial Regulation Edited by David G. Mayes and Geoffrey E. Wood 38 Monetary Policy in Central Europe Miroslav Beblavý 39 Money and Payments in Theory and Practice Sergio Rossi 40 Open Market Operations and Financial Markets Edited by David G. Mayes and Jan Toporowski

41 Banking in Central and Eastern Europe, 1980–2006 A comprehensive analysis of banking sector transformation in the former Soviet Union, Czechoslovakia, East Germany, Yugoslavia, Belarus, Bulgaria, Croatia, the Czech Republic, Hungary, Kazakhstan, Poland, Romania, the Russian Federation, Serbia and Montenegro, Slovakia, Ukraine and Uzbekistan. Stephan Barisitz 42 Debt, Risk and Liquidity in Futures Markets Edited by Barry A. Goss 43 The Future of Payment Systems Edited by Stephen Millard, Andrew G. Haldane and Victoria Saporta 44 Credit and Collateral Vania Sena 45 Tax Systems and Tax Reforms in Latin America Edited by Luigi Bernardi, Alberto Barreix, Anna Marenzi and Paola Profeta 46 The Dynamics of Organizational Collapse The case of Barings Bank Helga Drummond 47 International Financial Co-­operation Political economics of compliance with the 1988 Basel Accord Bryce Quillin

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

48 Bank Performance A theoretical and empirical framework for the analysis of profitability, competition and efficiency Jacob Bikker and Jaap W. B. Bos 49 Monetary Growth Theory Money, interest, prices, capital, knowledge and economic structure over time and space Wei-­Bin Zhang 50 Money, Uncertainty and Time Giuseppe Fontana 51 Central Banking, Asset Prices and Financial Fragility Éric Tymoigne 52 Financial Markets and the Macroeconomy Willi Semmler, Peter Flaschel, Carl Chiarella and Reiner Franke 53 Inflation Theory in Economics Welfare, velocity, growth and business cycles Max Gillman 54 Monetary Policy over Fifty Years Heinz Herrman

55 Designing Central Banks David Mayes and Geoffrey Wood 56 Inflation Expectations Peter J. N. Sinclair 57 The New International Monetary System Essays in honour of Alexander Swoboda Edited by Charles Wyplosz 58 Taxation and Gender Equity A comparative analysis of direct and indirect taxes in developing and developed countries Edited by Caren Grown and Imraan Valodia 59 Developing Alternative Frameworks for Explaining Tax Compliance Edited by James Alm, Jorge Martinez-­Vazquez and Benno Torgler 60 International Tax Coordination An interdisciplinary perspective on virtues and pitfalls Edited by Martin Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

International Tax Coordination

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

An interdisciplinary perspective on virtues and pitfalls Edited by Martin Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

First published 2010 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Avenue, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group, an informa business This edition published in the Taylor & Francis e-Library, 2010. To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk. © 2010 selection and editorial matter, Martin Zagler; individual chapters, the contributors All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this has been requested ISBN 0-203-84902-7 Master e-book ISBN

ISBN10: 0-415-56948-6 (hbk) ISBN10: 0-203-84902-7 (ebk) ISBN13: 978-0-415-56948-4 (hbk) ISBN13: 978-0-203-84902-6 (ebk)

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Per Cristiana, l’amore infinito M. Z.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Contents



List of figures List of tables Notes on contributors Acknowledgments List of abbreviations

  1 Introduction: international tax coordination – an interdisciplinary perspective on virtues and pitfalls

xiv xv xvi xx xxi

1

M artin Z a g l e r

1.1  Motivation  1 1.2  Interdisciplinary research in international taxation  2 1.3  Chapter synopsis  10   2 Rethinking tax jurisdictions and relief from international double taxation with regard to developing countries: legal and economic perspectives from Europe and North America

13

P as q u a l e P iston e an d T imot h y J . Goo d sp e e d

2.1  Introduction  13 2.2  Economic views of international tax jurisdiction  14 2.3 The legal rationale of international tax jurisdiction in    the light of European law  21 2.4  Rethinking tax jurisdictions: a possible new model  31   3 How to combat tax evasion in tax havens? A legal and economic analysis of OECD and EU standards on exchange of information in tax matters with a special focus on capital income Di e tmar A i g n e r an d M ic h a e l T u mp e l

3.1  Introduction  37

37

xii   Contents 3.2  OECD standards on exchange of information  39 3.3 EU standards on exchange of information on capital    income  47 3.4  An economic perspective  50 3.5  An economic evaluation  54 3.6  Conclusions  59

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

  4 Double tax avoidance and tax competition for mobile capital

61

M ar k u s L e ibr e c h t an d T h omas R ix e n

4.1  Introduction  61 4.2 Double tax avoidance as the institutional foundation of    tax competition  63 4.3 Empirical evidence on the role of DTAV in tax    competition  71 4.4  Summary and conclusions  78 Appendix  80   5 Intra-­firm dividend policies: evidence and explanation

98

C h ristian B e l l a k an d N a d in e W i e d e rmann - ­O n d r e j

5.1  Introduction  98 5.2  A review of empirical results on dividend smoothing  99 5.3 Dividend smoothing in the intra-­firm context: why   parent companies do or do not smooth dividends of subsidiaries  101 5.4  Summary  112   6 Cross-­border hybrid finance and tax planning: does international tax coordination work? Ewa l d A sc h a u e r , E v a Eb e r h artin g e r an d W o l f g an g P anny

6.1  Introduction  115 6.2  Cross-­border hybrid finance and international taxation  117 6.3  Model  118 6.4  Results  126 6.5  Conclusion and future research  132 Appendix: list of abbreviations  133

115

Contents   xiii   7 Investigating the shift toward a value-­added-type destination-­based cash flow capital income tax (VADCIT)

134

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

K l a u s Hirsc h l e r an d M artin Z a g l e r

7.1  Introduction  134 7.2  The VADCIT model  135 7.3  Legal questions  139 7.4  Analysis of VADCIT  140 7.5  Statistical effects  145 7.6  Fiscal effects  146 7.7  Economic distortions  147 7.8  Conclusions  147   8 The case for and against an EU tax

148

M ic h a e l Lan g an d M artin Z a g l e r

8.1  Introduction  148 8.2  Corporate income tax (CCCTB)  154 8.3  Personal income tax  159 8.4  Modulated VAT  163 8.5  Excise duties on tobacco and alcohol  166 8.6  Energy taxation  169 8.7  Climate charge on aviation  171 8.8  Communication tax  173 8.9  A tax on financial transactions  174 8.10  Transfer of seignorage revenue  175 8.11  Summary  176

Bibliography Index

179 195

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Figures

2.1 Tax rates and FDI stocks, 2002 6.1 Taxation of cross-­border hybrid finance without double tax conventions 6.2 Taxation of cross-­border hybrid finance under double tax conventions 6.3 Decision tree cross-­border hybrid finance without a double tax convention 6.4 Decision tree cross-­border hybrid finance with a double tax convention 6.5 Consequences in the event of a tax audit 6.6 Results depending on the probability of a tax audit in the source state 6.7 Difference in the probability of non-, single, double, or triple taxation, depending on the probability of a tax audit in the source state 6.8 Results depending on the probability of a tax audit in the residence state 6.9 Results depending on the fine factor, γ 6.10 Results depending on the probability of a corporate tax in the source state 7.1 Group’s cross-­border transactions (1) 7.2 Group’s cross-­border transactions (2) 8.1 Monopoly taxation in a closed economy 8.2 Monopoly taxation in an open economy

17 119 120 122 122 123 128 129 129 130 130 143 144 151 152

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Tables

2.1 Comparison of FDI incentives in OECD and developing countries 2.2 US FDI stock abroad by country, 2003 ($ billions) 5.1 Summary of the results of earlier studies of dividend smoothing 6.1 Variables, withholding tax reduced by a DTC 6.2 Variables, withholding tax not reduced by a DTC 6.3 Results for E, depending on the corporate tax rate in SS and RS 6.4 Results for p, depending on the corporate tax rate in SS and RS 6.5 Results for var, depending on the corporate tax rate in SS and RS 7.1 Tax base and tax revenues according to the current legal situation 7.2 Tax base and tax revenues with VADCIT (1) 7.3 Tax base and tax revenues with VADCIT (2) 7.4 Tax bases of P Ltd and S Ltd 7.5 Tax bases of P Ltd and S Ltd (if the profit margin is low) 7.6 Profit and loss account of S Ltd 8.1 Economic evaluation criteria

19 20 100 126 127 131 131 131 136 137 138 143 144 144 154

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Contributors

Dietmar Aigner is Assistant Professor at the Johannes Kepler University Linz. He studied in Linz, Vienna and Graz. His research interests are European tax law, taxation of interest savings, corporate reorganizations and international tax law. He publishes in international journals and is the author of several books on the taxation of capital income, corporate reorganizations, etc. Ewald Aschauer is Assistant Professor at the Department of Finance, Accounting and Statistics at the WU (Vienna University of Economics and Business). His doctoral thesis focuses on business valuation in the case of involuntary separation (2008). His research interests are business valuation, accounting and tax law. Since 2009, he has been working on the International Taxation and Business Finance project of the SFB. Christian Bellak is Associate Professor at the Economics Department of the WU (Vienna University of Economics and Business), where he also earned his doctoral degree in Economics in 1992 and his habilitation in 2003. He was a visiting scholar at the University of Reading (UK), the Helsinki School of Economics, the Hamburg Institute of International Economics, the Leverhulme Centre for Research on Globalisation and Economic Policy (Nottingham, UK) and the Oxford University Centre for Business Taxation (Oxford, UK). His main research interests include foreign direct investment, multinational enterprises, taxation and industrial policy. He has published in international journals such as Management International Review, Multinational Business Review, Applied Economics Letters, the Journal of Business History, Economic Dynamics and Structural Change, Applied Economics Quarterly, the International Journal of Money and Finance, Transnational Corporations, The International Trade Journal, the Oxford Bulletin of Economics and Statistics, the Journal of Economic Surveys, the International Journal of the Economics of Business, Applied Economics, etc.). He is also author of 43 chapters in edited volumes. Eva Eberhartinger is Professor for Tax Management at the WU (Vienna University of Economics and Business). In her research, she focuses on hybrid finance and international taxation and on the relation between commercial

Contributors   xvii

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

and tax accounting. From 2000 to 2002, she was Professor at the University of Münster Institute for Accounting and Business Taxation in Germany. She held visiting positions at McGill University in Montreal, at the University of Illinois at Urbana-­Champaign and at the HEC in Paris, France. In addition to her research activities, she has been the Vice-­Rector for Financial Affairs of WU since January 2006. Tim Goodspeed has been Professor of Economics at Hunter College since 1994. From 1990 to 1994, he was Assistant Professor of Economics at Florida International University. After receiving his Ph.D. in 1986 from the University of Maryland, he worked as an economist at the US Department of the Treasury in the Office of Tax Analysis in both the Individual and the International Tax Divisions from 1986 until 1990. Since 2002, he has been an Adjunct Professor in SIPA at Columbia University and has also been a Visiting Researcher at the Fundación de Estudios de Economía Aplicada (FEDEA) in Madrid and a Visiting Professor at the Universidad Carlos III de Madrid. His research relates to taxation, intergovernmental fiscal relations, state and local public finance, tax competition and international taxation, including the tax treatment of multinational corporations and transfer pricing. He has consulted for the OECD. Klaus Hirschler is Associate Professor at the Department for Finance, Accounting and Statistics, Tax Management group at the WU (Vienna University of Economics and Business), where he earned his doctorate degree in 1996 and his habilitation in 2000. His work experience is mainly in business tax law and accounting. Michael Lang is Professor of Tax Law at the Institute of Austrian and International Tax Law of the WU (Vienna University of Economics and Business) and has a special focus on international, European and business tax law. He has been Head of the Institute since 1998 and Speaker of the Special Research Project (SFB) “International Tax Coordination” since 2003. Michael Lang is also a member of the Global Law Faculty at the New York University (NYU) School of Law, has been Visiting Professor at, among others, Sorbonne, Georgetown University, Peking University and the University of São Paulo, and is currently Chair of the Academic Committee (AC) of the European Association of Tax Law Professors (EATLP). Markus Leibrecht is a senior postdoctoral researcher at the Special Research Project on International Tax Coordination at the WU (Vienna University of Economics and Business). His main research interests include the role of public policies for the attraction of mobile tax bases and the role of the state in international consumption risk sharing. He has published in journals such as Economics of Transition, World Economy, Open Economies Review, Applied Economics, Structural Change and Economic Dynamics, Applied Economics Letters, Empirica, etc. He is also author of two books and several chapters in edited volumes.

xviii   Contributors

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Wolfgang Panny is Professor of Applied Computer Science at the Institute for Information Business at the WU (Vienna University of Economics and Business) and is also affiliated to the research institute for computational methods at the WU. From 1986 to 1987, he held the position of Professor for Information Systems at the University of Bamberg, Germany. His main research and teaching areas are algorithms and data structures, analysis of algorithms, databases, and information retrieval. Since 1998, he has been a member of Working Group 3 of ISO/IEC JTC1/SC32, which is responsible for the further development and standardization of the database language SQL. Pasquale Pistone is EURYI-­Award Professor of Tax Law at the Institute of Austrian and International Tax Law of the WU (Vienna University of Economics and Business). He has a special focus on international and European tax law. Since 2005, he has also been Professor of Tax Law at the University of Salerno. He is principal investigator of the Special Research Project (SFB), his topic of investigation being “Tax Coordination and Third Countries”. Pasquale Pistone is a faculty member of the Postgraduate Program in International Tax Law at the Vienna University of Economics and Business Administration. Thomas Rixen, political scientist and economist, joined the scientific staff of the research unit “Transnational Conflicts and International Institutions” at the Wissenschaftszentrum Berlin (WZB) in November 2007. He studied in Bonn, Hamburg, Paris and Ann Arbor, Michigan. Prior to his engagement at the WZB, he was a researcher with the Collaborative Research Center 597, “Transformations of the State”, at the Jacobs University of Bremen and a scientific staff member in the office of a Member of the German Federal Parliament (Bundestag). His research focuses on international political economy and the theory of institutions; his current research deals, in particular, with institutional (non-)change in international taxation. Michael Tumpel is Professor and head of the Department of Tax Management at the Johannes Kepler University Linz. He studied in Vienna and was a visiting researcher at New York University (NYU). His special research interests are VAT, European tax law, corporate reorganizations and international tax law. He publishes in international journals and is the author of several books on European and international tax law. Nadine Wiedermann-­Ondrej works as a financial officer for the Austrian Ministry of Finance. From 2002 to 2008, she was Research Assistant at the Tax Management Department of the WU (Vienna University of Economics and Business) and earned her doctoral degree in 2006. Her research interests are hybrid financial instruments, multinational corporations and international tax law. Martin Zagler is Associate Professor of Economics at the WU (Vienna University of Economics and Business). He studied in Linz, Aix-­en-Provence,

Contributors   xix

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Vienna and Florence and was visiting professor at the University of Rome 1 (La Sapienza), Harvard University, the Catholic University of Piacenza, the Free University of Bozen/Bolzano and the University of Sassari. He held fellowships at the European University Institute, University College London and the Robert Schuman Center for Advanced Studies in Florence. His research interests are macroeconomics, economic policy and public finance. He publishes in learned journals and is the author of two books on economic growth with Palgrave Macmillan and a Springer textbook on public finance.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Acknowledgments

This book is a result of the Special Research Group (SFB) on International Tax Coordination, funded by the Austrian Science Fund (FWF ) and hosted at the WU (Vienna University of Economics and Business). We are very grateful for the substantial financial support from the FWF, and to the WU for providing a stimulating and supportive environment. An interdisciplinary endeavor of this size would not have been possible without this assistance. It gave the authors the possibility of discussing and interacting across disciplinary boundaries, to find a common language and to initiate joint research projects. You can find out more about the SFB on our website at www.sfb-­itc.at.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Abbreviations

ACE ACP AktG ALS APA AStG BEA BIT BRIC CBIT CCCTB CDU CEN CFC CIC CIN CIT DC DTAV DTC DTT ECB ECJ EEA EPA EUCIT FDI GDP GewStG GNI HIC ICAO

allowance for corporate equity African, Caribbean and Pacific Group of States Aktiengesetz (corporate law) arm’s-length standard advanced pricing agreement Außensteuergesetz (international transaction tax law) Bureau of Economic Analysis (United States) bilateral investment treaty Brazil, Russia, India and China comprehensive business income tax common consolidated corporate tax base Christlich Demokratischen Union Deutschlands (Christian Democratic Union (Germany)) capital export neutrality controlled foreign company/companies capital-­importing country capital import neutrality corporate income tax developing country double tax avoidance double tax convention double tax treaty European Central Bank European Court of Justice European Economic Area European Protection Agreement European Union corporate income tax foreign direct investment gross domestic product trade tax law (Gewerbesteuergesetz) gross national income highly industrialized country International Civil Aviation Organization

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

xxii   Abbreviations ILADT Instituto Latinoamericano de Derecho Tributario KStG corporate income tax law (Körperschaftsteuergesetz) LATC Latin American Tax Convention (on income) LFTTD Linked/Longitudinal Firm Trade Transaction Database MC Model Convention MFN most favored nation MNE multinational enterprise NIC newly industrializing country NN national neutrality OECD MTC Model Tax Convention on Income and Capital of the Organisation for Economic Co-­operation and Development OLS ordinary least squares PE permanent establishment; permanent enterprise ROCE return on capital employed RS residence state SS source state TIEA Tax Information Exchange Agreement 2SLS two-­stage least squares VADCIT value-­added-type destination-­based cash flow capital income tax WTO World Trade Organization

1 Introduction International tax coordination – an interdisciplinary perspective on virtues and pitfalls Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Martin Zagler 1.1  Motivation International taxation is a major research topic. Indeed, scholars from business, economics, law, history, political science, sociology, psychology and a number of related fields have extensively investigated the issue of international taxation. For a field of research that is at the intersection of so many disciplines, it is surprising that very little has been done across disciplinary boundaries. This book attempts to fill the gap between some of these disciplines by presenting interdisciplinary research on international tax coordination. We believe that this book can contribute to the advancement of knowledge in each discipline through its interdisciplinary approach. Issues of international taxation are a growing concern in a world of integrating economies such as the enlarged European Union, and these issues have undergone an evolutionary process themselves. When the European Union was set up, we saw the Neumark Report’s vision of a single economic area where only economic parameters would be taken into account in individual allocative decisions. Differences in taxation are an obstacle to the optimal allocation of resources within the Union, which is why the harmonization of tax systems was the choice for international taxation at that time. Such a positive approach toward harmonization is still found in recent documents, mainly driven by anticipated effects such as a significant reduction of tax compliance costs and the “neutrality” and “efficiency” of tax regimes. The tax harmonization view interprets differences in taxation as a hindrance to the optimal allocation of resources. The opposing idea of tax competition has been extensively discussed by economists (Tiebout 1956; McLure 2001). However, European institutions started to consider this line of argument only in the early 1990s. In this concept, the idea of competition is extended to sovereign players that offer individuals or firms different ranges of tax costs. With both approaches – harmonization and competition – having potential benefits from an economic point of view, international tax coordination is seen to cover the whole range between tax harmonization and tax competition. Tax coordination differs from tax harmonization by allowing a certain degree of national freedom in designing the tax regime. International tax coordination is

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

2   M. Zagler perceived to increase transparency, thus making competition more visible, which may bring markets toward equilibrium and may improve economic welfare. In this aspect, tax coordination may be superior to tax competition. However, tax coordination may also be superior to tax harmonization, in particular if different countries have different preferences for the size of the public sector. In this case, a harmonized level of taxation would induce one country to have a public sector larger than desired and the other to cut public expenditures below its social optimum. This book focuses on various issues in the area of international tax coordination, but has a clear research question: What degree of international tax coordination within the European Union and in relation to third countries is suitable for solving problems arising from national tax sovereignty, and what problems may arise from tax coordination itself? Measures of tax coordination fill the gap between tax harmonization and tax competition. The contributions in this volume will combine interdisciplinary teams from business, economics, information science, law and political science to offer a unique and innovative interdisciplinary approach on the issue of international tax coordination. All the chapters are written in collaboration between at least two authors from two different disciplines. The distinguishing feature of the book is its clear interdisciplinary approach. All the authors have been collaborating during the past six years in a special research program funded by the Austrian Science Fund. During this period, the authors have come to realize that the essence of many questions in international tax coordination cannot be solved within a single discipline. An interdisciplinary approach offers a strenuous but worthwhile approach to the understanding of many issues of international tax coordination. It must be emphasized that interdisciplinary tax research that goes beyond the mere coexistence of the disciplines that are involved is hardly ever found. In this regard, the book adds considerably to existing literature and gives new insights into international tax coordination. Further, the book contributes novel aspects to the understanding of international taxation. It provides methodological answers to the question of how to conduct interdisciplinary research, and introduces research questions and provides answers to those that are important to related disciplines.

1.2  Interdisciplinary research in international taxation Many current researchers have emphasized that the production of interdisciplinary knowledge is a major issue that needs to be considered in today’s “knowledge-­based society” (Gibbons et al. 1994). The debate on interdisciplinarity is still young and the process of interdisciplinary research is still being developed. However, a growing community of interdisciplinary researchers stimulates vivid debate aimed at developing means to deal with ensuing challenges in research. Nowadays, interdisciplinary research is strongly promoted as a desirable direction in the development of sciences, in terms of both teaching and research.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Introduction   3 This is true not only in the natural sciences but also in the social sciences, which are said to benefit tremendously from interdisciplinary approaches. These approaches may help overcome artificial boundaries and improve the overall quality of social science research. As Popper (1963) points out, “We are not students of some subject matter, but students of problems. And problems may cut right across the borders of any subject matter or discipline.” Sciences and society are confronted with problems that cannot be fully comprehended by a simple deterministic disciplinary view. The main issues nowadays have a high level of complexity that requires insights from multiple disciplines. Interdisciplinary research and education are mainly and traditionally inspired by the drive to solve complex questions and problems generated by scientific curiosity or by society. They guide researchers from different disciplines to meet and work at the interfaces and frontiers of those disciplines, and even cross such frontiers to create new disciplines. Obviously, not all subjects lend themselves to interdisciplinary research. International taxation, however, has always been a topic of interest in more than one discipline. Because taxes are introduced by law, legal scholars certainly have an interest in the issue. Taxes influence business decisions, so researchers in business administration examine it. Moreover, in turn, taxes have an impact on the economy. Hence, economists, too, have an interest in the subject. The growth of scientific and technical knowledge in recent decades has prompted researchers from various disciplines to address complex problems that require deep knowledge from different perspectives. The results obtained so far confirm the high level of productivity and effectiveness of research teams composed of partners with diverse expertise. On the one hand, the attempt to solve complex problems on the one hand may be considered the main factor that has led to the growth in the application of interdisciplinary research methods. On the other hand, increasing social interest in such research may be attributed to the following factors: •

• • •

It is argued that research performed in an interdisciplinary context is likely to be more creative and innovative. Mixing people with different backgrounds and/or ideas from different fields is likely to generate “breakthrough” and “innovative” research results. Second, interdisciplinary research has been emphasized as being more likely to lead to applicable results because it is compatible with problem-­solving approaches (Foray and Gibbons 1996). Moreover, there is concern that disciplinary research has become too narrowly specialized, leading to diminishing returns for this traditional research approach. Furthermore, it is now increasingly recognized that the prevalent tendency in most disciplines for narrow, deep specialization tends to make research less relevant to laypeople or to society, fosters imperialism rooted in ideological thinking and limits the exchange of ideas across disciplines, ultimately impeding the progress of science.

4   M. Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

For all these reasons, interdisciplinary research has become a real subject of interest for science policy in recent times. The following statement explains why interdisciplinary research is highly relevant in the current scientific debate: Interdisciplinary research is an appropriate form of research when searching for science based solutions to problems in the lifeworld with a high degree of complexity in terms of factual uncertainties, value loads and societal stakes. Through bridging different scientific and social knowledge components, it can significantly improve the quality, acceptance and sustainability of such solutions. However, deliberation about facts, practices and values are ongoing when bringing results to fruition in the lifeworld as well as in scientific communities. (Wiesmann et al. 2008) Interdisciplinary research can be one of the most productive and inspiring human pursuits in the sense that it provides the connections that can lead to new knowledge. The potential of conducting research this way is very useful in terms of new discoveries and technologies. 1.2.1  A brief historical overview Interest in interdisciplinary research has not been always so vivid. In ancient times, education and philosophy were interdisciplinary in the sense that philosophers did not accept any boundaries or limitations to the validity of their truths. Aristotle was the first to introduce a division of knowledge by splitting it into the theoretical and the practical, thus balancing “pure” thinking (rhetoric, logic, mathematics, ethics, etc.) with the observation of nature (physics, astronomy, etc.). This first division of “philosophical” knowledge paved the way for the further division of knowledge into more and more specialized fields of science. It seemed that the unity of knowledge had been irrevocably lost. In the early twentieth century, a new philosophical school of thought under the name “logical positivism” emerged in Germany and Austria. Some of the logical positivists were committed to the idea of a “unified science” based on the development of a universal scientific language. Science was considered to be an interdisciplinary enterprise. This led to the question of how the knowledge of the natural sciences could have been used to further understanding in the social sciences. For the positivists, all the academic disciplines share the same universal scientific rationality. Another development in modern philosophy of science has been the rise of the “descriptive history of science”. Postmodernists who felt it important to divide scientific knowledge for the purpose of serving the interest of their respective knowledge communities placed serious limits on the possibility of interdisciplinarity. In the late 1980s, a new school of thought, so-­called social epistemology, claimed that the disciplines are efficient in producing new knowledge and in maintaining rigor of inquiry, which would be lost in an interdisci-

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Introduction   5 plinary science (Bridges 2006). All these theories argued that the rules of testing and validating hypothesis are different among disciplines and that there are obvious constraints in doing disciplinary synthesis. By contrast, in the social sciences the new school of the so-­called interdisciplinarians (Newell 2001; Thompson Klein 2004) focuses on the problem that academic work generally happens within narrow and arbitrary or artificial disciplinary boundaries, which sometimes prevents academics from seeing the close connections between different phenomena and the different disciplines. Moreover, they claim that complex discipline-­transgressing phenomena cannot be understood adequately by using reductionist disciplinary approaches. According to this new orthodoxy, scientists should aim to develop fruitful relationships with disciplines other than their own and perhaps should even transcend disciplinary thinking altogether. The natural sciences were the first to introduce the concept of complexity to shape their methodology (Prigogine 1980; Casti 1994). The idea was that knowledge is no longer simple, and this raises the need to combine different kinds of knowledge or different points of view to describe a phenomenon. Different disciplines have to work together to outline objects in a scientific way. In addition, a phenomenon has to be seen as a whole whose parts work together. Despite this apparent mutual interest, joint interdisciplinary research is rare in international taxation. In Central Europe, law and economics were historically thought of as Staatswissenschaften (state sciences). The division into Nationalökonomie (national economics) and law came before interdisciplinarity emerged as a modern approach to conduct science. By that time, public economists were following the Anglo-­Saxon tradition in economics, and ignored most of legal science. Only recently have the disciplines moved closer to each other. This book is certainly a step in this (in our view, correct) direction. 1.2.2  Definition and measurement While there is agreement that interdisciplinarity is in principle desirable, there is still a good deal of disagreement on what interdisciplinarity means in actual practice and on the identification of a universally accepted definition of the word “interdisciplinarity”. There are difficulties in including the diverse range of activities described under the heading of “interdisciplinary research” in a single definition. Understanding the real concept of interdisciplinary research remains difficult even if a number of recent studies (e.g. Sanz-­Menendez et al., 2001) have sought to improve its comprehensibility. The term “interdisciplinarity” is loosely and insufficiently defined such that it is made almost meaningless. The most general definition of “interdisciplinarity” found in the literature is “any form of dialogue or interaction between two or more disciplines” (Moran 2001). This is certainly too vague a definition but at least it captures what the word “interdisciplinarity” brings to mind for most people. It means crossing disciplinary boundaries, which could be considered a starting point to fix the basic idea.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

6   M. Zagler The main problem with the notion of “interdisciplinarity” is that people who use it do not explicitly state what they understand to lie within a particular discipline, when a disciplinary boundary is being crossed and what the consequences are. Following this approach means that any useful definition of “interdisciplinarity” would require a workable definition of the academic disciplines first – which is certainly not easy. The term “academic discipline” incorporates many elements. At the same time, it has also become a technical term for the organization of learning and the systematic production of new knowledge. Often disciplines are identified with taught subjects, but, clearly, not every subject taught at the university can be called a discipline. There is more to a discipline than the fact that it is a subject taught in an academic setting. In fact, there is a whole list of criteria and characteristics that indicate whether a subject is indeed a distinct discipline. Some try to define “interdisciplinarity” without the use of the concept of disciplines. Among the numerous definitions found in the literature, the following are the ones that in our view seem best to address the complexity of the issue: Interdisciplinary research (IDR) is a mode of research by teams or individuals that integrates information, data, techniques, tools, perspectives, concepts, and/or theories from two or more disciplines or bodies of specialized knowledge to advance fundamental understanding or to solve problems whose solutions are beyond the scope of a single discipline or field of research practice. (National Academy of Sciences et al. 2004) Interdisciplinary research is research that includes cooperation within the scientific community and a debate between research and the society at large. Interdisciplinary research therefore transgresses boundaries between scientific disciplines and between science and other societal fields and includes deliberation about facts, practices and values. (Wiesmann et al. 2008) These definitions make it clear that research is truly interdisciplinary only when it is not merely a combination of two disciplines to create one product but is rather an integration and synthesis of different ideas and methods. Along with the difficulties in defining what interdisciplinary research is, interdisciplinarity is made up of a range of very different concepts such as inter- or supradisciplinarity, multidisciplinarity, crossdisciplinarity, transdisciplinarity and megadisciplinarity, which are often talked about as though they are just one concept. Some authors (Tabak 2004; Krishnan 2009) have attempted to clarify these different concepts: •

Inter- or supradisciplinarity: this means that researchers work together on common questions or problems. This interaction may create a new research

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Introduction   7

•

•

•

field or scientific discipline. In the inter- or supradisciplinary approach, disciplines collaborate on developing some common perspective. Several chapters of this book fall into this category. A good example is the contribution by Aschauer, Eberhartinger and Panny (Chapter 6), who use probability trees to analyze the complexity of international taxation in hybrid finance. Another example is Hirschler and Zagler (Chapter 7), who use accounting techniques to discuss the impact of introducing a new corporate income tax. Finally, Bellak and Wiedermann-­Ondrej (Chapter 5) look at dividend payments from both an economic and a tax accounting point of view within a common framework. Multidisciplinarity is probably the most common approach in collaborative research and does require some limited collaboration of researchers from different disciplines. It means that the researchers work together on particular common problems. In multidisciplinary research, a team of researchers work toward a common aim or problem, but each discipline represented works independently or in sequence. The final product of this way of working normally consists of a compilation of disciplinary research on a common theme or object. Alternatively, this collaboration may result in an integrated research product that synthesizes the disciplinary perspectives into a coherent picture. Several chapters fall into this category. Pistone and Goodspeed (Chapter 2) analyze international development from an economic and legal perspective, Leibrecht and Rixen (Chapter 4) investigate double tax agreements from both an economic and a political science approach, and Lang and Zagler (Chapter 8) analyze economic and legal (dis)advantages of EU taxes. Crossdisciplinarity is in some sense the easiest form of interdisciplinarity. It does not actually require the collaboration of researchers with various disciplinary backgrounds, or in fact any collaborative research effort. The general idea in practicing crossdisciplinarity is to look at what other disciplines have to say about a particular phenomenon or object within the scope of their own discipline, or, alternatively, to apply concepts and methods of their own discipline to phenomena or objects of other disciplines. The researcher then applies instruments belonging to other disciplines without actively collaborating with other researchers. No contribution in this volume strictly falls into this category. While Aigner and Tumpel (Chapter 3) are clearly interdisciplinary between law and business, their chapter also uses economic concepts extensively. Transdisciplinarity involves seeking research collaborators outside of the academic or university context and producing research that is primarily for the purpose of application, for example the development of new technology or the formulation of policy. This phenomenon is still more common in the natural sciences, engineering and computer science. Even so, there are opportunities for academics from the social sciences to venture outside the university and the academic funding system and do contract research for the public, private or non-­profit sectors.

8   M. Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

•

Megadisciplinarity: Each discipline has increasingly deepened its scope during the twentieth century. This phenomenon is so widespread that there is a large overlap among the different disciplines. Traditional disciplinary divisions do not seem to make much sense anymore and it is now common for researchers to cross disciplinary boundaries to collaborate with members of other disciplines who share the same specialization. This phenomenon has fueled the debate about the opportunity to merge many of the current disciplines into a few super- or megadisciplines that could more effectively organize the accumulated knowledge of existing disciplines. The few remaining superdisciplines could not only integrate several social sciences into a unified social science discipline, but also bridge the traditional division between the natural and the social sciences.

1.2.3  How to practice interdisciplinary research We have stated how important it is nowadays to adopt an interdisciplinary approach in research. At the same time, practicing interdisciplinarity is fraught with difficulties. When one tries to define “interdisciplinarity”, the ambiguity of the definition itself has an impact on the understanding of the analytical approaches adopted by interdisciplinarians. The overlap of different disciplines leads to the increased use of the interdisciplinary research approach. Because of overlapping subject areas, disciplines are now identified through the methodology they apply to certain topics or research fields, rather than through the topics or research fields themselves. A very clear example is the field of international taxation, which lies at the crossroads of several disciplines. For this reason, several interdisciplinary centers have emerged. This book is special as it is the first effort to combine interdisciplinary research from three disciplines: business, economics and law. However, it is also more common for scientists belonging to different disciplines actually to use the same methodologies to analyze a topic. This could indicate that the distinction among disciplines is artificial and an accident of history rather than the result of any scientifically substantial difference. In this volume, we present papers by accountants using econometrics, and economists who use accounting case studies. According to Gibbons et al. (1994), the production of knowledge is organized outside disciplinary and academic contexts and is focused on creating knowledge directly related to its application. Traditional discipline-­specific knowledge production within academic departments is becoming increasingly obsolete and less relevant for society. The many possibilities for understanding disciplines and disciplinarity lead to rather different conclusions on the value of disciplines and the practicality of interdisciplinarity. The starting point on whether to use a traditional disciplinary approach or an interdisciplinary one greatly depends on the problem that the researchers are aiming to solve. One way of conducting interdisciplinary research is by mixing methods to ask distinctive and intersecting questions.

Introduction   9

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

In order to run an interdisciplinary research project, team members should adopt the following steps: • • • • • • • •

Define common project goals. Define a project design that works. Find a common language. Build trust. Overcome epistemological barriers. Develop interdisciplinary theory. Engage in interpersonal communication. Facilitate integration.

It is important to clarify the specific characteristics of interdisciplinarity to better understand how this type of research can be conducted in practice. •

• •

Adopting an interdisciplinary approach involves recognizing that the social world and the issues to be studied are multidimensional, and that different dimensions might exist and coexist, rather than being integrated within a single plane or dimension. Moreover, as interdisciplinary research analyzes complex empirical questions, it aims to deal with problems better and investigates how existing practices can be changed or improved. Interdisciplinary research is committed to developing scientific innovation. It is driven by the tension between specialization in interdisciplinary methods and triggers the transformation of disciplines. Therefore, interdisciplinary researchers form a mixed college of peers.

From a practical perspective, this type of approach is challenging because by definition it pushes knowledge and practice to the boundaries of the social science philosophy. Practicing interdisciplinary research demands certain requirements due to its specific nature. This type of approach requires: •

• • •

Considerable skill and commitment from researchers and teams, who need to have the capacity and inclination to see beyond disciplinary, epistemological and ontological distinctions, without simply seeking to critique others from a single perspective, or to subsume all other perspectives into one. Taking intellectual risks and being interested in alternative approaches without fear of reprisal or contradiction by other members of the working group. Respecting the distinctive nature of the different approaches so that it is allowed to flourish, rather than by reducing all to a bland lowest common denominator which is assumed to be “interdisciplinarity”. Creative tension between the different disciplines. This means that the final output of the research should not be an integrated account or explanation of

10   M. Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

•

•

whatever is being researched, or a series of parallel accounts, but a multinodal explanation based on the dynamic relation of more than one way of seeing and researching. This requires that researchers include the different ways of asking and answering questions. A research design that is basically a recursive and negotiation process (structuring, searching for solutions, bringing results to fruition, etc.). Interdisciplinarity implies that the precise nature of a problem to be addressed and solved is not predetermined and needs to be defined cooperatively by actors from science and beyond. During the research process, bodies of knowledge have to be integrated to produce different types of knowledge. Interdisciplinarity thus changes the traditional sequence leading from scientific insight to action. The potential to stimulate innovation in participating disciplines, which is intrinsically implied by interdisciplinarity, is brought very fruitful. This requires an emerging college of peers who are able to link disciplinary and interdisciplinary specialization.

Practicing interdisciplinary research is complex work that implies a number of important critical aspects. There are some recurrent roadblocks that researchers have to face when working with an interdisciplinary approach. True collaboration and participation of all the members of an interdisciplinary group is fundamental for the success of this approach. True collaboration does not imply the denying of a diversity of goals, values and expectations among different scientists. Good and concrete interdisciplinary practice must be integrated through its scientific foundations and its scientific recognition. Such efforts must go beyond systematizing interdisciplinary research procedures and aim at theoretical, methodological and topical development and innovation at the interface with participating disciplines to benefit both sides. Facing these challenges requires the development of extended peer networks and other collaborative networks that bridge interdisciplinary and disciplinary reference and quality control systems.

1.3  Chapter synopsis This book attempts to fill the gap between several disciplines by presenting interdisciplinary research on international tax coordination, in particular its virtues and pitfalls. This book is a collection of seven chapters, not counting this introduction, that in themselves are novel and innovative in international taxation. The book begins with the current problems of international taxation and ends with potential solutions. Chapters 2 and 3 show how current well-­intended efforts of tax coordination can exhibit unwanted side effects. Pasquale Pistone and Tim Goodspeed in Chapter 2 show how economic development could be fostered by altering allocation rules and introducing standards of legal protection equivalent to the ones applied within the European Union. Dietmar Aigner and Michael Tumpel in Chapter 3 look at measures to combat tax avoidance and tax

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Introduction   11 fraud in tax havens. They analyze EU and OECD measures, with a special focus on capital income. Three issues of coordination are also investigated. First, Markus Leibrecht and Thomas Rixen in Chapter 4 investigate double tax agreements and their effect on decisions by multinational enterprises concerning where to locate. Christian Bellak and Nadine Wiedermann-­Ondrej in Chapter 5 consider dividend repatriation, and Ewald Aschauer, Eva Eberhartinger and Wolfgang Panny in Chapter 6 analyze hybrid finance and tax planning. All these three chapters investigate concerns about current efforts of tax coordination, which may have suboptimal side effects for capital flows, dividend payments and tax planning compliance costs. The final two chapters investigate potential future directions of tax coordination. First, Klaus Hirschler and Martin Zagler in Chapter 7 look at a different way to tax corporate income with a value added-­type destination-­based cash flow capital income tax. The last chapter, by Michael Lang and Martin Zagler, analyzes the logical next step in international taxation, supranational taxes. The interdisciplinary approaches chosen by the authors provide a fresh perspective on these issues. We believe that we can contribute to the advancement of science in each discipline through this interdisciplinary approach. The first main chapter in this volume, by Pistone and Goodspeed, uses economic arguments to investigate potential gains from changes in legislation. The authors ask how European tax law can contribute to the fostering of economic growth in developing countries. The combination of researchers from law and economics allows the analysis to go into the legal details of the credit and exemption method, which were previously beyond economic consideration. The authors go further by applying sound economic principles in investigating legal issues. The merit of this chapter is that it both communicates legal issues to economists and sensitizes lawyers to the economic consequences of legal aspects. Chapter 3, by Aigner and Tumpel, looks at the implications of information exchange for tax fraud. The authors search for collective or individual solutions appropriate for national governments in order to combat tax evasion and tax fraud in tax havens. They find that economic channels that act through detection and penalties do not prevent tax evasion, as legal loopholes allow agents to circumvent taxation. In fact, moral suasion is more likely to prevent tax evasion. Once again, interdisciplinarity between economic, legal and business studies has generated a result that would not have been derived if the problem had been considered only from the viewpoint of one scientific field. In Chapter 4, Markus Leibrecht and Thomas Rixen use institutional analysis typically adopted in the political sciences to discuss both the legal and the economic literature on double tax agreements. The authors manage to establish a point that has not yet been raised in the economic and legal literature: that double tax agreements may foster tax competition instead of reducing it, because they provide a legal framework for tax planning to investigate the intended and unintended loopholes of national tax codes.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

12   M. Zagler In Chapter 5, Bellak and Wiedermann-­Ondrej examine what happens after capital has moved across borders. They provide empirical evidence on dividend policies of multinational firms and discuss the determinants of dividend smoothing within the firm. The analysis is unique as it is the first attempt to look at dividend payments from an economic and accounting point of view. Interestingly, it is not so much the relationship between the parent and the affiliate, as suggested by the principal–agent theory but, rather, tax accounting that can shed light on dividend streams, particularly when dividends are subject to high withholding taxes. The contribution by Aschauer, Eberhartinger and Panny (Chapter 6) uses mathematical tools of probability trees to analyze the uncertainty surrounding the taxation of hybrid instruments. The authors show that the taxation of hybrid instruments of finance has a major effect on yields. They also demonstrate that double tax agreements can substantially reduce the yield volatility associated with hybrid instruments. Interestingly, the authors find that the presence of double tax agreements also reduces the expected tax burden. The innovative crossdisciplinary methodology adopted in this chapter will certainly have an impact on the tax accounting literature as well as the economic literature on double tax agreements. Chapter 7, by Hirschler and Zagler, applies methods of accounting science to scrutinize an economic concept. The authors use a balance sheet case study to analyze the impact of the introduction of a value added-­type destination-­based cash flow capital income tax. The concept itself has stirred some interest in the debate about the reform of corporate income tax. However, the authors convincingly demonstrate a series of pitfalls that render this particular tax infeasible. Once again, the use of methods of one discipline, in this case tax accounting, can shed new light on the question raised in another discipline, here economics, and thus render novel insights. The final chapter of this book looks at the final step toward tax coordination in Europe, the introduction of taxes collected by the European Union itself. Currently, the Union depends on transfers from its Member States to finance its budget. A proposal by the European Union to introduce taxes itself could change this. This is the first attempt in a single piece of writing to investigate both the legal and the economic virtues and pitfalls of the introduction of EU taxes. The authors conclude that only VAT and ecological taxes are qualify as candidates for the European Union’s own taxes.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

2 Rethinking tax jurisdictions and relief from international double taxation with regard to developing countries Legal and economic perspectives from Europe and North America Pasquale Pistone and Timothy J. Goodspeed 2.1  Introduction Countries have historically claimed the right to tax all income generated within their boundaries. With respect to the income generated by domestic factors (labor or capital) outside of their borders, countries have followed one of two approaches. One approach is to effectively cede jurisdiction to the foreign country by exempting such income from domestic taxation; this is termed territorial taxation. The second approach is to claim jurisdiction and subject such income to domestic tax as well as to give a credit for foreign taxes paid (up to what would have been paid if the income had been generated domestically). The credit method maintains domestic tax jurisdiction but cedes the revenue generated by the foreign tax rate to the foreign country. If the tax obligation owed to the foreign country is less than the obligation that would have been owed to the home country, then the home country collects the difference under the tax credit method, whereas no home tax is owed under the territorial method. Thus, the method of relief of international double taxation is central to the effective tax jurisdiction claimed by the home country. Furthermore, the specifics of double taxation relief also often depend on bilateral tax treaties, and sometimes on multilateral tax agreements or legal rulings (as in the case of the European Union, for example). The credit method allows for taxation in the country of source, as well as allowing taxation of the overall income or capital in the taxpayer’s country of residence (Cnossen 1987). The latter characteristic is argued to be important in ensuring capital export neutrality and consistency in a nation’s application of the ability-­to-pay principle (Surrey 1965). Capital export neutrality implies efficiency in the worldwide allocation of capital; however, other sorts of neutralities and efficiency criteria have come to the fore in recent discussions. With respect to the ability-­to-pay criterion, it is argued that without additional tax levied by the home country, those with high ability to pay may evade correspondingly high tax by locating themselves in low-­tax jurisdictions. However, this rationale is not universally accepted. High-­tax jurisdictions applying the exemption

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

14   P. Pistone and T. J. Goodspeed method implicitly accept any inconsistency in the ability-­to-pay principle embodied in their system. The ability-­to-pay view has also been questioned in the legal literature (Fleming et al. 2001), and additional criticism can be raised in the light of European law as well as in the context of economic globalization.1 Although taxation is important for the location of foreign direct investment (FDI) in developed countries, it is probably not the most important factor influencing FDI in developing countries.2 More fundamental aspects, such as low corruption and good infrastructure, are probably more important. Nevertheless, the international tax policies of one country have an impact on the revenue and international tax policies of other countries, and this external impact is particularly acute (at least in a political sense) when one country is developed and the other is developing. Given that developed countries use various tools, including direct aid to developing countries, the role of tax jurisdiction is worth rethinking (Barker 2007: 365; Hines 1998; OECD 2001). In this chapter, we review and reexamine the issue of international tax jurisdiction from economic and legal viewpoints, with particular emphasis on the impact of developed country rules on developing countries. We also consider the impact of European law on the relations with non-­EU Member States (hereafter, “third countries”). The remainder of the chapter is organized as follows. The next section examines the economic views of tax jurisdiction and how it affects the relationships between developed and developing countries. This section also examines how the recent US proposals to change the way in which the United States taxes foreign income might impact developing countries. Surprisingly, two recent proposals head in exactly opposite directions: the 2005 Bush Tax Reform Commission recommended moving toward a territorial system, while the recent Obama administration proposals appear to tighten worldwide taxation. The following section presents a legal view of tax jurisdiction, with particular emphasis on the way in which European law and recent European Court cases affect the developing country’s relationships with Europe. The final section argues that a move toward territorial taxation with respect to developing countries would be consistent with the recent European court rulings, would give more autonomy in tax matters to developing countries and could be seen as an alternative development tool. However, the biggest problems would involve cross-­border flows of passive income and maintaining domestic ability-­to-pay goals because of the difficulty of taxing wealthy taxpayers, and we have made several suggestions concerning ways to minimize this problem.

2.2  Economic views of international tax jurisdiction 2.2.1  Tax jurisdiction and efficiency Economic views of tax jurisdiction tend to center around questions of world efficiency (how can tax jurisdictions be defined to maximize world income?) versus national goals (how can countries define tax systems to maximize national income?). Initial studies concluded that worldwide income could be maximized

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Rethinking tax jurisdictions   15 by taxing worldwide income and allowing a credit for foreign taxes, while national income could be maximized with a deduction system (Richman 1963; Hines and Gordon 2002). These results follow from highly stylized models. For instance, the advantages of a worldwide system are most easily thought of in the context of a world of two countries where one is a capital exporter and the other is a capital importer. In this context, the worldwide system with an (unlimited) foreign tax credit will lead to the efficient allocation of capital in the world. This is because no matter where a company invests, it will pay the tax rate of its home country on the margin. If the company invests at home, it will pay the home tax rate on its investment income. On the other hand, if it invests abroad, then it will first pay foreign taxes. However, if the foreign tax rate is less than the tax rate at home, then the company will owe the difference to the home country; if the foreign tax rate is higher than the home tax rate, then the company will get a refund of the difference (assuming an unlimited tax credit). Hence, no matter where the company invests, it will pay exactly the same tax on the income generated and will have no tax-­induced incentive to invest in one country over another. This is called capital-­export neutrality and results in an efficient allocation of capital across countries. A second view is that the territorial system leads to an efficient allocation of savings across different countries. A territorial tax system exempts foreign-­ source income from tax, and consequently all investors in a country face the same tax rate on the margin. For this reason, it is said to comply with capital import neutrality. From an economic perspective, it is viewed as promoting savings efficiency, because any international saving (interest from foreign bonds for instance) is not influenced by the tax system. Both of these views have some validity. Worldwide residence taxation promotes investment efficiency but distorts saving decisions. A territorial system promotes the efficient allocation of savings but distorts investment decisions. A third view, developed primarily in the literature on subnational taxation, is the benefit view. Under this view, taxes are payments for the provision of public goods and services. People and companies willingly pay taxes in exchange for goods and services provided by the government. As people and businesses tend to consume public services where they are located, the tax jurisdiction should also be defined with respect to where they are located. A territorial system approximates this situation, and thus, under the benefit view, the territorial system may lead to efficient location decisions. Tax competition promotes efficiency in this view, because it tends to force governments to provide goods and services such that the marginal benefit of the public good is just equal to the marginal cost of the tax payment. A fourth view that has recently been raised is termed capital-­ownership neutrality (Desai and Hines 2003; Devereux 1990). According to proponents of this view, ownership of assets is important, because owners who are better able to run certain types of companies can generate a higher return. Hence, it is important to distribute owners of capital to the type of capital in which they have a comparative advantage. This view sees efficiency as requiring the tax system to

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

16   P. Pistone and T. J. Goodspeed be neutral with respect to the ownership of capital. Capital-­ownership neutrality can be achieved if all countries use the same tax system, whether it is a worldwide system or a territorial system. To summarize, economic views of tax jurisdiction have concentrated on various types of worldwide efficiency criteria. These include capital-­export neutrality, capital-­import neutrality, capital-­ownership neutrality and the benefit view of taxation. Capital-­export neutrality is associated with a worldwide tax system and implies efficiency in the allocation of capital across countries. Capital-­import neutrality is associated with a territorial system and implies efficiency in international savings decisions. Capital-­ownership neutrality can be achieved under either a worldwide or a territorial system, and implies efficiency in the matching of assets and owners. The benefit view requires those who benefit from public services to pay the associated costs, which are closely approximated by a territorial division of tax jurisdictions. 2.2.2  Tax jurisdiction and ability-­to-pay principles One reason that countries may prefer to use a worldwide system and extend tax jurisdiction abroad is to preserve domestic progressivity of the tax system and the ability-­to-pay principles. Highly taxed factors that are not compensated with correspondingly good public services will have an incentive to evade the high taxes by positioning themselves in low-­tax localities. However, the degree to which this is possible probably differs between persons and companies. Many countries that have territorial taxation with respect to corporate income maintain worldwide taxation with respect to personal income. In addition, even in countries with corporate territorial systems, corporate passive income is usually subject to special rules that effectively tax such income at the home rate. Hence, territorial tax systems tend to have many exceptions that attempt to preserve domestic ability-­to-pay concerns. 2.2.3  Tax jurisdiction and foreign investment in developing countries 2.2.3.1  The importance of all public policies The empirical literature on taxes and FDI has in general found significant tax effects, though the estimated elasticity varied significantly between them, depending on the data set used and whether the study was cross-­sectional or panel (Hines 1999). A set of papers from the 1980s used a time series of aggregate Bureau of Economic Analysis (BEA) data and found significant effects of taxation on FDI with an elasticity of about −0.6 (Hartman 1984; Boskin and Gale 1987; Young 1988). Others found significant effects when using the cross-­sectional depth of the BEA data to examine FDI across different countries for a given year (Hines and Rice 1994; Grubert and Mutti 1991). These studies also found significant effects, though with more variation in the point estimate of the effect of taxes. Other studies used firm-­level data, usually in a panel, and again observed significant tax effects.3

Thus, public finance literature, in general, found significant effects of taxes on FDI. Furthermore, studies also found that the impact was greater on developed countries than on developing countries, and the important factors for developing countries may include both the provision of adequate infrastructure and lower corruption.4 Figure 2.1 is illustrative and shows that the average FDI stocks are higher when taxes are lower in both developing and developed countries, though the impact is somewhat greater on developed countries. This suggests that the benefit view of tax effects may be particularly important in the case of developing countries: developing countries need to pay close attention to providing the right kinds of public inputs and business environment (such as low corruption) that will foster foreign investment, and not be concerned solely with low taxes. 2.2.3.2  The developing country problem in designing tax incentives Developing countries can have difficulties designing tax policies to attract investment when major foreign investors use the worldwide system of taxation. Suppose, for instance, that a developing country attempts to use a tax holiday to attract investment. A tax holiday is a period of time (usually several years) during which a company enjoys a reduced (usually zero) rate of tax. In the simplest case, a foreign investor from a country that practices worldwide taxation will not benefit from the tax holiday. This is because any profits generated from the investment that are repatriated would be subject to the home country tax. 200,000 180,000 160,000 140,000 $100 million

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Rethinking tax jurisdictions   17

120,000 100,000 80,000 60,000 40,000 20,000 0

Average FDI stocks for low-tax-rate developed countries

Average FDI stocks for high-tax-rate developed countries

Average FDI stocks for low-tax-rate developing countries

Average FDI stocks for high-tax-rate developing countries

Figure 2.1 Tax rates and FDI stocks, 2002 (source: Updated chart based on Goodspeed et al. (2009)).

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

18   P. Pistone and T. J. Goodspeed However, the same may not be true for foreign investment from a country that uses a territorial tax system. As the territorial country exempts foreign-­ source income, any tax holiday or other tax incentive would not be offset by home-­country taxes. Indeed, tax holidays and tax exemption are common tax incentives used in developing countries. However, these policies are ineffective when investment comes from countries with worldwide tax systems. Thus, the overall impact of a developing country’s tax incentives on inbound investment will depend on the extent of foreign investment from territorial or tax-­sparing countries (versus worldwide countries) as well as the general impact of taxes on foreign investment. Developed countries with worldwide tax systems have varying policies with respect to tax holidays of developing countries. Some countries, such as Japan, preserve the incentive created by the tax holiday by allowing a credit for taxes that would have been paid in the absence of the tax holiday (a policy called tax sparing); however, other countries, most notably the United States, do not. Thus, foreign investment from these latter countries cannot take advantage of any tax incentive offered by the developing country. 2.2.4  Tax jurisdiction as an optional tool to aid developing countries Developed countries have explicit policies with the goal of increasing investment and incomes in developing countries. Direct aid is an often-­used tool, and indeed the United Nations has established a goal for overseas aid to developed countries at 0.7 percent of GDP. By some accounts, direct aid has not been particularly successful in increasing the incomes of developing countries (Easterly 2001, 2003). In part, this may be because important components of market systems are absent or flawed in many developing countries. For instance, expropriation of assets and high corruption levels are not uncommon. The same might be said for taxation in developing countries. As was argued earlier, a good business environment (low corruption) and use of public funds for public inputs desired by foreign investors are probably more important than tax jurisdiction per se. Nevertheless, the tax system may be one underutilized tool, especially for investment coming from certain developed countries such as the United States that have a worldwide tax system but do not allow tax-­sparing provisions in their tax treaties. Ironically, US domestic social policy has increasingly moved from direct programs to aid through the tax system. This is exemplified by the earned income tax credit, a refundable tax credit that has become the largest welfare program in the United States. Of course, it is more complicated to use the tax system to try to encourage capital investment, and particularly to try to direct more investment towards low-­income countries. Nevertheless, Table 2.1 shows that developing countries tend to use tax holidays and lower tax rates to try to attract investment. It may be possible to lead developed countries toward more efficient and effective means of attracting foreign investment through more coordinated policies between developed and developing countries.

Rethinking tax jurisdictions   19 Table 2.1  Comparison of FDI incentives in OECD and developing countries FDI incentive

% OECD countries

% Developing countries

Tax exemption/holiday Lower tax rate

20  5

55 45

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Source: Goodspeed (2006).

Although the United States has historically rejected tax sparing, it is reconsidering its international tax policies (as noted below), and this may be a time for it to rethink tax sparing as well. Tax sparing can be criticized on a number of fronts. These include the fact that it encourages investment only for profitable companies, and hence does not provide much incentive for new start-­up companies, which typically incur initial losses, possible transfer-­pricing problems, and the possibility that tax sparing can change repatriation patterns, encouraging a more rapid repatriation of profits. However, probably the most serious set of issues is the more general problem associated with tax havens, which stems from the implied low tax rates of tax-­ sparing agreements. A multinational can abuse a worldwide tax system (or indeed a dividend exemption system) when one country has a zero or very low tax rate on capital, while others have significantly higher rates. Among the games that can be played are loans from a low-­tax country that can be deducted in the high-­tax country and be counted as income in the low-­tax country; hybrid transactions, particularly when debt in one country is viewed as equity in another; and pure abuse of transfer pricing rules. This is a serious and difficult problem and would probably be the single most important aspect in any tax-­ sparing proposal. This is underlined by data in Table 2.2 which show that roughly 84 percent of US FDI stock in developing countries is in Latin Amer­ ican and Caribbean countries, of which 40 percent is located in the British Virgin Islands. Recently, there have been a flurry of proposals to reform the US international tax system, and any discussion should first underline the possible ways in which US international taxation may change. First, President Bush appointed a bipartisan Tax Reform Panel that made recommendations on reforming the US international taxation in 2005. Second, President Obama named a bipartisan Tax Reform Commission that is due to release recommendations in December of 2009. Third, the US Treasury and President Obama issued a press release on 4 May 2009 that began to detail certain aspects of the President’s plan to reform US international tax laws and enforcement. Perhaps surprisingly, the 2005 recommendations of the Tax Reform Panel and the proposals of 4 May 2009 from the Treasury would move the US tax system in exactly opposite directions. In 2005, the President’s Tax Reform Panel recommended changing the US tax system to a territorial system. President Obama has suggested reforming the US tax system by limiting deferral, and hence moving the United States more toward a strict worldwide tax system. The differences are striking. By and large, the

20   P. Pistone and T. J. Goodspeed Table 2.2  US FDI stock abroad by country, 2003 ($ billions)

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Total developed Western Europe Other

1,272 978 294

Total developing Latin America and Caribbean   South America    Venezuela   Other    British Virgin Islands    Netherlands Antilles    Panama    Mexico    Bermuda    Bahamas Asia Africa

83 70 6 5 64 28 4 8 7 6 1 12 2

Other

23

Total

1,378

Source: UNCTAD, www.unctad.org (accessed March 2009).

2005 proposal would exempt foreign-­source income from US tax, while the 2009 Treasury proposal would apply US tax more heavily to foreign-­source income. How would all this affect developing countries? At first glance, one would think that a move toward a territorial system by the United States would make investment in low-­tax developing countries more attractive for US multinationals. A territorial system would exempt dividend repatriations, and thus would seem to encourage investment in these countries. Surprisingly, recent studies suggest that a move to a territorial system would not greatly impact US investment in low-­tax countries (Altshuler and Grubert 2001; Grubert 2006).5 In part, this is because multinationals are fairly adept at repatriating income without triggering additional US tax. However, this conclusion also depends on what developing countries and multinationals do in reaction to such a change. If multinationals restructure agreements to lower royalty payments and increase dividend payments, or if developing countries change their tax rates (or bases), it could make investment in low-­income countries more attractive. In contrast to the 2005 report by the President’s Tax Reform Panel, President Obama and the US Treasury appear to be favoring a stricter version of worldwide taxation, arguing that such a change would increase investment in the United States. The Treasury proposal suggests: (i) reforming deferral rules to disallow deductions for expenses (except for research and experimentation) on foreign-­source income until taxes are paid on the profits; (ii) closing certain loopholes associated with the foreign tax credit; and (iii) eliminating “check-­thebox” rules that enable multinational firms to choose their organizational form by checking a box, which can facilitate the use of hybrid entities to avoid US tax.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Rethinking tax jurisdictions   21 The details of these reform proposals are important (but not yet available). However, it is safe to say that they would constitute a stricter version of worldwide taxation. This would probably increase the difficulties of developing countries in attracting FDI. If the United States moves toward a stricter version of worldwide taxation, which appears possible, one option would be “partial” tax-­sparing agreements with developing countries. “Partial” tax sparing not only encourages investment in developing countries but also minimizes the problems that can arise with tax-­ sparing agreements. The idea would be to discourage tax holidays per se, but still try to allow a developing country’s tax system to encourage FDI, by allowing credit of a certain percentage of US-­owed tax. For instance, the United States might allow tax sparing of, say, one-­third of the US marginal tax rate. Hence, if the US tax rate was 33 percent, the repatriated profits that had been earned under a zero-­tax holiday would be subject to a 21 percent tax when repatriated to the United States. This could be made available to developing countries that negotiate a tax treaty with the United States, along with an exchange of information agreement to limit abuse in tax-­haven countries. Such credits could be kept in a different basket or carried out on a per-­country basis to avoid using those credits to shield income coming from high-­tax countries. Furthermore, developing countries would have an incentive to implement, say, a 21 percent tax, rather than a complete tax holiday, because otherwise the United States would receive the revenues. This might lead to higher revenues in the developing country, which could be used for infrastructure investment needs and could serve the dual purpose of encouraging investment in developing countries and limiting transfer pricing or other problems with tax havens (because of both the limit to the lower tax rate and the exchange of information agreement). One could also attach conditions related to good governance – limitations on corruption and expropriation of assets, for instance – although good measures for such conditions might be difficult to obtain.

2.3  The legal rationale of international tax jurisdiction in the light of European law 2.3.1  The exercise of national taxing powers within the European legal system The impact of European law on direct taxation only began to be analyzed in the early 1990s – that is, after the European Court of Justice started to set a growing number of limits to the exercise of taxing powers retained by the EU Member States. However, the attention paid to such issues has generally been focused on their dimension within the internal market, thus almost showing that European law is a matter for Europeans and within Europe. In contrast, for a long time an aura of uncertainty has surrounded various scenarios involving third countries, almost ignoring the fact that the evolution of European law, in particular since the

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

22   P. Pistone and T. J. Goodspeed liberalization of capital movements, has given it an external dimension. Its repercussions in the field of direct taxation, mainly due to how the European Court of Justice interpreted fundamental freedoms in its case law, represent an important element for the purposes of our studies as well as to achieve a critical revision of the interjurisdictional allocation of taxing powers with regard to developing countries. Preliminarily, for the purposes of our study, one should consider that (at least from a tax and legal perspective) third countries are not a single homogeneous block, but rather represent a complex category made up of countries that share the element of not being Member States of the European Union, despite being, in fact, very different with regard to their respective features and needs, including the degree to which they share the goals of the internal market. This characterization in function of European law is required, because from the perspective of EU Member States such countries are obliged to exercise their tax sovereignty in such a way as to ensure the supremacy of European law, and from the perspective of non-­EU Member States they may have to agree, on the basis of an agreement under public international law, to limit their sovereignty in line with the goals of the internal market not only as regards the EU Member States, but also as regards the European Union itself. Accordingly, previous research has grouped third countries into several categories, ranging from countries that are members of the European Economic Agreement and share the goals of the internal market almost entirely, to countries that do so in a more limited way.6 For the purposes of our study, it seems essential, though, to go beyond the traditional label of “developing country” and search for elements that define the complex features of such category in a more precise way. On the other hand, such elements should reflect the growing difference in economic growth and international tax policy needs arising within this category, as well as comply with the legal requirements set by European law for the exercise of national tax sovereignty by its Member States. The first specification plays a particularly important role for the purpose of justifying whether, and to what extent, developed countries should pursue goals of fostering economic development without paying specific attention to a reciprocal tax treatment, whereas the second is required to allow for a more favorable treatment toward nationals than that which would apply to EU nationals within the internal market. 2.3.2  Developing countries: a complex category Several states traditionally considered to fall within the category of developing countries have recently been included in economic steering groups, such as G20, showing that their role in the globalized economy is no longer marginal, but rather essential for achieving effective international coordination. Some of them still import capital from developed countries, while being net capital exporters to less developed countries (such as Brazil, Russia, India, and China, which make up the so-­called BRIC group, as well as several other coun-

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Rethinking tax jurisdictions   23 tries, like South Africa and Chile, which have been given full OECD membership after adapting their international tax policy to the new role). This shows that the traditional general differentiation between capital-­importing and capital-­ exporting countries would be inappropriate for our purposes. Hence, some developed countries have terminated previously concluded treaties, including elements (such as notional credit) that were aimed at supporting economic development.7 Furthermore, from the perspective of European law, some of them enjoy preferential treatment on the basis of European international agreements, through which Europe attempts to increase their economic development on a non-­reciprocal basis. All these arguments suggest that for the purposes of this study, instead of considering such countries as a single category it would be more appropriate to split them into two subcategories: those that could be defined as newly industrialized countries (NICs), comprising countries with a so-­called economy in transition (such as the former Member States of the Soviet Union), and the remainder, which are truly capital-­importing countries (CICs). Thus, each of the two categories may require a different analysis according to whether or not they have concluded European international agreements. NICs are extremely active in broadening their tax treaty network to further increase their competitiveness on the worldwide scene, which is already enhanced by their low labor costs and moderate budgetary needs, thus making it possible for them to afford more moderate corporate tax rates than those applied by highly industrialized countries (HICs). The old treaties of NICs with HICs currently boost their economic development when including mechanisms like notional credits, giving a tax incentive for capital investment. Such conditions generally do not raise significant problems of compatibility from the perspective of European tax law, though problems may arise in some specific situations. The most relevant example can be seen by looking at the combined action of tax treaties with European international agreements, including a most-­favored-nation clause, which has so far received little attention from legal and tax scholars. We believe that relations with third countries should be protected from compensatory measures, extending the same line of arguments as those applied by the European Court of Justice to the controlled foreign companies (CFC) legislation within the internal market, with regard to the Cadbury Schweppes decision. On the other hand, CICs mainly seek to attract international capital to their territory to have sufficient resources available to foster their economic growth, thus having a very limited negotiating power with respect to multinational enterprises (MNEs). CICs include several territorial jurisdictions,8 have ultra-­low labor costs and often possess insufficient financial resources to cover their budgetary needs. Fostering the economic development of CICs should still represent a priority for HICs, making it extremely important for such countries to master their own tax decisions, including those concerning the grant of tax incentives, without compensatory countermeasures by HICs. The same conclusion could be reached with respect to the policy that NICs should follow in their relations with CICs.

24   P. Pistone and T. J. Goodspeed

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

No problem of compatibility with European law should arise even when EU Member States provide relief for taxes not effectively paid in the CICs, such as in the case of notional credits, provided that such measures are applied in a non-­ discriminatory way. The main reason for such a conclusion is that European law acknowledges the importance of promoting economic development. Furthermore, it can be argued that the conclusions reached by the European Court of Justice in the Cadbury Schweppes case can be applied to any tax measure by the EU Member States compensating lower taxation applicable on the basis of measures pertinent in CICs. 2.3.3  The exercise of taxing powers by EU Member States in compliance with European law within the internal market and in relations with developing countries The exercise of taxing powers can be considered with respect to three specific issues, namely (i) the scope of personal taxation, (ii) the exercise of taxing sovereignty on non-­residents, and (iii) the relief for foreign taxes. The goal of such analysis is to describe how taxing powers should be exercised in such situation from a legal perspective, according to the common understanding in international taxation, as well as to evaluate the impact of European law on such rules, including some consideration on the relations with third countries and, in particular, with developing countries. 2.3.3.1  The scope of personal taxation Insofar as taxes are levied on a personal basis, a tax system will exercise its jurisdiction on any personal economic event related to the taxpayer. Accordingly, on the one hand the taxpayer’s global income and capital will be linked to the taxing jurisdiction of his or her country of residence (or, in some cases, nationality), whereas on the other hand, that country will be obliged to allow for deductions that are related to the taxpayer’s personal situation. European law has pushed this obligation even further, disconnecting it from the revenue taxed by the country of residence and rejecting the idea that the loss of tax revenue could constitute a valid justification for violating EU fundamental freedoms. Currently, no elements exist to exclude that the same conclusions may not be reached with respect to the relations with third countries, including NICs and CICs. Similarly, considering that losses will immediately affect the overall situation of the taxpayer, they should be taken into account by any system that exercises tax sovereignty on a personal basis. However, many systems currently take into account foreign-­source losses only to the extent that the tax jurisdiction is correspondingly exercised on foreign profits. This is based on a symmetry theory that tax treaties support, and yet prevent, the taking of losses into account whenever international double taxation is not relieved by the state of residence through the credit method, but rather by means of exemption. It turns the worldwide system in the latter situation into some kind of proxy for a territorial system, despite the

Rethinking tax jurisdictions   25

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

structural differences that should exist on the basis of their different rationale. In the Marks & Spencer case, European law has supported the obligation for the state of residence to import final losses even when incurred by a foreign subsidiary in the presence of a group taxation regime. Although this case was decided on the basis of the right of establishment, similar conclusions could be reached in relations with third countries (including developing countries) on the basis of the free movement of capital. However, the current evolution of European law seems to show that the potential for application of the Marks & Spencer decision is very limited. 2.3.3.2  Objective taxation and comparison with residents of source state The legal basis of territorial taxation is objective, or in rem, thus linking income (or capital) as such to the taxing jurisdiction, because of its place of origin or source, and regardless of any further circumstance that may be related to the person who produces (or owns) it. This legal difference has often been invoked to reject any possible comparison between taxation of residents and taxation of non-­residents under international tax law, with the former being liable to tax on a personal basis and the latter on an objective basis. This view is also reflected in Article 24 of double tax treaties based on the Model Tax Convention on Income and Capital drafted by the Organisation for Economic Co-­operation and Development (hereafter, OECD MTC), affecting the very foundation of the non-­discrimination principle contained therein. Thus, such treaties admit a comparison between residents and non-­residents only to a very minor extent. In contrast, a much broader comparison is possible in the light of European tax law, with respect to the Schumacker (ECJ, decision of 14 February 1995, case C-­279/93, Schumacker) and Saint-­Gobain (ECJ, decision of 21 September 1999, case C-­307/97, Saint-­Gobain) decisions. Insofar as the situation of a non-­ resident EU national in an EU Member State other than that of his/her/its residence is equivalent to that of the resident taxpayers of that Member State (a matter which is usually determined by means of their relation to what part of their income or capital is subject to the taxing sovereignty of such country), then he/she/it shall also be entitled to national treatment in such country, regardless of whether he/she/it is entitled to the benefits of a double tax treaty under international tax law. This conclusion is necessitated by the primacy of European law over national law (including double tax treaties). In contrast, whenever no such similarity exists, the taxpayer neither enjoys the right to national treatment within the internal market, nor – under the current stage of evolution of European law – could he/she/it claim equal treatment among non-­residents under a most-­favored-nation (MFN) approach (ECJ, decision of 5 July 2005, case C-­376/03, D., paras 61–62). However, the right to MFN exists with respect to some third countries, namely whenever European international partnership agreements, especially with developing countries, explicitly acknowledge it.9 As

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

26   P. Pistone and T. J. Goodspeed such treaties do not explicitly exclude taxation, and no good arguments have so far been put forward for their doing so implicitly at the level of interpretation, third-­country nationals may in fact enjoy a better situation than that of any EU national whenever none of such persons are entitled to invoke national treatment.10 Although one may question whether granting such better tax treatment to non-­EU nationals is in fact consistent with the rules and goals of the internal market, a valid reason for doing so with respect to developing countries may be represented by the intention to foster their economic growth. However, different reasons should be sought with respect to other countries, such as strengthening economic links or opening up their markets to the Europeans. However, a more circumscribed analysis of such goals is beyond the scope of our chapter and should also be sought on the basis of factual elements existing at the time of the concluding of such agreements.11 As the current development of European law has caused it to have an immediate impact on cross-­border direct taxation, including in relations with third countries, attention should be paid in the future when including MFN clauses within European international agreements, at least insofar as such agreements do not specifically exclude taxation from the scope of clauses of this kind. Various reasons support this conclusion. First, MFN clauses often have undesirable tax implications as regards consistency with the international tax policy of a country, and in particular with its treaties, which are often negotiated as package deals. Second, the implications of case law of the European Court of Justice in the absence of a coordinated European tax policy can be very far-­reaching: removing obstacles to cross-­border international situations may not necessarily comply with the need to consistently support a sound and well-­structured practice of fostering economic development. Third, the contracting states to such European international agreements must comply with the rules contained in those agreements even when exercising their jurisdictions on the basis of the applicable tax treaties with EU Member States. Finally, a fourth argument arises from a completely different perspective, insofar as one considers that the very existence of such European international agreements may represent, in terms of European law, the legal basis for indicating that (at least a mixed) competence has been shifted to the Community level. Additional issues arise with respect to non-­EU Member States that have not concluded the European Partnership Agreements (EPAs) with the European Union. Under such circumstances, the impact of European law is mainly circumscribed to the scope of free movement of capital,12 whose worldwide scope puts an obligation on the EU Member State that is not connected to a condition of reciprocity. The decision of the European Court of Justice seems to support the view that the external dimension of such freedom expands whenever the right of establishment is not applicable (ECJ, decision of 18 December 2007, case C-­101/05, Skatteverket v. A.). Furthermore, the need to ensure an objective entitlement to such freedom should represent an important element to ensure its consistent application, thus also opening it up to non-­EU nationals who own such capital or derive income from it. Accordingly, one could conclude that the free

Rethinking tax jurisdictions   27 movement of capital may be invoked with respect to the measures put forward by EU Member States to dissuade investment in developing countries. However, the European Court of Justice has not yet reached these conclusions, possibly because it has not yet been asked to reach a judgment on a case raising a similar issue. It has instead rejected the external scope of free movement of capital in cases that were also relevant for the free movement of services, considering that the external dimension of the latter freedom lacks a direct effect.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

2.3.3.3  Relief for double taxation When tax treaties do not allocate exclusive taxing powers to one contracting state, further problems may arise from the perspective of European law as to how relief for international double taxation is given. Relief methods can be applied on a bilateral (i.e. through the provisions contained in the tax treaty itself ) or unilateral basis (i.e. on the basis of domestic tax law), usually in the form of exemption (sometimes with progression), or credit for foreign taxes. The two methods have different impacts on the conditions for tax neutrality with respect to foreign-­sourced income, as will be more clearly analyzed from an economic perspective. However, also from a legal perspective a number of relevant issues arise that require a proper evaluation of the impact of European law on the relief of double taxation. The use of either method by the state of residence for the giving of relief on taxes levied in the source state is mainly an issue that each state decides in conformity with its own tax policy. However, from the perspective of European law the impact of such relief methods on cross-­border situations within the internal market could create some compensatory effects of lower taxes that would not arise with respect to fully domestic situations – that is, cases that fall under the tax sovereignty of a single Member State. On the basis of the arguments employed by the European Court of Justice with regard to the CFC legislation as a restrictive measure for its compensatory effects on lower taxes levied by the foreign tax jurisdiction,13 one may therefore support the view that giving relief through credit would affect locational neutrality insofar as the residence state offsets the more advantageous tax treatment that could be available in the source country, thus making it impossible for the investor to compete on an equal footing in the latter EU Member State. Arguments to support this view could be drawn by looking at how the UK High Court of Justice (UK High Court, decision of 11 November 2008, case [2008] EWHC 2893 [CH]) implemented the FII decision of the European Court of Justice (ECJ, decision of 12 December 2006, case C-­446/04, FII) on the use of different methods for relieving economic double taxation in domestic and cross-­ border situations. Insofar as the exemption and credit always have a different impact on income received by the shareholder, Member States may be free to choose either of them (and this reading is also confirmed by the Parent–Subsidiary Directive), but are obliged to apply the same method to both domestic and cross-­ border situations. One may therefore wonder whether national tax sovereignty

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

28   P. Pistone and T. J. Goodspeed could create such difference in treatment by using a relief method, such as credit that, when applied with respect to juridical double taxation, structurally has the effect of compensating foreign taxes levied on the same income.14 Within the internal market, the issue is not just whether a taxpayer having cross-­border income can bear the same tax burden as applies to any other taxpayer in the state of residence deriving the same amount of income from the sources located within such country. It is likewise important to allow any taxpayer to compete on an equal footing with the residents of other EU Member States without there being tax measures in their state of residence that might dissuade him/her from investing his/her capital in countries other than that of his/her residence. The restrictive implications of foreign tax credit are particularly evident in the presence of per-­ country limitations and tax-­rate differentials, which prevent an effective consideration of the taxpayer’s overall ability to pay at the level of the internal market. Despite acknowledging that in some cases disparities among the tax systems can play a significant role, the need to remove tax obstacles within the internal market may lead the European Court of Justice to gradually remove the dissuasive effect created by the foreign tax credit. However, such issue is not an exclusive matter for the internal market, but also affects third countries in particular, owing to the broader geographical scope of the free movement of capital in the EC Treaty. When considered in relation to developing countries, such compensatory effect may turn a preferential tax treatment given by the source state – for the purpose of fostering investment on its territory – into additional revenue for the state of residence of the taxpayer. Thus, beyond any issue concerning whether such tax incentives should be given by the state of source, from a legal perspective one can conclude that the use of credit to relieve juridical double taxation is as such structurally inadequate to pursue a tax policy that supports economic development and does not interfere with the source state’s own decisions. Such issues do not arise (or do so only to a lesser extent) whenever a notional credit is given by either tax sparing at nominal rates of the source state regardless of the tax actually borne by the taxpayer, or matching credit at fixed rates. Accordingly, from the perspective of European law, no credits of such type would have compensatory features and raise a problem of compatibility with the fundamental freedoms with respect to third countries. Exemption would also make the tax policy decisions of the developing country become the final ones, as it does not imply an actual exercise of the taxing jurisdiction over economic events taking place outside the territory of the state of residence. However, for this very reason, exemption does not allow for the deduction of foreign losses, which can instead be creditable against the domestic profits of the state of residence under tax sparing or matching credit. Possibly, this reason would make notional credit more desirable than exemption with a view to supporting economic growth. However, the issue here is not regarding whether economic growth should be enhanced by giving tax incentives or subsidies, but rather whether tax treaties should still include clauses that make developed countries deprive developing countries of their own tax policy decisions or, in contrast, let the latter countries have their fair share of revenue from income sourced within their territory.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Rethinking tax jurisdictions   29 Until now, the rationale of worldwide taxation with relief by the credit method has been to support capital export neutrality, as opposed to capital import neutrality being pursued through the exemption method. However, the issue is one regarding whether developed countries may consider giving up their worldwide taxation in their tax treaties with developing countries to foster economic growth. This matter is particularly sensitive in Latin American countries that support territorial-­based taxation (but is also perceived in India, especially with respect to taxation of royalties and the justification of taxing powers based on the definition of their source), which often presume that they are being expropriated of their own wealth by the tax systems of developed countries that in fact associate with their taxing jurisdiction economic events that take place outside their territory. Shifting back to a purely territorial taxation could be a fairly complex issue to accept and implement for developing countries, at least in the present scenario of international taxation. Changing the existing tax rules with regard to one or more developing countries, perhaps on the basis of international treaties, could be a more realistic alternative. Exemption is possibly the less sophisticated way to implement such innovation without implying a major revolution.15 Many feared that such a method might automatically open up the way to low or no taxation on cross-­border business, especially considering the strong negotiating power of some MNEs with governments of some developing countries wishing to attract international capital to their territory. However, the significant reduction of corporation tax rates over the past decade, in Europe in particular, may lead to reconsideration of the impact of the issue on cross-­border business. Two further arguments can be mentioned to support a shift toward exemption. First, insofar as the developing country reduces the level of its corporation taxes for all taxpayers (thus, without giving rise to ring-­fenced regimes for international capital), it would be perfectly in line with European law if tax treaties did not include measures to compensate for such lower levels of taxation.16 Second, from a legal point of view a revised strategy to support economic growth could simply sever the legal connection between the taxing jurisdiction of the developed country and income that is sourced in the developing country and, through the applicable tax treaty, is exclusively subject to the latter’s jurisdiction. In such a situation, it should not be a major problem for the developed country to obtain all the information that is required for enforcing its own jurisdiction effectively. This conclusion seems to be supported by the dramatic change in the international tax cooperation that is currently opening up an unprecedented worldwide expansion of mutual assistance in gathering information and collecting taxes. Various alternative paths for exemption have been suggested in the tax literature, though two are particularly interesting for our purposes, and hence will be specifically addressed subsequently. According to one proposal (Barker 2007: 347–377), instead of sharing taxing powers on income from capital that MNEs export into developing countries,

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

30   P. Pistone and T. J. Goodspeed such countries could slice the tax base of such income, possibly in agreement with HICs, to enjoy exclusive taxing powers on its parts. In particular, such a proposal suggests that mobile rents – that is, those derived from mobile factors of production that can be exploited in many different locations because the market for consumption is worldwide – should be taxed only in the country of residence of the MNE, whereas locational rents – that is, net income arising from the factors of production belonging to a host country and consisting in the remuneration of capital exceeding the normal rate of return – should instead remain within the exclusive jurisdiction of the developing country, considering the important role that that country plays in the production of such income. The author also suggests the latter rent as the object of tax-­sparing provision in HICs. Attributing an exclusive jurisdiction to tax on cross-­border income represents, in principle, a positive development in taxation of cross-­border income, thus removing some of the problems related to the existing relief methods and achieving the important result of preventing double taxation, rather than completely or partially removing its effects. Another proposal (Avi-­Yonah 1996) also supports the need to eliminate forms of shared taxation on cross-­border income, by assigning exclusive taxing rights to source countries on cross-­border active business income and to residence countries of MNEs on portfolio income. Both theories undoubtedly play a significant role in the research activity related to our study, in particular with respect to the output that is being drafted for the application with regard to the Latin American countries. Preventing double taxation represents an especially important element to simplify the rules of taxation on cross-­border income and remove the potential exposure to (also temporarily) unrelieved double taxation, while keeping a reasonable degree of fairness that can be positively (and with limited costs) handled by unsophisticated tax authorities. This reduces the possible exposure to (also temporarily) unrelieved double taxation, which represents a potential obstacle in the light of European law with respect to capital movements also in relation to third countries. All such reasons lead to the question of whether the former theory is in fact a good alternative to the latter one, especially if considering that (especially when applied unilaterally by the developing country, but, in fact, also in the presence of agreements with HICs) significant difficulties may arise in slicing the tax base in a way that exactly prevents double taxation. Furthermore, both theories perhaps underestimate the reluctance of several developing countries (in particular, CICs, but also the BRIC countries such as India, in the case of royalties) to give up their sovereignty on what they consider as income totally produced on their territory. Hence, some form of compromise between simplicity and fairness might be considered to achieve a solution that is also reasonably feasible to implement. Formulae could in principle be a good solution (Avi-­ Yonah 2007), though their determination by common agreement could give rise to the same problems that the separation of locational from mobile rents is likely to cause. Furthermore, their compatibility with European law might sometimes be difficult to defend on a legal basis, taking into account that the European

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Rethinking tax jurisdictions   31 Court of Justice has so far shown a preference for ascertaining the existence of obstacles on the basis of effective situations. A constructive solution could be provided by making use of some form of revenue sharing which allows for the payment of a fee for the information gathered by the state of source and makes it possible for the latter country to maintain some revenue from cross-­border income yielded by the capital imported into its territory. Although economists (e.g. Keen and Ligthart 2006b: 88) have questioned its efficiency, from a legal perspective its application in the European common system for taxation of cross-­border savings payments to EU beneficial owners made by paying agents located in Austria, Belgium and Luxembourg, as well as in Switzerland and a few other countries, seems fairly successful. Another possibility is the use of notional tax credits or “tax sparing”,17 which maintains the levels of taxation of developed countries without depriving the developing country of the positive effects of its own tax policy decisions. However, the main obstacle with respect to shifting toward such a method in developing countries lies in the traditional opposition developed in the 1960s by the United States and its scholars (Kuhn 1963: 262), which considers that relieving taxes regardless of their actual payment to the revenue of the developing country can have several undesirable effects. The European scenario, however, provides for a significant number of treaties, including forms of notional relief for foreign taxes, though some countries have recently departed from such a pattern. Two examples provide evidence of this trend. First, Sweden terminated its treaty with Peru, considering that notional tax credit was leading to some undesired forms of tax arbitrage and abuse. Second, Germany terminated its treaty with Brazil, because it no longer regarded Brazil as a developing country and thus concluded that the rationale for such favorable treatment did not exist. Little attention has been paid to the possible implications of European law, in particular with regard to Brazil, whose treaties with several other European countries include forms of relief by notional tax credit similar to the one contained in the treaty terminated by Germany. In such a scenario, the right for a German company to enjoy national treatment in any of such European countries (e.g. when having a permanent establishment (PE) on their territory18) would automatically entitle it to national treatment, and thus to a relief by means of notional tax credit equivalent to that which would apply under the treaty of such a country with Brazil.19 In other words, the implications of European tax law for the international tax policy of the EU Member States show the need for better international tax coordination to break the wall of silence that has so far surrounded this domain. However, one may argue that the issue is not so much tax sparing as such, but how such a policy is actually drafted, structured and implemented.

2.4  Rethinking tax jurisdictions: a possible new model This final section contains the combined output of research aimed at developing a new approach to the problems of international taxation in relation to developing

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

32   P. Pistone and T. J. Goodspeed countries. Such an approach is drafted in such a way as to combine legal standards currently set by European tax law (in terms of an effective removal of obstacles on international movements of capital) with a sound international business tax policy from an economic perspective, and make it possible to foster economic development through taxation. Far from supporting the use of tax incentives for such purpose with new arguments, the goal of our work is to stimulate debate in tax-­related legal and economic literature on a possible reconsideration of the boundaries of tax jurisdictions and methods for relieving double taxation in relation to developing countries. More specifically, this section will focus its attention on the particular situation of Latin America, considering that most Latin American countries share the same legal and tax tradition20 despite the fact that economic development (of NICs in particular) has led some of them to abandon the traditional territorial boundaries of their national jurisdiction. The application to Latin America considers the relations among NICs and CICs of that region, as well as the relations of all such countries with HICs, with special emphasis on EU Member States bound by European law to ensure free movement of capital with respect to third countries on a unilateral basis.21 As a preliminary, we should recall the importance of European law in transforming the international tax scenario from a mere matter of allocation of taxing powers among the states – without any international rights for taxpayers – to an area in which the exercise of national sovereignty in the field of direct taxation is subject to conditions which seek to avoid tax treatments that actually discriminate against cross-­border situations or otherwise dissuade taxpayers from investing their capital abroad. The broad scope of prohibition of discrimination and restriction in the direct tax case law of the European Court of Justice, and the immediate right of EU nationals to activate legal remedies against violations by the Member States, represent a Copernican revolution in international taxation, which is now gradually expanding to non-­EU Member States also. From such a perspective, upgrading the legal standards of international taxation in the external relations of the EU and its Member States will soon become a necessity from the legal perspective. Applying this pattern to non-­EU Member States as well would represent a significant step for international taxation from a legal perspective, because it would bring about an effective dimension of rights in cross-­ border situations; in addition, also from an economic point of view, it would be significant because it would foster a spontaneous trend of economic integration that corresponds to the needs of the globalized economy. Such a scenario should be ideal for making taxation play a new role in fostering economic development: instead of giving financial subsidies or tax incentives, developed countries could reconsider allocation rules with respect to developing countries. In particular, a general shift toward territorial or exclusive allocation of taxing powers in cross-­border situations could be pursued with respect to developing countries. Exposure to international double taxation represents a considerable burden for MNEs, often leading them to set up complex schemes to operate in specific countries. Removing the risk of unre-

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Rethinking tax jurisdictions   33 lieved double taxation for international business, or the possibility that HICs might compensate lower taxation in the state of source, could perhaps stimulate the investment of international capital in developing countries and significantly contribute to their economic growth. An additional issue could arise from the perspective of preventing abusive practices in cases of shared allocation of taxing powers with a view to minimizing the tax burden in the state of source. Such practices, better known as treaty-­shopping techniques, would simply disappear as a consequence of the exclusive allocation of taxing power to one single state. This certainly would not remove the need to include measures to counter other types of abusive practices, though clear signs now indicate that this may be carried out by means of exchange of information22 without giving rise to disproportionate reactions by the tax systems; all such clauses (anti-­ abuse as well as broad clauses on exchange of information and assistance in collection of taxes) represent an essential tool for modern tax treaties, including those with developing countries. This proposal could alter tax-­competition incentives, in particular as far as it is concerned with corporation tax rates. However, to some extent this phenomenon has already occurred following the US 1986 tax reform, reforms in other European countries and the widening of the European Union to 27 states, as a consequence of the more moderate budgetary needs of the new Member States, their intention to attract foreign capital, and the obligation for all Member States to apply non-­discriminatory tax treatment in corporation–shareholder relations.23 This gave rise to a generalized reduction in corporation tax rates throughout the European Union, bringing down the average rates to well below 30 percent.24 One way to implement a move toward territorial taxation with respect to developing countries is presented in the current studies on a Draft Model Latin American Tax Convention on Income (the Draft LATC) (Quiñones 2010). An approach to balance out a surrender of taxing powers by HICs would be to give them exclusive taxing powers on passive income (dividend, interest, royalties and capital gains). However, a debate carried out on the framework of the preparation of the Draft LATC has shown that this objective is simply unrealistic because of the strong resistance, in particular within Latin American countries, to the idea of not taxing the income produced on their territory (Tenore 2010). A possible solution to the problem would be to admit a system of revenue sharing with respect to such type of income, while developing countries tax such income at source and then pass on the revenue to HICs after retaining a portion. The mechanism is similar to the one contained in the transitional regime set by Art. 11 of the EU Savings Directive and in the EU international tax agreements with Switzerland, Andorra, Liechtenstein, Monaco and San Marino, but the function is partly different. Instead of being a mere instrument to keep track of cross-­border flows of passive income, this mechanism is a way to replace shared taxing powers with sharing the tax revenue from such income (ibid.). For other categories of income, whenever an exclusive allocation of taxing powers is not possible, relief from international taxation could be given by the

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

34   P. Pistone and T. J. Goodspeed state of residence in a way that does not produce compensatory effects – hence by either exemption or tax sparing. Partial tax sparing, as proposed earlier in the chapter, could be a convenient way to avoid some of the drafting problems that tax sparing has produced in some of the existing double tax treaties. The adoption of the proposed rules on exclusive allocation of taxing powers and of methods for relieving double taxation other than foreign tax credit could indeed complicate the achievement of domestic ability-­to-pay principles. However, countries that use territorial taxation also tend to maintain their own ability-­to-pay criteria with respect to personal taxation and passive income. In other words, many countries that have opted for territorial taxation maintain their own home tax rate for income perceived to be particularly susceptible to tax avoidance (such as passive income) as well as for income deemed to be particularly important for domestic ability-­to-pay concerns (such as personal taxation). Achieving satisfactory arrangements for all countries is difficult and complex, especially with regard to developing countries, and taxation is no exception.

Notes   1 The European Association of Tax Law Professors analyzed the European dimension of the ability-­to-pay principle in its 2008 Congress, held in Cambridge. More details on this can be found at www.eatlp.org.   2 See, for instance, Goodspeed et al. (2009). Several studies suggest that taxation is not an important instrument with which to promote economic growth. See, for instance, McKenzie et al. (1997) and Easson (1999). Measuring the effectiveness of tax incentives has always been problematic, though perhaps their main weakness lies in the fact that they increase the complexity of a tax system. Accordingly, the IMF and the World Bank have generally opposed their introduction in developing countries, which should keep the tax system simple enough to be administered by unsophisticated tax authorities. See on this, among others, Gupta and Tareq (2008) and Bergsman (1999).   3 These studies used a variety of data sources. For instance, Compustat data were used by Auerbach and Hassett (1993) and by Cummins and Hubbard (1995). Commerce data were used by Ondrich and Wasylenko (1993). Altshuler et al. (2001) used US Treasury data. Hines (1996) also used a panel, but exploited state-­level tax differences using BEA data and found significant tax effects for foreign investment in US states.   4 For instance, Wheeler and Mody (1992) found a significant effect for agglomeration (as measured by infrastructure quality) and an insignificant effect for taxation. This study used pooled data, however, which could bias the results if there are significant country-­fixed effects. In another preliminary piece of work, Goodspeed et al. (2009) found a negative impact of taxes, a positive impact of infrastructure, and a negative impact of corruption on inbound FDI. These effects differ between developing and developed economies, however, with the tax effects being significantly larger for developed countries and insignificant for developing countries.   5 Altshuler and Grubert (2001) and Grubert (2006) indicate that it depends on how the territorial system is implemented. However, assuming some allocation of expenses and current taxation of royalties and interest, both studies suggest that not much would change.   6 A categorization of seven groups of countries has been put forward by Pistone (2006): (i) European Economic Area (EEA) countries, (ii) EPA countries, (iii) Switzerland, (iv) developing countries, (v) the United States, (vi) tax havens, and (vii) all other countries.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Rethinking tax jurisdictions   35   7 As in the case of the German double tax convention with Brazil.   8 However, not all territorial jurisdictions are CICs. See, for instance, the case of Singapore, an international financial center with good economic development, whose territorial tax system has a different justification.   9 This seems fairly clear from the wording of the agreements with several African, Caribbean and Pacific (ACP) countries, economies in transition (notably the former Member States of the Soviet Union), and countries of the Euro-­Mediterranean agreements, which include a significant number of directly applicable most-­favored-nation clauses, especially on aspects involving the secondary right of establishment, but sometimes also involving the supply of services. See further on this in the appendices to Bezborodov (2007: 708–710). Surprisingly, to date, no case law exists on the matter, which has also been unduly neglected by legal scholars. 10 Such a situation would, in our view, affect the entitlement to lower withholding taxes and give rise to a number of inconsistencies. See further on this in Pistone (2002: 130–131) and, for an example of the possible inconsistencies arising in tax treaties, Vogel (1995: 264). 11 Good examples of this kind are the agreements with Russia and Croatia. A paradox arises with respect to Croatia, where an exception is made for the free movement of services (currently not included in the European international agreement), and accession to the European Union would worsen its situation, downgrading it from MFN to national treatment for the right of establishment. 12 Nevertheless, relevant issues for the right of establishment could also arise in triangular cases involving non-­EU Member States. 13 ECJ, decision of 12 September 2006, case C-­196/04, Cadbury Schweppes, paras 45–46. Such a decision was applied to the right of establishment for the same reason that the European Court of Justice had previously applied in tax and non-­tax cases, for example with respect to the free movement of services in the decision of 26 October 1999, case C-­297/97, Eurowings, paras 44–46. 14 Surprisingly, the European Court of Justice took a different view in ECJ, decision of 6 December 2007, case C-­298/05, Columbus Container Service, para. 40, with respect to the application of German unilateral switchover clauses in a case concerning a Belgian coordination center and its German partners. This decision creates tax fragmentation within the internal market, and was not motivated on the grounds of seeking to achieve a proportionate reaction to abusive practices. 15 The shift to exemption has been argued in tax literature by Brown (2002), though McDaniel (2003) provided arguments to support investment in developing countries through a revised version of the US foreign tax credit rules, which are regarded (see McDaniel 2007) as better on the grounds of simplicity, efficiency and equity. 16 Although European law does not interfere in principle with the allocation of taxing powers, the outcome of such allocation should nevertheless be compatible with the primacy of European law (see ECJ, Saint Gobain, paras 56–58), thus possibly excluding measures that compensate for lower taxes levied in another country (leaving aside the case of abusive practices, of course), which in our opinion should also include third countries. 17 Advocates for tax sparing include Hines (1998) and Brooks (n.d.). 18 A clear-­cut case of such implications can be found in the decision of 9 May 2007 by the Finnish Supreme Administrative Court (KHO:2007:30), which, on the basis of the Saint-­Gobain decision of the European Court of Justice, applied to cross-­border royalties earned in China by a Luxembourg company with a permanent establishment in Finland a treatment equivalent to that provided by the China–Finland double tax convention (DTC). 19 Such a result is nonetheless possible to the extent that Germany exempts the income of the foreign permanent establishment.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

36   P. Pistone and T. J. Goodspeed 20 From a legal perspective, their tax systems are reasonably homogeneous, owing to the structural influence exercised on their current wording by the Latin American Model Tax Code (Modelo de Código Tributario para América Latina) drafted by the experts of the Latin American Institute for Tax Law (Instituto Latinoamericano de Derecho Tributario, ILADT) in the late 1950s. 21 Our study backs up a project for the drafting of two model tax conventions for Latin America, which is being conducted by an international research group (made up by legal experts and headed by Jacques Malherbe, Pasquale Pistone and Heleno Taveira Tôrres) on behalf of the Latin American Institute for Tax Law (ILADT), whereas preparatory studies have been presented at a seminar held in Montevideo on 11–12 May 2009 (on which see further Mazz and Pistone 2010). In particular, such a project supports the conclusion of a multilateral tax convention among Latin American countries, as well as a model bilateral tax convention for assisting tax treaty negotiators of Latin America when concluding their treaties with HICs. 22 Over the past couple of years, the OECD has undertaken a strategy to put significant pressure on all countries to cooperate to conclude tax treaties including broad exchange of information clauses. Now the effects of this policy can also be seen among countries that previously preserved banking secrecy and were considered tax havens. 23 The latter phenomenon is relatively well known among legal experts as the outcome of legal interpretation by the European Court of Justice. See ECJ, decision of 6 June 2000, case C-­35/98, Verkooijen; decision of 15 July 2004, case C-­315/02, Lenz; and decision of 7 September 2004, case C-­319/02, Manninen. 24 In the case of some old EU Member States such as Germany, corporation tax rates have been lowered to as little as 15 percent. This indicates that the traditional big gap between EU HICs and developing countries is no longer what it used to be decades ago.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

3 How to combat tax evasion in tax havens? A legal and economic analysis of OECD and EU standards on exchange of information in tax matters with a special focus on capital income Dietmar Aigner and Michael Tumpel 3.1  Introduction The liberalization of international capital movements has made significant progress in recent decades inside and outside the European Union (see Council Directive 88/361/EEC, June 1988). Although it is widely accepted among economic scholars that free trade of goods and services and free movement of capital is in principle the superior trade policy and can foster economic development, a “borderless world” with free movement of capital investments facilitates tax evasion (Bacchetta 1992; Gläser and Halla 2006). The global economic crisis and recent tax evasion scandals have spurred calls for fairness and transparency in the tax system. Removing practices that facilitate tax evasion is part of a broader drive to clean up one of the more controversial sides of a globalized economy. Combating tax evasion, which mainly arises from income on cross-­border investments, has been an issue within the European Union since the 1960s. It has become increasingly important because so-­called tax havens are being intensively used for aggressive tax planning. They deprive governments of revenues needed for vital infrastructure and undermine the confidence that citizens have in the fairness of their tax laws. National governments need to take action to ensure that taxpayers do not abuse this borderless world to evade their tax liabilities (Gläser and Halla 2006). The harmful effects of tax havens result from their lack of transparency, reliance on excessively strict bank secrecy, and the opportunities that they provide taxpayers to avoid or evade tax in their home countries. The European Commission mentions that fraud accounts for approximately 2–2.5 percent of GDP in the European Union (European Commission IP09/201 2009). Therefore, the interests at stake in the case of fiscal fraud are extremely important. Although the challenge of combating offshore tax evasion has grown more complex and more serious, given the increased scope for illicit use of the international financial system in a globalized world, the functioning of fiscal

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

38   D. Aigner and M. Tumpel systems is still primarily the responsibility of the Member States. They are responsible for fiscal administration, tax auditing and the recovery of taxes. When countries take firm action to stop the above-­mentioned loss of revenue, they need to rely on international organizations such as the Organisation for Economic Co-­operation and Development (OECD) or the European Union to take action in a coordinated and integrated manner (see IP/09/201 2009). At the OECD level, the work on harmful tax practices has developed standards of transparency and exchange of information, which have been widely accepted as an international norm. The initiatives aim at eliminating harmful tax practices of preferential tax regimes by identifying “tax havens” and seeking their commitments to the principles of transparency and effective exchange of information. So far, all the countries identified as tax havens by the OECD have made formal commitments to implement these principles. The OECD principles are laid down in an OECD Model Agreement on Exchange of Information on Tax Matters and in Article 26 of the OECD Model Tax Convention on Income and Capital. At the EU level, the European Commission considers that tax evasion could better be controlled through joint action by the Commission and the Member States. As a part of the so-­called tax package, the European Union has enacted the Council Directive 2003/48/EC of 3 June 2003 on taxation of savings income in the form of interest payments (referred to hereafter as the “Savings Directive”). Furthermore, measures equivalent to or the same as those provided for by the Savings Directive have also come into effect in Andorra, Liechtenstein, Monaco, San Marino, Switzerland and certain relevant dependent or associated territories of the Member States.1 The ultimate aim of the Savings Directive is to enable savings income in the form of interest payments made in one Member State of the European Union to the so-­called beneficial owners, who are individuals residing in another EU Member State, to be made subject to effective taxation in accordance with the laws of the latter Member State. This is to be achieved through an automatic exchange of information on cross-­border interest payments to individuals. However, owing to structural differences that are mainly caused by national banking secrecy, Austria, Belgium and Luxembourg are permitted, for a transitional period, to levy a withholding tax on the savings income of residents of other Member States, instead of providing the other Member States with information. This will guarantee a minimum effectiveness in the field of savings taxation. Therefore, the main objective of the Savings Directive is to reduce tax evasion by EU residents who gain interest income from assets abroad and fail to report this income to their domestic tax authorities (Gläser and Halla 2006). In this chapter, we have evaluated the effectiveness of OECD and EU measures to combat tax evasion concerning capital income from a legal and an economic point of view. We have to conclude that there exist restricted scope and loopholes within the OECD and EU measures. Furthermore, as tax-­related information exchange is a strategic variable of the countries to alter the attractiveness of their financial markets, it is at least debatable whether the existence of loopholes is due to hidden non-­cooperative behavior by the Member States.

How to combat tax evasion in tax havens?   39

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

This chapter is organized as follows. After a review of the background and the general principles of the OECD and EU measures to combat tax evasion concerning capital income in sections 3.2 and 3.3, respectively, in section 3.4 the economics of tax information exchange and contrasts of the OECD measures and the Savings Tax Directive with the former are examined. In section 3.5, the effectiveness of the OECD and EU measures in reducing cross-­border tax evasion and tax avoidance is discussed. The final section concludes the chapter and discusses broader policy implications.

3.2  OECD standards on exchange of information 3.2.1  Model agreement on exchange of information in tax matters 3.2.1.1  Background Under common international law and the principle of sovereignty, states are allowed to tax income if there is a genuine link to their territory. Nearly all states make use of the resident taxpayer as the genuine link to their territory. According to the principle of residence, these states are allowed to tax the worldwide income of taxpayers who have their permanent residence within their territory. Even income derived from investments in a source state is subject to tax in the state of residence, if the beneficiary of this income is a resident of the latter state. Because the state of residence has no information on this income, the taxpayer is generally obliged to declare the worldwide income to the competent tax authority of the state of residence. Clearly, the principle of residence requires that tax authorities have comprehensive information about all the sources of income of their residents in foreign states. A problem that often arises is that taxpayers do not declare their income from foreign sources. As capital is a very flexible commodity and there is generally no (automatic) tax-­related information exchange outside the European Union between the state of source and the state of residence, investing in foreign countries appears to be a straightforward way to evade taxes (Gläser and Halla 2006). The OECD launched its work to address harmful tax competition in its 1998 report Harmful Tax Competition: An Emerging Global Issue. This report describes the lack of effective exchange of information as an important factor in determining harmful tax practices (Oberson 2003). The OECD advocates exchange of information between tax authorities on request in cases of specific tax inquiries to better equip tax authorities to tackle tax evasion, and released a model agreement on exchange of information in tax matters (hereafter “Model Agreement”) on 18 April 2002. The Model Agreement represents the standard for effective exchange of information for the OECD’s initiative on harmful tax competition. At the same time, the OECD published a list of seven “uncooperative jurisdictions”. However, 31 other jurisdictions decided to cooperate with the OECD and have bound themselves either directly or indirectly to the Model Agreement. The purpose of this Agreement is to promote international

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

40   D. Aigner and M. Tumpel cooperation in tax matters through exchange of information. It was developed by the OECD Global Forum Working Group on Effective Exchange of Information. The Working Group consisted of representatives from the OECD member countries as well as delegates from Aruba, Bermuda, Bahrain, the Cayman Islands, Cyprus, the Isle of Man, Malta, Mauritius, the Netherlands Antilles, the Seychelles and San Marino. The lack of effective exchange of information is one of the key criteria in determining harmful tax practices. The mandate of the Working Group was to develop a legal instrument that could be used to establish effective exchange of information. The Model Agreement is a non-­binding instrument that provides the basis for an integrated bundle of bilateral treaties. A party to the multilateral Agreement would only be bound by the Agreement vis-­ à-vis the specific parties with which it agrees to be bound. Thus, a party that seeks to be bound by the multilateral Agreement must specify, in its instrument of ratification, approval or acceptance of the party or parties to which it wishes to be bound. The Agreement then enters into force, and creates rights and obligations, though only between those parties that have mutually identified each other in their instruments of ratification, approval or acceptance that have been deposited with the depositary of the Agreement. The bilateral version is intended to serve as a model for bilateral exchange of information agreements. As such, modifications to the text may be agreed in bilateral agreements to implement the standard set in the model (Oberson 2003). The Agreement is intended to establish the standard of what constitutes the effective exchange of information for the purposes of the OECD’s initiative on harmful tax practices. However, the purpose of the Agreement is not to prescribe a specific format for how this standard should be achieved. Thus, the Agreement in either of its forms is only one of several ways in which the standard can be implemented. Other instruments, including double taxation agreements, may also be used, provided both parties agree, given that other instruments are usually wider in scope. A number of bilateral agreements have already been based on this Agreement. 3.2.1.2  Key regulations According to Article 1 of the Model Agreement on Exchange of Information in Tax Matters, competent authorities of the contracting parties shall provide assistance upon request through exchange of information that is foreseeably relevant to the administration and enforcement of the domestic laws of the contracting parties concerning taxes covered by the Model Agreement. The agreement is limited to information that is foreseeably relevant to the determination, assessment and collection of taxes covered, the recovery and enforcement of tax claims, or the investigation or prosecution of tax matters. Information is to be exchanged in accordance with the provisions of the Agreement and shall be treated confidentially. The standard of foreseeable relevance is intended to provide for exchange of information in tax matters to the widest possible extent, and at the same time to clarify that contracting parties are not at liberty to engage

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

How to combat tax evasion in tax havens?   41 in fishing expeditions or to request information that is unlikely to be relevant to the tax affairs of a given taxpayer. Parties that choose to enter into bilateral agreements based on the Model Agreement may agree to an alternative formulation of this standard, provided that such alternative formulation is consistent with the scope of the Model Agreement (Oberson 2003). The rights and safeguards secured to persons by the laws or administrative practice of the requested party remain applicable to the extent that they do not unduly prevent or delay the effective exchange of information. Such rights may include, depending on the circumstances, a right of notification, a right to challenge the exchange of information following notification, or rights to challenge information-­gathering measures taken by the requested state. Such procedural rights and safeguards also include any rights secured to persons that may flow from relevant international agreements on human rights, and the expression “unduly prevent or delay” indicates that such rights may take precedence over the Agreement. Therefore, Article 1 of the Model Agreement strikes a balance between the rights granted to persons in the requested state and the need for effective exchange of information. It ensures that the rights and safeguards are not overridden simply because they could, under certain circumstances, operate to prevent or delay effective exchange of information. However, Article 1 obliges the competent authority of the requested state to ensure that any such rights and safeguards are not applied in a manner that unduly prevents or delays effective exchange of information. For instance, a bona fide procedural safeguard in the requested state may delay a response to an information request. However, such a delay should not be considered as “unduly preventing or delaying” the effective exchange of information unless the delay is such that it calls into question the usefulness of the information exchange agreement for the applicant’s competent authority. Another example is with respect to the notification requirements. A requested party whose laws require prior notification is obliged to ensure that its notification requirements are not applied in a manner that, in the particular circumstances of the request, would frustrate the efforts of the party seeking the information. For instance, notification rules should permit exceptions from prior notification (e.g. in cases in which the information request is of a very urgent nature, or when the notification is likely to undermine the chance of success of the investigation conducted by the applicant party). To avoid future difficulties or misunderstandings in the implementation of an agreement, the contracting parties should consider discussing these issues in detail during negotiations and in the course of implementing the agreement, to ensure that information requested under the agreement can be obtained as expeditiously as possible while ensuring adequate protection of taxpayers’ rights. Agreements on Exchange of Information in Tax Matters in Article 3 cover, at a minimum, four categories of direct taxes (i.e. taxes on income or profits, taxes on capital, taxes on net wealth, and estate, inheritance or gift taxes), unless both parties agree to waive one or more of them. Article 5 of the Model Agreement on Exchange of Information in Tax Matters provides the general rule that the competent authority of the requested party

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

42   D. Aigner and M. Tumpel must provide information upon request for the purposes referred to in Article 1. The paragraph makes clear that the Agreement only covers the exchange of information upon request (that is, when the information requested relates to a particular examination, inquiry or investigation) and does not cover automatic or spontaneous exchange of information. However, contracting parties may wish to consider expanding their cooperation in matters of information exchange for tax purposes by covering automatic and spontaneous exchanges and simultaneous tax examinations. The reference in the first sentence to Article 1 of the Agreement confirms that information must be exchanged with respect to both civil and criminal tax issues. Information in connection with criminal tax issues must be exchanged irrespective of whether or not the conduct being investigated would also constitute a crime under the laws of the requested party. National tax authorities have to take action to obtain the information requested and cannot solely rely on the information possessed by its competent authority. Reference is made to information “in its possession” rather than that “available in the tax files”, because some contracting parties do not have tax files because they do not impose direct taxes. Upon receipt of an information request, the competent authority of the requested party must first review whether it has all the information necessary to respond to a request. If the information in its own possession proves inadequate, it must take “all relevant information gathering measures” to provide the applicant party with the information requested. The term “information gathering measures” is defined in Article 4, para. 1 of the Model Agreement on Exchange of Information in Tax Matters. An information-­gathering measure is “relevant” if it is capable of obtaining the information requested by the applicant party. The requested party determines the information-­gathering measures that are relevant in a particular case. Para. 2 further states that information must be exchanged irrespective of whether the requested party needs the information for its own tax purposes. This rule is needed because a tax interest requirement might overwhelm the effective exchange of information, for instance in cases where the requested party does not impose an income tax or the request relates to an entity not subject to taxation within the requested party. Article 5, para. 3 includes a provision intended to require the provision of information in a format specifically requested by a contracting party, to satisfy its evidentiary or other legal requirements to the extent allowable under the laws of the requested party. Such forms may include depositions of witnesses and authenticated copies of original records. Under Article 5, para. 3, the requested party may decline to provide the information in the specific form requested if such form is not allowable under its laws. A refusal to provide the information in the format requested does not affect the obligation to provide the information. By referring explicitly to persons who may enjoy certain privilege rights under domestic law, Article 5 makes clear that such rights cannot form the basis for declining a request unless otherwise provided in Article 7. For instance, the inclusion of a reference to bank information in Article 5, para. 4, subparagraph a) rules out any possibility that bank secrecy could be considered as a part of

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

How to combat tax evasion in tax havens?   43 public policy (ordre public). Similarly, Article 5, para. 4, subparagraph a), together with Article 7, para. 2, makes it clear that information that does not otherwise constitute a trade, business, industrial, commercial or professional secret or trade process does not become a secret simply because it is held by one of the persons mentioned. In accordance with the OECD report Improving Access to Bank Information for Tax Purposes (2000), access to information held by banks or other financial institutions may be, by direct means or indirectly, through a judicial or administrative process. As stated in the report, the procedure for indirect access should not be burdensome and time-­consuming, such that it acts as an impediment to access to bank information. Typically, requested bank information includes account, financial and transactional information as well as information on the identity or legal structure of account holders and parties to financial transactions. Furthermore, Article 5, para. 4 mentions information held by persons acting in an agency or fiduciary capacity, including nominees and trustees. A person is generally said to act in a “fiduciary capacity” when the business that he/she transacts, or the money or property that he/she handles, is not his/her own or for his/her own benefit, but for the benefit of another person related to him/her, implying and necessitating confidence and trust on one part and good faith on the other part. The term “agency” is very broad and includes all forms of corporate service providers (e.g. company formation agents, trust companies, registered agents, lawyers). The competent authorities of the contracting parties must have the authority to obtain and provide ownership information that the other contracting party may legitimately expect to receive in response to a request for ownership information so that it may apply its own tax laws, including its domestic definition of beneficial ownership. In connection with companies and partnerships, the legal and the beneficial owner of the shares or partnership assets will usually be the same person. However, in some cases the legal ownership position may be subject to a nominee or similar arrangement. In cases where the legal owner acts on behalf of another person as a nominee or under a similar arrangement, that person, rather than the legal owner, may be the beneficial owner. Thus, the starting point for the ownership analysis is legal ownership of shares or partnership interests, and all contracting parties must be able to obtain and provide information on legal ownership. Partnership interests include all forms of partnership interests: general, limited, capital or profits. However, in certain cases legal ownership may be no more than a starting point. For example, in any case where the legal owner acts on behalf of any other person as a nominee or under a similar arrangement, the contracting parties should have the authority to obtain and provide information about that other person, who may be the beneficial owner, in addition to the information about the legal owner. An example of a nominee is a nominee shareholding arrangement where the legal title-­holder, who also appears as the shareholder of record, acts as an agent for another person. Within the constraints of Article 2 of the Agreement, the requested party must have the authority to provide information about the persons in an ownership chain.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

44   D. Aigner and M. Tumpel Article 7 identifies the situations in which a requested party is not required to supply information in response to a request. If the conditions for any of the reasons for declining a request under Article 7 are met, then the requested party is given the discretion to refuse to provide the information; however, it should carefully weigh the interests of the applicant party with the pertinent reasons for declining the request. However, if the requested party does provide the information, then the person concerned cannot allege an infraction of the rules on secrecy. In the event that the requested party declines a request for information, it should inform the applicant party regarding the reasons for its decision at the earliest opportunity. Article 10 establishes the requirement for contracting parties to enact any legislation necessary to comply with the terms of the Agreement. The contracting parties are obliged to enact any necessary legislation with effect from the date specified in the Agreement, as well as, implicitly, to refrain from introducing any new legislation contrary to their obligations under this Agreement. 3.2.2  Article 26 of the OECD Model Tax Convention on Income and Capital 3.2.2.1  Background Before the introduction of tax information exchange agreements, international exchanges of information regarding direct tax issues had, since the 1990s, basically been carried out within the context of double taxation treaties. Article 26 of the OECD Model Tax Convention (OECD MTC) provides the most widely accepted legal basis for bilateral exchange of information for tax purposes. More than 3,500 bilateral tax treaties are based on the OECD MTC. The provision creates an obligation to exchange information that is foreseeably relevant to the correct application of a tax convention as well as for the purposes of the administration and enforcement of domestic tax laws of contracting states. Countries do not have the autonomy to engage in “fishing expeditions” or to request information that is unlikely to be relevant to the tax affairs of a given taxpayer. In formulating requests, the requesting state should demonstrate the foreseeable relevance of the requested information. In addition, it should also pursue all domestic means to access the requested information, except those that would give rise to disproportionate difficulties (Engelschalk 2008). Since the 2005 update of Article 26, it has been made clear that a state cannot refuse a request for information solely because it has no domestic tax interest in the information or solely because a bank or other financial institution holds the information. Bank secrecy is not incompatible with the requirements of Article 26, and virtually all countries have bank secrecy or confidentiality rules. Meeting the standard of Article 26 requires only limited exceptions to bank secrecy rules, and would not undermine the confidence of citizens in the protection of their privacy. Finally, where information is exchanged, it is subject to strict confidentiality rules. It is expressly stated in Article 26 that information communicated

How to combat tax evasion in tax havens?   45 shall be treated as a secret and that it can only be used for the purposes provided for in the convention (see Article 26, paras 4 and 5 OECD MTC).

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

3.2.2.2  Key regulation As mentioned earlier, the competent authorities of the contracting states shall exchange information that is foreseeably relevant for carrying out the provisions of this Convention, or to the administration or enforcement of the domestic laws concerning taxes of every kind and description imposed on behalf of the contracting states, or their political subdivisions or local authorities, insofar as the taxation there under is not contrary to the Convention. According to the Commentary on the OECD MTC, the standard of “foreseeable relevance” is intended to provide exchange of information on tax issues to the widest possible extent, and, at the same time, to clarify that contracting states are not at liberty to engage in “fishing expeditions” or to request information that is unlikely to be relevant to the tax affairs of a given taxpayer. Contracting states may agree to an alternative formulation of this standard that is consistent with the scope of Article 26 (e.g. by replacing “foreseeably relevant” with “necessary” or “relevant”). The scope of exchange of information covers all tax issues without prejudice to the general rules and legal provisions governing the rights of defendants and witnesses in judicial proceedings. Exchange of information on criminal tax issues can also be based on bilateral or multilateral treaties on mutual legal assistance (to an extent, they also apply to tax crimes). To retain the exchange of information within the framework of the Convention, a limitation is set so that the information is given only insofar as the taxation under the domestic taxation laws concerned is not contrary to the Convention (Commentary on Article 26, para. 1 of the OECD MTC). Article 26 of the OECD MTC embodies rules under which information may be exchanged to the widest possible extent, with a view to lay the proper basis for the implementation of the domestic tax laws of the contracting states, as well as for the application of specific provisions of the Convention. The text of Article 26 makes it clear that the exchange of information is not restricted by the scope of the OECD MTC as defined in Articles 1 and 2, and that the information may include particulars about non-­residents and may relate to the administration or enforcement of taxes not referred to in Article 2. Before 2000, Article 26 only authorized the exchange of information and the use of the information exchanged in relation to the taxes covered by the Convention under the general rules of Article 2. As drafted, the paragraph did not oblige the requested state to comply with a request for information concerning the imposition of a sales tax, because such a tax was not covered by the Convention. The paragraph was then amended so as to cover the exchange of information concerning any tax imposed on behalf of the contracting states, or their political subdivisions or local authorities, and to allow the use of the information exchanged for the purposes of the application of all such taxes. The rule laid down in Article 26, para. 1 of the OECD MTC allows information to be exchanged in three different ways:

46   D. Aigner and M. Tumpel •

•

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

•

on request, with a special case in mind, and understanding that the regular sources of information available under the internal taxation procedure should be relied upon in the first place before a request for information is made to the other state; automatically, for example when information about one or various categories of income having their source in one contracting state and received in the other contracting state is transmitted systematically to the other state; spontaneously, for example in the case of a state having acquired information that is considered to be of interest to the other state, as a result of certain investigations.

Any information received under Article 26, para. 1 by a contracting state is to be treated as a secret in the same manner as the information obtained under the domestic laws of that state, and is to be disclosed only to persons or authorities (including courts and administrative bodies) concerned with assessment or collection, enforcement or prosecution, determination of appeals in relation to the taxes referred to in Article 26, para. 1, or the oversight of all the above-­ mentioned roles. Such persons or authorities are to use the information only for such purposes. They may disclose the information in public court proceedings or judicial decisions. In no case are the provisions of Article 26, paras 1 and 2 to be construed so as to impose the obligation on a contracting state: • • •

to carry out administrative measures at variance with the laws and administrative practice of that or the other contracting state; to supply information that is not obtainable under the laws or in the normal course of the administration of that, or the other, contracting state; to supply information that would disclose any trade, business, industrial, commercial, or professional secret or trade process, or information disclosure that would be contrary to public policy (ordre public).

If the information is requested by a contracting state in accordance with Article 26, then the other contracting state is to use its information-­gathering measures to obtain the requested information, even though that other state may not need such information for its own tax purposes. The obligation contained in the preceding sentence is subject to the limitations of para. 3, but in no case are such limitations to be construed as permitting a contracting state to decline to supply information solely because it has no domestic interest in such information. This paragraph was added in 2005 to deal explicitly with the obligation to exchange information in situations where the requested information is not needed by the requested state for domestic tax purposes. Prior to the addition of Article 26, para. 4, this obligation was not expressly stated in the article but was clearly evidenced by the practices followed by the member countries. According to Article 26, para. 5, in no case are the provisions to be construed as permitting a contracting state to decline to supply information solely because

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

How to combat tax evasion in tax havens?   47 the information is held by a bank, other financial institution, nominee or a person acting in an agency or a fiduciary capacity, or because it relates to the ownership interests in a person. Article 26, para. 5, added in 2005, represents a change in the structure of the article, and should not be interpreted as suggesting that the previous version of the article did not authorize the exchange of such information. The vast majority of the OECD Member Countries already exchanged such information. Article 26, para. 5 does not preclude a contracting state from invoking Article 26, para. 3 to refuse to supply information held by a bank, financial institution, a person acting in an agency or fiduciary capacity, or information relating to ownership interests. However, such refusal must be based on reasons unrelated to the person’s status, such as a bank, financial institution, agent, fiduciary or nominee, or the fact that the information relates to the ownership interests. Austria, Switzerland, Luxembourg and Belgium have reserved the right not to include Article 26, para. 5 in their double tax conventions (DTCs). However, Austria is authorized to exchange information held by a bank or other financial institution where such information is requested within the framework of a criminal investigation carried out in the requesting state concerning the commitment of tax fraud. Switzerland has proposed to limit the scope of Article 26 to information necessary for carrying out the provisions of the OECD MTC. This reservation may not be applicable to cases involving acts of fraud subject to imprisonment according to the laws of both contracting states. Belgium reserves that in conventions in which Article 26, para. 5 is included, the exchange of information held by a bank or other financial institution is restricted to the exchange on request of information concerning both a specific taxpayer and a specific financial institution. The current edition of the OECD MTC, updated on 17 July 2008, indicates that Austria, Belgium, Luxembourg and Switzerland have entered reservations to Article 26. However, in March 2009 each of these countries notified the OECD that they were withdrawing those reservations (see the Commentary to Article 26, paras 4 and 5 of the OECD MTC).

3.3  EU standards on exchange of information on capital income 3.3.1  The savings tax directive 3.3.1.1  Background Within the European Union, awareness of the problem of tax evasion concerning capital income emerged in the 1960s (Debatin 1965; Fischer 1968; Groeben 1962; Grund 1965; van Themaat 1966). At that time, it was widely accepted that the best way to combat these opportunities for tax evasion would be the introduction of a tax-­related system of information exchange between the national tax authorities, and the abolition of existing withholding taxes. Other, different

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

48   D. Aigner and M. Tumpel systems, such as the then current withholding tax procedure, were considered as conceivable but imperfect (van Themaat 1966). By the implementation of Council Directive 88/361/EEC of 24 June 1988, the free movement of capital between Member States of the European Union and between Member States and third countries was further increased. Article 6, para. 5 of this directive authorized the Commission to submit proposals aimed at eliminating or reducing risks of distortion, tax evasion and tax avoidance linked to the diversity of the various national systems for the taxation of savings and for controlling the application of these systems. After several proposals, the Council agreed on Directive 2003/48/EC of 3 June 2003 on taxation of savings income in the form of interest payments (Gläser and Halla 2006). 3.3.1.2  Key regulations According to Article 1, the ultimate aim of the Savings Directive is to enable savings income in the form of interest payments made in one Member State to beneficial owners who are individual residents in another Member State for tax purposes, subject to effective taxation in accordance with the laws of the latter Member State. This aim will in principle be achieved by providing the beneficial owners’ state of residence with sufficient information regarding the interest payment by the state from which the interest was paid. With reference to banking secrecy, Austria, Belgium and Luxembourg are allowed to operate a withholding tax system instead of the information system during a transitional period when interest payments are made from paying agents within their territory to beneficial owners who are residents of other Member States. The withholding tax rate currently is 20 percent and will rise to 35 percent by July 2011 for the remainder of the transitional period. Three-­quarters of the withholding tax levied by Austria, Belgium and Luxembourg must be transferred to the states of residence of the beneficial owners. The remaining quarter may be kept by the withholding tax states. Nevertheless, the other Member States are obliged to provide Austria, Belgium and Luxembourg with information regarding interest payments made to their residents. The three withholding tax states are free to change to the information procedure at any time during the transitional period (Gläser and Halla 2006). According to Article 10 of the Savings Directive, the Council has to agree on the fulfillment of several conditions to oblige the three withholding tax states to change to the information procedure. While Belgium may switch to the information procedure before the end of the transitional period (Eynatten 2006; Springael 2004), Austria, on the other hand, may not (Kuttin 2003). To our knowledge, there is no evidence that Luxembourg is likely to switch before the end of the transitional period. Nevertheless, considering the conditions2 that have to be fulfilled, the end of the withholding tax procedure seems to be far in the future (Gläser and Halla 2006). Under the rules of the Savings Directive, the paying agents (any economic operator who pays interest or secures the payment of interest for the immediate

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

How to combat tax evasion in tax havens?   49 benefit of the beneficial owner) is responsible for the identification of the beneficial owner and his/her state of residence, calculation of the amount of interest paid and forwarding the information or deduction, and forwarding the withholding tax to the national tax authority. Beneficial owners may only be individuals who receive interest payments for their own benefit. Interest payments made for the benefit of companies or legal persons are not within the scope of the Savings Directive. The regulations of the Savings Directive are only applicable if the beneficial owner is a resident in a Member State other than where the paying agent is established. If the withholding tax needs to be deducted in Austria, Belgium or Luxembourg, then the beneficial owner has to be offered an opportunity to avoid this deduction. Therefore, the Savings Directive offers two procedures,3 of which the withholding tax countries are required to implement at least one (Gläser and Halla 2006). The deduction of the withholding tax under the regulations of the Savings Directive does not release the beneficial owner from his/her tax liability in the state of residence. Rather, it is the state of residence that is obliged to eliminate any double taxation that might result from the imposition of this withholding tax, along with any national income tax on the interest payment. The state of residence is obliged to grant a credit equal to the amount of the tax withheld in accordance with its domestic law. If the deducted withholding tax exceeds the amount of tax due in accordance with domestic law, then the state of residence is obliged to refund the excess amount of tax withheld to the beneficial owner (Article 14 of the Savings Directive). 3.3.1.3  Equivalent measures in third countries and dependent territories The application of the Savings Directive is conditional on the simultaneous application of treaties providing for equivalent measures in third countries such as Andorra, Liechtenstein, Monaco, San Marino and Switzerland, as well as the dependent and associated territories of the United Kingdom (Anguilla, the British Virgin Islands, the Cayman Islands, Guernsey, the Isle of Man, Jersey, Montserrat, the Turks and Caicos Islands) and the Netherlands (Aruba and the Netherlands Antilles). The British Virgin Islands, Guernsey, the Isle of Man, Jersey, the Netherlands Antilles and the Turks and Caicos Islands exercise the withholding tax procedure, while Anguilla, Aruba, the Cayman Islands and Montserrat provide the state of residence with information regarding the interest payment. The detailed provisions are equal to those of the Savings Directive. Remarkably, only Aruba, the British Virgin Islands, Guernsey, the Isle of Man, Montserrat and the Netherlands Antilles receive information or a part of the deducted withholding tax when interest is paid from the paying agents established in the Member States of the European Union to their residents, while Anguilla, the Cayman Islands, Jersey, the Turks and Caicos Islands and all the third countries get neither the information nor a part of the withholding tax (Gläser and Halla 2006).

50   D. Aigner and M. Tumpel

3.4  An economic perspective From an economic perspective, there are at least two aspects of the OECD and EU principles on exchange of information that are discussed in the economic literature. The first aspect is the analysis of administrative mechanism of the tax-­ related information exchange system, and the second aspect deals with individual tax evasion.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

3.4.1  Exchange of information in tax matters and tax competition From an economic point of view, it is interesting to analyze the conditions under which a country is willing voluntarily to exchange tax-­related information with other countries. Keeping in mind that economic theory basically assumes that all agents (including countries and their tax authorities) act in their own interest, it has to be considered that providing information to a foreign tax authority makes a country less attractive to foreign investors (Gläser and Halla 2006). Moreover, the administrative costs that arise during the gathering and transmitting of information are not compensated. From this point of view, a tax-­related exchange of information seems to be a serious drawback for participating countries. On the other hand, a country providing information to others could expect to receive tax-­related information in return to enforce its own taxes. Considering this trade-­off, it is a priori not clear whether a country should engage in a tax-­ related information exchange system (ibid.). Gläser and Halla assumed that if all countries were identical, signifying that investors would be indifferent regarding where to shift their assets, then a tax-­ related exchange of information would be the preferred strategy by all the countries. However, by introducing heterogeneity, the benefits from a tax-­related information exchange system may not accrue equally to all participating countries. This denotes that the net exporters of capital (net importers of information) might gain from information exchange, whereas the net importers of capital (net exporters of information) might not (Gläser and Halla 2006). Tanzi and Zee (2001) argued that a tax-­related information exchange system can never be self-­enforcing in a decentralized world with asymmetric countries. Bacchetta and Espinosa (1995, 2000) as well as Keen and Ligthart (2006a) highlighted the strategic elements of a tax-­related information exchange system, which might promote non-­cooperative solutions. They showed that governments may have a further incentive to transmit information through strategic motives if commitments on the degree of information exchange are made prior to the choice of the tax rate. If all countries were to agree on information exchange, then investors could not simply evade taxes by investing abroad. This would reduce the incentive for investors to invest abroad, and therefore soften the extent of tax competition. As a result, countries would be able to increase their own tax rate. According to Gläser and Halla (2006), these additional strategic considerations permit a system of tax-­related information exchange to be optimal in situations of non-­cooperation.

How to combat tax evasion in tax havens?   51

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

In summary, the existence of information exchange systems between asymmetric countries can be explained by certain strategic motives (Keen and Lig­ thart 2006b). However, there are at least two persistent facts underlying all these models that have to be considered. First, there are few fundamental incentives for governments to participate officially in a tax-­related information exchange system and to provide information to foreign tax authorities. Therefore, loopholes in the information exchange systems and uncooperative behavior by countries are basically severe problems. Furthermore, one has to consider that net importers of capital (typically, small countries) can never gain from information sharing as much as can net exporters of capital (Gläser and Halla 2006). 3.4.2  Individual tax evasion In an economic context, the term “tax evasion” includes illegal activities to reduce personal tax liability (e.g. the under-­reporting of income) and tax avoidance, which is consistent with the existing law (e.g. taking advantage of tax loopholes) (Slemrod and Yitzhaki 2002; Cross and Shaw 1981). In the following section, we will focus on tax evasion, but many of the considerations do equivalently apply to tax avoidance.4 Since the first formal economic analysis of tax evasion (Allingham and Sandmo 1972), an overwhelming number of studies have been carried out (e.g. Cowell 1990; Andreoni et al. 1998; Slemrod and Yitzhaki 2002). In this study, we first present the fundamental findings of the economic literature, especially the traditional determinants of tax evasion and tax morale, and then evaluate whether an exchange of information in tax matters based on OECD or EU principles is capable of reducing cross-­border tax evasion (Gläser and Halla 2006). 3.4.2.1  The economics of crime and the traditional model of tax evasion The economics-­of-crime approach is based on the assumption that criminals are rational people (Becker 1968). Therefore, whether or not a person will commit a crime is analyzed in a similar way to any other economic decision. In making an economic decision, economic agents are confronted with different combinations of costs and benefits. A decisive element of this approach is that all criminal behavior is characterized by uncertainty. Economists traditionally examined individual behavior under uncertainty within the so-­called expected utility framework. Under the assumption of the expected utility framework, an economic agent always chooses the alternative with the maximal expected payoff.5 Therefore, to increase the deterrence of criminal activity a state can either increase its expenditures on police and courts to catch a larger proportion of offenders or can impose more severe punishments. Of course, in a wider approach toward individual criminal behavior, the amount of crime is determined not only by the legal framework but also by the economic and social environment, such as opportunities for legal employment, schooling, and living

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

52   D. Aigner and M. Tumpel conditions in general (Becker 1997). The optimal policy for combating illegal activity is the problem of finding an optimal allocation of resources, and is therefore at the heart of economics. Tax evasion is a particular form of crime that can be analyzed within such a framework (Gläser and Halla 2006). Allingham and Sandmo (1972) and Yitzhaki (1974) carried out the first formal economic analysis of tax evasion (see also Srinivasan 1973). These models differ only in their assumptions on the penalty structure of tax evasion. Allingham and Sandmo presumed that the penalty depends on the income understatement, while Yitzhaki presumed that the tax understatement is decisive (Slemrod and Yitzhaki 2002). To sum up, the Allingham–Sandmo–Yitzhaki (referred to as “ASY” hereafter) model of tax evasion predicts that tax evasion decreases with an increase in the probability of detection and with an increase in the penalty rate; and if preferences exhibit decreasing absolute risk aversion, then tax evasion also decreases with an increase in the tax rate and a falling income (Gläser and Halla 2006). Evidently, this simple, static model is in some respects a sharp simplification of the real world and the taxpayer’s decision problem. First, it is uncertain whether the taxpayer really knows the probability of detection. In reality, the tax authority might not announce the number of audits that are to take place in the upcoming year. Second, the taxpayer might not know the penalty rate. Many institutional settings to some extent leave it to the discretion of a distinct authority to set the actual penalty rate. Third, it is not realistic that the probability of an audit is constant for all taxpayers. Usually, a taxpayer has to pay taxes in consecutive years. This time dimension will probably influence the taxpayer’s behavior as well as the state’s enforcement policy. Prudent tax authorities might be more likely to screen a former tax evader than a citizen who has been honest for many years. An evader who has been detected might anticipate this strategy. Finally, it is not clear whether an audit can always help the state to find out the exact amount of the taxpayers’ actual income, and this model does not allow for intrinsically motivated taxpayers (Gläser and Halla 2006). The basic ASY framework has been generalized and extended in many directions (Andreoni et al. 1998); however, in particular, two results of the ASY model have been strongly criticized. First is the predicted negative influence of the tax rate on the income evaded. Probably not many economists would suggest that a positive relationship between the tax rate and tax evasion is much more in accordance with intuition and common sense. In fact, most of the empirical and experimental analysis conducted reports that tax evasion rises with an increase in the marginal tax rate. To solve this problem, a more recent study attempted to analyze tax evasion within the so-­called non-­expected utility frameworks (Starmer 2000). The expected utility framework has been the working horse for economists to study decisions made under uncertainty for decades. However, there exists accumulated evidence that people do not behave in accordance with it. Dhami and al-­Nowaihi (2005) employed Kahneman and Tversky’s cumulative prospect theory (Tversky and Kahneman 1992), which is probably the most important non-­expected utility framework for modeling an individual’s decision

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

How to combat tax evasion in tax havens?   53 regarding tax evasion. Their model predicts that an increase in the tax rate will lead to an increase in evaded income. This result is contrary to the prediction made by the ASY model under the expected utility assumption, but in accordance with intuition and common sense (Gläser and Halla 2006). The second heavily criticized result of the ASY model concerns its quantitative prediction of tax evasion. To square the predicted degree of tax evasion of the ASY model under the expected utility with empirical and experimental evidence, taxpayers would need to be risk-­averse to an absurd degree. The realistic audit probabilities may be very low, ranging from 0.01 to 0.03, and the actually observed penalty rates may be between 1.5 and 2. Given these parameter values, one has to assume a coefficient of relative risk aversion of about 70 to obtain realistic estimates of evaded taxes predicted by the ASY model. However, realistic magnitudes of relative risk aversion lie between 1 and 2 (Alm et al. 1992). In general, it seems that incorporating non-­economic motivation, such as psychological and sociological variables – especially this intrinsic motivation to pay taxes – is a decisive point in solving this quantitative puzzle. The evidence from experimental work showing that this intrinsic motivation to pay taxes (called tax morale) affects compliance behavior is overwhelming (Torgler 2002). For instance, Baldry (1986) reported that some individuals do not act as gamblers and never evade taxes, even though doing so would be rational for them. Frey (1992) used the term “ipsative possibility set” to capture the idea that individuals may have a personal set of potential actions that does not coincide with the objective set (Gläser and Halla 2006). 3.4.2.2  Policy implications to combat tax evasion What can policymakers do to combat tax evasion on the basis of the results from economic research on tax evasion? First, if one takes the ASY model seriously, the determinants of tax evasion are found to be the probability of detection, the penalty rate, the tax rate and the individual income. If they refrain from using individual incomes and the tax rate as potential policy instruments to reduce tax evasion, policymakers can only directly use the probability of detection and the penalty rate to affect tax evasion. In addition, they can try to reduce tax evasion via an increase in tax morale. This leaves us with two questions: (i) What combination of audit probability and penalty rate should be chosen? (ii) How can politicians encourage a high level of average tax morale in the population? To answer the first question, one has to recognize that the two direct policy instruments differ in respect of the cost of making changes. Although it is in principle virtually costless to increase the penalty rate, an increase in the probability of detection involves a considerable cost. Increasing the probability of detection is equal to auditing a larger fraction of the population. Therefore, the state faces a trade-­off between higher expenditures on audits and lower tax revenues because of tax evasion. On the basis of this reasoning, the optimal policy would be not only to audit only a very small fraction of the population, but also to impose very high fines – or, as Kolm (1973) strikingly put it, “hang tax

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

54   D. Aigner and M. Tumpel evaders with probability zero”. However, in reality there are legal and social constraints on the upper limit of the penalty. If one takes an upper limit for the penalty into account, it turns out that the optimal policy will imply at least some tax evasion (ibid.). The only possible attempt to answer the second question is to take a look at the empirical literature on tax morale. By roughly considering only the factors that the policymakers may influence, these results tell us that when more citizens trust the state, they become more satisfied with the work of their government and their tax morale becomes higher. However, policymakers may not gain much from this, as they should try to do a good job anyway. Moreover, the question of the direction of causality clearly is: do “good” people have “good” governments, or do “good” governments have “good” people? There is probably no simple way, if any, to identify the true direction of causality. However, in any case it seems natural that any policy intervention that attempts to increase awareness (in a “benevolent” way) of the problem of tax evasion should at least not increase tax evasion via a deterioration in tax morale (Gläser and Halla 2006). In the next section, we will evaluate the OECD and EU standards on exchange of information in tax matters on the basis of the results from economic research on individual tax evasion.

3.5  An economic evaluation 3.5.1  OECD principles on exchange of information in tax matters 3.5.1.1  Aim of the OECD principles The OECD Model Agreement on Exchange of Information in Tax Matters and Article 26 of the OECD MTC basically provide for an exchange of information between the competent authorities of the contracting parties upon a request that is foreseeably relevant to the administration and enforcement of the domestic laws of the contracting parties concerning taxes covered by the Model Agreement. In both cases, the agreement is limited to the exchange of information that is foreseeably relevant to the determination, assessment and collection of such taxes, the recovery and enforcement of tax claims, or the investigation or prosecution of tax matters. However, the OECD principles on exchange of information in tax matters only provide for an exchange of information upon request; they do not implement automatic exchange of information. The standard of foreseeable relevance is intended to provide for the exchange of information in tax matters to the widest possible extent, and at the same time to clarify that contracting parties are not at liberty to engage in fishing expeditions or to request information that is unlikely to be relevant to the tax affairs of a given taxpayer. In this respect, one has to notice that the state of residence needs to have some kind of more or less detailed information about resident taxpayers who may evade taxes by, for

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

How to combat tax evasion in tax havens?   55 example, making capital investments in foreign capital markets. As fishing expeditions are not intended either by the regulations of the OECD Agreements on Exchange of information in tax matters or Article 26 of the OECD MTC, national tax authorities need to have personalized information about their resident taxpayers to fulfill the requirements for an exchange of information upon request that meets the standard of foreseeable relevance. Therefore, the OECD rules on exchange of information do not necessarily lead to a substantial improvement in national tax administrations. Resident taxpayers can easily evade taxes on capital income as long as the state of residence does not have any idea of foreign capital investments. Successful avoidance of the information transfer due to a complete lack of information in the state of residence simply enables taxpayers to conceal interest income from their state of residence, as if the OECD rules had never been introduced. However, in that case the introduction of the OECD rules on exchange of information in tax matters does not cause any increase in the probability of detection. In the following section, we will point out situations where an exchange of information according to the OECD standards on exchange of information can be revised. 3.5.1.2  Refusal of exchange of information upon request Article 7 of the Model Agreement contains the customary reasons for declining a request, which can also be found in other instruments for exchange of information. However, Article 7 also contains several important innovations and more precise formulations. As mentioned earlier, the Model Agreement does not allow a request to be denied solely because the requested party does not need the information for its own administrative practice. If the conditions for any of the reasons for declining a request under Article 7 are met, then the requested party is given discretion to refuse to provide the information; however, it should carefully weigh the interests of the applicant party with the pertinent reasons for declining the request. If the requested party does provide the information, then the person concerned cannot allege an infraction of the rules on secrecy. In the event that the requested party declines a request for information, it must inform the applicant party of the reasons for its decision at the earliest opportunity (Barnard 2003). The first sentence of Article 7, para. 1 makes it clear that a requested party is not required to obtain and provide information that the applicant would not be able to obtain under similar circumstances under its own laws for purposes of administration or enforcement of its own tax laws. This rule is intended to prevent the applicant party from circumventing its domestic law limitations by requesting information from the other contracting party, thus making use of greater powers than it possesses under its own laws. For instance, most countries, under their domestic laws, recognize that information cannot be obtained from a person to the extent that such person can claim the privilege against self-­ incrimination. Therefore, a requested party may decline a request if the applicant party is precluded by its own self-­incrimination rules from obtaining the

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

56   D. Aigner and M. Tumpel information under similar circumstances. Article 7, para. 2 states that a contracting party is not obliged to provide information that would disclose any trade, business, industrial, commercial or professional secret or trade process. Most information requests will not raise issues of trade, business or other secrets. For instance, information requested in connection with a person engaged only in passive investment activities is unlikely to contain any trade, business, industrial, commercial or professional secret, because such a person is not likely to be conducting any trade, business, industrial, commercial or professional activity. Financial information, including books and records, does not generally constitute a trade, business or other secret. However, in certain limited cases the disclosure of financial information might reveal a trade, business or other secret. For instance, a requested party may decline a request for information on certain purchase records, where the disclosure of such information would reveal the proprietary formula of a product. The Agreement overrides any domestic laws or practices that may treat information as a trade, business, industrial, commercial or professional secret or trade process merely because it is held by a person identified in Article 5, para. 4, or merely because it is ownership information. Thus, in connection with information held by banks, financial institutions, etc., the Agreement overrides domestic laws or practices that treat the information as a trade or other secret when it is in the hands of such a person, but does not afford protection when it is in the hands of another person, for instance the taxpayer under investigation. In connection with the ownership information, the Agreement makes it clear that information requests cannot be declined merely because domestic laws or practices may treat such ownership information as a trade or other secret. A contracting party may decline a request if the information requested is protected by attorney–client privilege. However, when the equivalent privilege under the domestic law of the requested party is narrower than the definition contained in para. 3 (for example, the law of the requested party does not recognize a privilege in tax matters, or it does not recognize a privilege in criminal tax matters), a requested party may not decline a request unless it can base its refusal to provide the information on Article 7, para. 1. The attorney–client privilege attaches to any information that constitutes 1 2 3 4

“confidential communication” between “a client and an attorney, solicitor, or other admitted legal representative”, if such communication “is produced for the purposes of seeking or providing legal advice” or is “produced for the purposes of use in existing or contemplated legal proceedings”.

Article 7, para. 5 clarifies that an information request must not be refused on the basis that the tax claim related to it is disputed. Another innovation is that Article 7, para. 3 states that information which would breach attorney–client privilege need not be provided. This establishes a

How to combat tax evasion in tax havens?   57 very important principle and ensures that there is no loophole enabling a country to use creative regulations and/or interpretations to circumvent its duty to provide information (Barnard 2003).

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

3.5.1.3  Limits to exchange of information according to Article 26 of the OECD MTC Under Article 26 of the OECD MTC, the competent authorities of contracting states may exchange the information necessary for the correct application of the treaty or the domestic laws of contracting states.6 This system, however, has its limits. First, it includes only countries with a treaty network and does not work in relation to countries, notably tax havens, that have not concluded any tax treaties, in particular because they do not levy any significant taxes. For these countries, the issue of international double taxation does not arise. Second, even in a treaty context the requested contracting state is notably not obliged to carry out administrative measures at variance with its laws and administrative practice, or to supply information that is not obtainable under its laws or in the normal course of the administration of that state (see Article 26 para. 2 (a) and (b) of the OECD Model). Therefore, the requested state is not bound to go beyond its own domestic laws and administrative practice in putting information at the disposal of the other contracting state. Third, some countries, such as Switzerland, have made a reservation on Article 26 of the OECD MTC and consider that Article 26 only obliges a state to grant information necessary for the correct application of the treaty, and not for the application of domestic law (see para. 24 of the Commentary on Article 26 of the OECD Model). However, based on recent negotiations with treaty partners, a new development in this respect is for Switzerland to grant assistance also in the case of “tax fraud” (see Article 26 of the 1997 United States–Switzerland treaty and the protocol of 7 December 2001 to the Germany–Switzerland treaty; Oberson 2003). In general, Article 26, para. 5 of the OECD MTC states that provisions of a contracting state which permit it to decline to supply information solely because the information is held by a bank, other financial institution, nominee, or person acting in an agency or a fiduciary capacity, or because it relates to ownership interests in a person, may not be in line with the aims and purpose of the OECD MTC. However, Article 26, para. 5 does not preclude a contracting state from invoking Article 26, para. 3 to refuse to supply information held by a bank, financial institution, a person acting in an agency or fiduciary capacity, or information relating to ownership interests. Such a refusal must be based on reasons unrelated to the person’s status, such as a bank, financial institution, agent, fiduciary, nominee, or the fact that the information is related to ownership interests. 3.5.1.4  Interim results on OECD principles on exchange of information The traditional economic model of tax evasion states that tax evasion is determined by the probability of detection, the penalty rate, the tax rate and the gross

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

58   D. Aigner and M. Tumpel income of the investor. By accepting these as the relevant factors influencing tax evasion, the OECD principles on exchange of information impact only the probability of detection. As OECD principles only provide for an exchange of information upon request, it is uncertain whether there will be a significant increase in the probability of detection. As fishing expeditions are not allowed under OECD rules, the state of residence of the taxpayer may need to have information about unreported income from other sources to be able to apply for an exchange of information under the Tax Information Exchange Agreement (TIEA) or Article 26 of the OECD MTC. Furthermore, existing possibilities to refuse an exchange of information based on the OECD information exchange systems and uncooperative behavior by the tax havens may have a negative impact on the probability of detection (Gläser and Halla 2006). By considering these factors, it can be predicted that the introduction of the OECD principles on exchange of information will not cause a significant increase in the probability of detection. If this is true, then the OECD principles will not lead to a decrease in tax evasion, because beneficial owners willing to evade taxes are not confronted with a significant increase in the probability of detection (Gläser and Halla 2006). Furthermore, the fact that small countries that are net importers of capital (typically tax havens) can never gain from information sharing as much as net exporters of capital will be an important issue in the future. Although all tax havens identified by the OECD have made formal commitments to the OECD principles on exchange of information in tax matters, an effective exchange of information would require the full cooperation of low-­tax jurisdictions. As tax havens providing full information to a foreign tax authority will face allocative disadvantages, because they will become less attractive to foreign investors than those that are less cooperative, a tax-­related exchange of information seems to be a serious drawback for tax havens. As there is almost no profit from tax-­ related information that they receive in return, and as the administrative costs of gathering and transmitting information are not compensated, there is almost no incentive for a low-­tax jurisdiction to engage or fully cooperate in a tax-­related information exchange system. The partial existence of loopholes and the possibilities to refuse an exchange of information may result in these countries showing a slightly hidden non-­cooperative behavior. 3.5.2  Savings Directive 3.5.2.1  Aims of the Savings Directive The automatic exchange of information under the Savings Directive aims to guarantee that the state of residence of an individual gets information about interest payments that the individual receives from paying agents in other Member States of the European Union. Contrary to this so-­called information procedure, Austria, Belgium and Luxembourg are operating the withholding tax procedure during a transitional period. In these countries, the deduction of with-

How to combat tax evasion in tax havens?   59 holding tax on cross-­border interest payments will guarantee a minimum of effective taxation by avoiding non-­taxation if the interest income is not declared by the beneficial owner in his/her Member State of residence. Hence, the basic aim of the Savings Directive is to minimize the opportunities to evade taxes, and thereby to decrease effective tax evasion while increasing public revenues.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

3.5.2.2  Interim results on the Savings Directive As the Savings Directive basically provides for an automatic exchange of information, it seems to be more effective in avoiding tax evasion by individuals than the above-­mentioned OECD principles. However, literature on the savings tax directive has shown that there are plenty of legal and illegal ways to circumvent the information procedure or withholding tax procedure under the Savings Directive. By considering the basic aim of the Savings Directive, one has to note that the legal or illegal circumvention of the regulations of the Savings Directive does not necessarily mean an effective evasion of taxes by the beneficial owner. A successful circumvention of these rules only enables the beneficial owner to conceal interest income from his/her state of residence, as if the Savings Directive had never been introduced. The beneficial owner has the option to fully declare or conceal the interest income. However, in that case the introduction of the Savings Directive may not cause any increase in the probability of detection (Gläser and Halla 2006). There are plenty of loopholes provided by the Savings Directive itself and its national implementations which enable investors, irrespective of whether or not they prefer European financial markets, to circumvent the regulations to avoid the intended information transfer or withholding tax deduction (Aigner 2009; Gläser and Halla 2006). Therefore, it can be concluded that the Savings Directive will not lead to a significant decrease in tax evasion, because beneficial owners willing to evade taxes are not confronted with an increase in the probability of detection. Furthermore, the Savings Directive has no influence on the penalty rates in the Member States and does not lead to an increase in the tax rates in the Member States.

3.6  Conclusions The traditional economic model of tax evasion states that tax evasion is determined by the probability of detection, the penalty rate, the tax rate and the gross income of the investor. Empirical and experimental evidence has shown that the addition of moral aspects can fruitfully extend the standard economic model. If one accepts these as the relevant factors that influence tax evasion, the OECD principles on exchange of information and the Savings Directive of the European Union do affect the probability of detection. Nevertheless, the OECD principles and the measures of the Savings Directive might not necessarily have succeeded in increasing the probability of detection. Considering the highlighted possibilities to refuse an exchange of information and the loopholes provided by the Savings Directive itself and its national implementations, investors, whether they

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

60   D. Aigner and M. Tumpel prefer European financial markets or not, have plenty of opportunities within the European financial markets to circumvent the regulations to avoid the intended information transfer. As tax-­related information exchange is a strategic variable used by the EU Member States to alter the attractiveness of their financial markets, it is uncertain whether the partial existence of loopholes in the national implementations is due to slightly hidden non-­cooperative behavior on the part of the countries involved. Unless the OECD and the European Union do not enforce cooperative behavior, an outflow of funds may not necessarily take place (Gläser and Halla 2006). However, in their content the OECD and EU principles on exchange of information are a balanced and refined instrument. A significant advantage of exchange of information and, by extension, assistance in recovery, etc. is that there is less need to harmonize national tax systems. An exchange of information would be far more effective for countries than modifying their respective national tax systems (Barnard 2003).

Notes 1 Anguilla, Aruba, the British Virgin Islands, the Cayman Islands, Guernsey, the Isle of Man, Jersey, Montserrat, the Netherlands Antilles and the Turks and Caicos Islands. 2 According to Article 10, para. 2, the transitional period will end at the end of the first full fiscal year following the later of: (i) the date of entry into force of an agreement between the European Community and the last of the key third countries (the last of the Swiss Confederation, the Principality of Liechtenstein, the Republic of San Marino, the Principality of Monaco and the Principality of Andorra) to provide exchange of information upon request as defined in the OECD Model Agreement on Exchange of Information on Tax Matters, released on 18 April 2002, with respect to interest payments, as defined in this directive, made by the paying agents established within their respective territories to beneficial owners residing in the territory to which the directive applies, in addition to the simultaneous application of a withholding tax, by those same countries, on such payments at the rate defined for the corresponding periods referred to in Article 11 para. 1; or (ii) the date when the European Council unanimously agrees that the United States is committed to the exchange of information upon request as defined in the OECD Model Agreement, with respect to interest payments, as defined in this directive, made by paying agents established within its territory to the beneficial owners residing in the territory where the directive applies. 3 This option may be either the authorization for the paying agent by the beneficial owner to report the interest payment to the state of residence, or the presentation of a certificate issued by the competent authority of the state of residence, which proves that the beneficial owner declared the existence of foreign income. 4 Tax evasion and tax avoidance may significantly differ if one introduces moral aspects into the analysis. Empirical evidence on that is given by Kirchler et al. (2003). 5 The expressions “payoff ” and “utility” are used interchangeably throughout this text. 6 Before 2000, Article 26 para. 1 of the OECD Model authorized the exchange of information only in relation to the taxes covered by the OECD Model under the rule in Article 2. Article 26 para. 1 has been amended so as to apply to any taxes imposed by a contracting state (see para. 11.1 of the Commentary on Article 26 of the OECD Model).

4 Double tax avoidance and tax competition for mobile capital

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Markus Leibrecht and Thomas Rixen

4.1  Introduction Capital tax competition has received increasing attention in recent years, not least because capital mobility, both de jure and de facto, has risen. It is frequently argued that the attraction of mobile capital, other things being equal, creates welfare gains for a country as additional tax revenues accrue and/or the domestic capital and technology stocks increase. Thus, it is not surprising that countries strive to increase the capital stock within their borders. There are two largely distinct bodies of literature and corresponding research areas in international taxation. On the one hand, there is a large and continuously growing body of theoretical and empirical literature on tax competition (see, for example, Wilson and Wildasin 2004; Hochgatterer and Leibrecht 2009), mainly in the field of economics; on the other, there is a large body of mostly legal literature dealing with international double tax avoidance (see, for example, Lang 2003). The fact is, however, that there are only very few contributions that address both issues at the same time (see, for example, Roin 2008). This chapter tries to narrow the gap. Specifically, it shows why measures of double tax avoidance, even though they comprise measures of bilateral tax coordination, may actually act as instruments of tax competition.1 The rules of double tax avoidance (DTAV) are contained in bilateral double tax treaties (DTTs), in domestic tax rules and, in the European Union, in supranational tax laws. All these different levels of rules are influenced by conventions promoted by international organizations. This set of rules is designed to ensure that international investments are not over-­taxed in comparison with national investments. The goal is to achieve “single taxation” by coordinating different countries’ taxing rights so as to avoid jurisdictional overlap. However, one consequence of the way DTAV rules are constructed and the way the various elements interact may be “double non-­taxation”: sophisticated taxpayers can use the rules in a way that helps them to minimize their tax payments. They engage in so-­called tax arbitrage across different jurisdictions and DTAV rules. However, the fact that taxpayers exploit differences in the rules in order to  minimize their tax payments is by itself not sufficient to diagnose tax

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

62   M. Leibrecht and T. Rixen competition. An additional condition is that governments use taxpayers’ arbitrage activities to attract capital investments.2 This becomes clear if one considers that horizontal tax competition is defined as non-­cooperative tax setting by independent governments, where each government’s respective policy choices influence the allocation of mobile tax bases among the regions or jurisdictions represented by these governments (Wilson and Wildasin 2004). This means that independent jurisdictions interact strategically in setting taxes. Moreover, it implies that at least three conditions must be met in order for tax competition to exist. First, governments use tax rules intentionally to attract capital or to keep it in the country. This condition can be met in various ways such as a discretionary reduction in tax rates, fashioning preferential tax regimes for foreign investors or creating (or not closing) tax loopholes. Second, capital tax bases must be sensitive with respect to tax law differences, so that there is a real effect of governmental actions on the allocation of mobile tax bases. Third, capital mobility must be de jure possible and it must de facto occur (see also Bellak and Leibrecht 2008). As has already been noted, an extensive body of empirical literature exploring the presence of tax competition has emerged,3 but this literature focuses upon just one tax policy instrument to attract mobile capital, namely, statutory or effective tax rates on corporate income. Yet tax competition might occur via more subtle channels (see also Altshuler and Grubert 2005). In particular, we argue that there is a direct link between DTAV and tax competition. The most obvious link is that if double taxation is curbed, the net rate of return on, say, a foreign direct investment (FDI) is increased. This phenomenon is expected to be particularly pronounced in the case of tax-­sparing rules in some treaties between developed and developing countries, as these rules ensure the advantages granted by specific tax provisions. Hence, with DTAV rules in force there is an increased propensity among firms to make foreign investments, and thus cross-­border capital flows are induced. A more interesting link is based on the fact that DTAV rules generally rely on legal form rather than economic substance. Thus, these rules may leave open or create opportunities (“loopholes”) for tax arbitrage, enabling businesses to shift paper profits. Although it is likely that some of these loopholes are unintended, it is also conceivable that governments actively use DTAV rules to tailor specific offers to taxpayers’ needs in order to attract tax base. For example, governments can offer corporations the possibility to set up brass plate (“letterbox”) companies or other artificial legal entities; this encourages these enterprises to engage in tax residence “shopping”. Alternatively, governments may be reluctant to include thin capitalization measures in their DTAV rules in order to make their country more attractive for real capital investments (see, for example, Haufler and Runkel 2008). Thus, in addition to statutory and effective tax rates, governments may also use DTAV rules as instruments for competing over mobile tax bases. This chapter is structured as follows. In section 4.2, we outline the principles and rules of international double tax avoidance and explain how they can be

Double tax avoidance   63 used for purposes of tax arbitrage. Moreover, we outline why DTAV rules can provide an important institutional foundation of tax competition. After that, we review in section 4.3 the empirical literature suggestive of the role of DTAV rules in tax competition. Finally, we summarize our analysis and draw some preliminary policy conclusions.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

4.2  Double tax avoidance as the institutional foundation of tax competition In this section, we show how DTAV rules provide an institutional foundation of tax competition. We first describe the principles and rules of DTAV that operate on the domestic, bilateral and multilateral levels (subsection 4.2.1). We then move on to show how the complex set of DTAV rules can be used for purposes of tax arbitrage by taxpayers (subsection 4.2.2). Subsection 4.2.3 will then illustrate how governments react to taxpayers’ arbitrage activities and position themselves in the resulting tax competition. This description also provides a stylized historic account of the development of the international tax system. 4.2.1  Principles and rules of double tax avoidance Double taxation is defined as the imposition of comparable taxes in at least two countries on the same taxpayer with respect to the same subject matter and for identical periods (Baker 2001). It results from an overlap of jurisdiction to tax between a residence state, where the recipient of income lives, and a source state, where the income was generated. If both states exert their power to tax to the full extent, the total burden on trans-­border economic activities is prohibitively high. In order to obtain the benefits of international liberalization, governments have a common interest in avoiding double taxation. The basic conflict is: which country has the right to tax the income, and which country must restrict its tax claims (see, for example, Li 2003)? While there were some domestic provisions and a few DTTs in continental Europe prior to World War I, after the war the problem became more pressing when many countries put taxation on a broader basis, including general income taxes. In the 1920s, the International Chamber of Commerce put the issue of double taxation on the international agenda. In response to this, the League of Nations commissioned reports by experts and convened several conferences of technical experts and government officials. During the League years, the basic principles and rules were developed that have guided the avoidance of double taxation up to today. In the 1950s and 1960s, the OECD took over the position of the League of Nations (and briefly the United Nations) as the main multilateral policy forum for discussions of international tax issues (Rixen 2008: 86–99; Picciotto 1992: 14–35). Initially, the objective of these activities was to draft a multilateral tax treaty. But governments persistently rejected the idea of a binding multilateral treaty. They were nonetheless supportive of developing a model convention (MC) that

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

64   M. Leibrecht and T. Rixen could be employed as a template for bilateral DTTs. They insisted on keeping the MC non-­binding to allow the necessary flexibility to make nationally differing tax systems compatible with one another (Rixen 2008, 2010). The development of the MC was marked by a conflict over the allocation of taxing rights. Should the MC be based on the residence or the source principle? Both principles can be normatively justified. Those emphasizing the principle of ability to pay will favor residence taxation because the residence country is in a better position to assess the taxpayer’s worldwide income. Conversely, if one regards the benefit principle as the more appropriate standard, this would suggest source taxation because the source country provides infrastructure that allows the generation of income in the first place. Both of these arguments are simple and intuitive. None of the scholars who have discussed the issue of a desirable allocation of taxing rights has come out in favor of one or the other principle; rather, they have favored some solution that accords different weights to these considerations (see, for example, Musgrave 2006; Li 2003). This normative indeterminacy was further aggravated by a distributive conflict between net capital importers and net capital exporters. Exporting countries favor the residence principle, whereas importers favor source taxation. In each case, the respective principle would result in the allocation of a bigger share of the international tax base to one or the other country (see, for example, Dagan 2000; Davies 2004).4 Accordingly, it has not been possible to achieve a general consensus on either principle. Instead, the solutions embodied in the various MCs that have been developed since the 1920s (which have remained fundamentally unchanged from that time right up to today) represent a compromise. Broadly speaking, the primary (or exclusive) right to tax active business income is granted to the source country (Article 7 of the OECD MC) if there is at least a “permanent establishment” (Article 5 of the OECD MC) – that is, a fixed place of business – in that country. The residence country, by contrast, has the primary right to tax passive income ­– that is, income from financial investment such as interest, dividends or royalties – with the source country having the right to levy withholding taxes (Articles 10–12 of the OECD MC).5 Overall, the MC differentiates between various kinds of income and assigns each to either the source country or the residence country. In general, the OECD MC places greater emphasis on residence taxation.6 For passive investment income, the rates for royalties are different from those for dividends or income. The OECD MC is thus based on a schedular system of taxation. The residence country is obliged to provide relief from double taxation in cases of full or limited source taxation. This can be done by granting either a foreign tax credit for the tax paid at source on the tax due in the home country or a full exemption of that income from home tax. In addition, bilateral treaties contain provisions on information exchange between tax administrations (Article 26 of the MC). This is the only provision of DTTs that is not primarily concerned with the avoidance of double taxation but focuses instead on the problem of potential tax evasion and avoidance.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Double tax avoidance   65 In parallel to the drafting of the non-­binding multilateral MC (and in some cases even prior to it), governments have developed domestic tax rules to avoid double taxation (Rixen 2008). Basically, all countries provide for relief of double taxation unilaterally. Their respective national tax codes contain provisions for exempting foreign income from taxation, crediting the taxpayer or granting a deduction (partial relief ) for taxes paid on such income in the source country. In addition, governments conclude bilateral DTTs on the basis of the MC. Basically, all of the more than 3,000 bilateral tax treaties adopt the basic structure and the provisions of the MC (Owens 2008). Thus, most DTTs are similar in terms of the topics covered, their structure and even the language used. Seventy-­ five percent of the wording in any given DTT will be identical to that of any other DTT (Avi-­Yonah 2007). Often, technical innovations that come up in bilateral treaties or in domestic provisions are integrated into the multilateral MC so that experience gained at the national and bilateral levels can be disseminated multilaterally to all parties. Other innovations are initiated by the OECD, which constantly strives to modernize and adapt the MC. Through this interaction of domestic, bilateral and multilateral activities, a common understanding of double tax avoidance is developed.7 In other words, the complex of rules which we summarily refer to as DTAV rules consists of domestic rules of international taxation and of binding bilateral DTTs, both of which are informed by the principles developed within multilateral policy forums, most importantly the OECD (Rixen 2010). In line with the interests of governments, the particular solution embodied in the MC is sovereignty-­preserving – that is, it ensures that countries are as free as possible to apply their own national tax laws (Rixen forthcoming). The MC defines a series of legal constructs intended to allocate the right to tax among the jurisdictions involved (see, for example, Bird 2000). For example, the concept of a permanent establishment (PE) codifies what is taxable as a separate entity of a multinational enterprise (MNE) in the country of source. This and other constructs are chosen in such a way as to interfere as little as possible with national tax laws. They merely allocate the right to tax to governments, without prescribing whether or how they ought to exercise this right. Governments retain full authority to design all elements of their respective tax regimes – namely the tax base, rate and system – independently of other governments. The idea is that of territorial disentanglement of different tax systems. There is no harmonization of national tax systems, but only coordination of different regimes. In this sense, the term “international tax” is a misnomer, since there is no overriding international law of taxation, but only rules of allocation that operate at the interfaces of different national tax systems (Li 2003). Emblematic of the sovereignty-­ preserving setup of international taxation are the rules for allocating the profits of MNEs to the various countries in which the MNE operates. Under the “separate entity approach”, the allocation should be the same as would result if the different entities of a multinational group were independent actors transacting on a market. This so-­called arm’s-length standard (ALS) makes it unnecessary to agree on a common tax base.

66   M. Leibrecht and T. Rixen Thus, while the interfaces of different tax systems in the form of DTAV rules have been harmonized to a considerable extent – even though important differences remain among bilateral tax treaties – national tax rules remain dissimilar.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

4.2.2  How double tax avoidance enables tax arbitrage Although rules of double tax avoidance explicitly address only governments and tell them how to structure their international tax relations, the same rules (implicitly) pre-­structure taxpayers’ avoidance or planning strategies. There are several techniques of tax arbitrage, which is defined as “tak[ing] advantage of inconsistencies between different countries’ tax rules to achieve a more favourable result than that which would have resulted from investing in a single jurisdiction” (Shaviro 2002: 319). First, the manipulation of arm’s-length transfer prices is a method of avoidance that builds on an important aspect of the DTT regime. According to Article 7 (2) of the OECD MC, the profits of an MNE have to be attributed to the different affiliates as if their transactions with each other were those of independent enterprises exchanging their products and services at regular market prices. However, it is often difficult to determine the “correct” market prices, since there is no regular market on which the goods in question are traded.8 Accordingly, taxpayers enjoy legal maneuvering room to set transfer prices in a way that leads to the allocation of larger shares of profit to low-­tax countries. A second tax planning technique is treaty shopping – that is, the “accessing of treaty benefits by persons who are not resident of either treaty state through the use of an entity that qualifies as a resident of one of the states” (Li 2003: 106). For example, if a company wishes to get access to cheap capital from residents of a country with which there is no tax treaty, it may be advisable to let a financial subsidiary in a treaty partner country issue the bonds (see, for example, Papke 2000). Third, the schedular structure of DTTs allows taxpayers to reclassify their financial flows in a tax-­optimal way. Most importantly, they can substitute debt for equity. Very often, this reclassification is facilitated by the possibility of hybrid financing and hybrid entities, which we discuss here in more detail for three reasons. First, the use of hybrid financial instruments provides a vivid illustration of how the structure of DTTs can be utilized to reclassify financial flows tax optimally. Second, hybrid financing can be interpreted as a sophisticated way to substitute debt for equity with the intention of circumventing thin capitalization rules (see the following subsection, subsection 4.2.3) – in many cases without actually increasing the debt to equity ratio. Third, although hybrid instruments may be issued for a variety of non-­tax reasons (Duncan 2000), taxation issues do have considerable impact on management’s financing decisions. What makes hybrid instruments such a potent tool in tax planning is the simple fact that in the majority of countries, financial instruments must be treated either wholly as equity or wholly as debt for fiscal purposes.9 In other words,

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Double tax avoidance   67 income is treated either as a distribution of profit (or dividends, in a broad sense), or as interest. This in turn is crucial for two reasons: first, the classification generally determines whether or not the issuer can treat the payments as tax deductible; second, in many cases it determines whether the income from the instruments is tax-­exempt in the hands of the recipients (Eberhartinger 2005; Helminen 2004; Six 2007; Wittendorff and Banner-­Voigt 2000). Only in rare cases is the fiscal treatment of an instrument split. De facto, a wide variety of financial instruments incorporate elements of both equity and debt.10 Because these financial tools usually cannot be classified clearly as either one or the other, they are referred to as “hybrid instruments”, or collectively as “mezzanine financing”. The spectrum of hybrid instruments ranges from corporate shares with features typical of loans (such as certain preference shares) to loans with features usually associated with equity investments (such as participation in profits and losses). Such equity-­type loans include, among other things, jouissance rights, silent partnerships, participation bonds, convertible bonds, warrant bonds, profit-­participating loans and preference shares. For fiscal authorities, the wide variety of hybrid instruments in use and their rapid evolution makes an unambiguous classification of any given hybrid instrument all but impossible. The need to apply general criteria in the classification of hybrid instruments opens up significant opportunities for tax planning. Evidently this is even more important for cross-­border transactions. Hybrid instruments are used effectively as flexible, tailor-­made forms of finance because in cross-­border situations both countries involved must decide on the classification of the instruments and could, in some cases, decide differently. Thus, if a payment is deductible as interest in the source state and exempt as remuneration of equity (dividend) in the parent company’s state of residence, the result will be double non-­taxation. In the opposite case, where the hybrid instrument is treated as equity in the source state and as debt in the state where the parent company resides, the difference in classification may result in double taxation, with the payment subject to withholding tax in the source state and to income tax in the parent company’s state of residence. This is where the DTTs should step in to prevent double taxation. The relevant distributive rules in DTTs for the treatment of income derived from hybrid financial instruments are Articles 10 (on dividends) and 11 (on interest). Articles 10 and 11 of the OECD MC give the residence state an unlimited right to tax the income received; at the same time, these rules do not deny the source state a right to tax that same income by levying a withholding tax. To avoid double taxation, the OECD MC therefore obliges the state of residence to credit the taxpayer with the tax withheld by the source state against the corporate tax that the former desires to levy. Articles 23A and 23B of the OECD MC provide for the use of the standard credit method with a per-­country limitation in these cases. The OECD MC curtails the rights of the source state by limiting the amount of withholding tax it may levy: Article 10 limits it to 5 percent in the case of associated companies11 and to 15 percent in all other cases; Article 11 limits the

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

68   M. Leibrecht and T. Rixen permitted withholding tax to 10 percent. These limits are the subject of bargaining in treaty negotiations, and individual conventions may differ considerably from the OECD MC.12 Similarly, it is also possible for a specific DTT to deny the source state the right to levy withholding tax; this is more frequently the case with interest payments than with dividends.13 Whereas the actual levels of withholding tax provided for in a specific DTT are relevant in practice, it is important to point out that DTTs may give the source state the right to levy withholding tax on both dividends and interest, on neither of them or – most importantly – on only one of them. This last case is clearly important for hybrid finance because the classification of the return as dividend or interest then determines whether withholding tax can be levied or not.14 In the standard case where both countries apply the same classification criteria to decide whether the return on a certain hybrid instrument falls under Article 10 or Article 11, it is merely a matter of tax planning to choose hybrid instruments whose returns fall under the more favorable of the two articles. More difficult to resolve is the situation where a DTT provides for a withholding tax in accordance with only one of the articles but the countries concerned do not apply the same article to the return of a particular hybrid instrument. In one particularly problematic case, the source state (country X) applies the article that provides for withholding tax (say, the dividend article) and levies the appropriate tax. The recipient’s state of residence (country Y), on the other hand, applies the article that does not provide for withholding tax. Must Y then credit the taxpayer with the tax withheld in X, even if according to Y’s interpretation of the treaty no such withholding tax is permitted?15 Although the prevailing German legal opinion (see, for example, Wassermeyer 2004) in part seems to suggest that the state of residence is not obliged to credit the taxpayer in this case, which would result in double taxation of the return despite the DTT, the issue remains unresolved. It seems reasonable to assume that some countries will credit the withholding tax while other countries will decline to do so in these circumstances. In any event, this problem makes it clear why the application of the two relevant articles is important when hybrid instruments are used, particularly if the accepted view is that the DTT must be interpreted autonomously (Lang 1993, 1998, 1999). The essential criterion for distinguishing between these two distributive rules is the existence of a corporate right. The yield of a hybrid instrument qualified as a corporate right under Article 10 is defined as a dividend in accordance with the purposes outlined in that article. In all other cases, the yield constitutes interest under Article 11. In order for an investment to qualify as a corporate right under the OECD MC, the investor must accept the risk of possible loss of the investment in a way comparable to the risk assumed by a regular shareholder. According to the prevailing German doctrine, this is the case if the investment involves participation in both the profits and the liquidation proceeds of the issuing company. Only if these two conditions are satisfied is the investor’s risk of loss comparable to that of a regular shareholder.16 The OECD MC, however, does not explicitly mention these conditions or indicate how they can be satisfied for purposes of qualifying a financial

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Double tax avoidance   69 instrument as equity. Hybrid financial instruments have numerous possible characteristics regarding participation in entrepreneurial risk, which makes it difficult to determine whether these conditions are satisfied in a particular case. This in turn will inevitably lead to situations where two contracting states disagree on whether the characteristics of a hybrid instrument are such as to qualify that instrument as a corporate right and thus as generating dividends. It is therefore possible that, in the situation of a particular taxpayer and a given form of hybrid finance, one state applies the dividend article (Article 10) while the other applies the interest article (Article 11). Depending on the relevant tax treaty, this might result in double taxation or double non-­taxation of the taxpayer’s income.17 From the perspective of tax planning this situation holds considerable potential because it implies that a group in a given set of countries may be able to choose an appropriate hybrid instrument that allows for double non-­taxation, thus creating in fact tax-­free income. An enterprise or business group with particular financial needs would be able to choose appropriate countries for its subsidiaries in order to optimize or even eliminate the tax burden on payments received (treaty shopping). With these and other similar techniques, taxpayers make sure that profits are taxable in low-­tax countries, while losses occur in high-­tax states. What these methods have in common is that none of them relies on the relocation of real business activities, but rather on the mere shifting of paper profits and losses. This form of tax arbitrage is possible because the DTT regime gives states the freedom to design their own national tax laws. The sovereignty-­preserving approach to double tax cooperation creates an opportunity structure for taxpayers to engage in tax arbitrage. 4.2.3  DTAV rules as instruments of tax competition The sovereignty-­preserving character of DTAV rules creates not only an opportunity structure for taxpayers but also one for governments to engage in tax competition. This manifests itself in different ways. To begin with, since DTTs may reduce the overall tax burden on international investment but at the same time contain provisions that increase investors’ legal certainty, governments may want to conclude such treaties in order to signal the attractiveness of their respective countries as a location for international investment (see, for example, Dagan 2000). Whereas this would make DTTs important instruments in the competition for mobile capital, clauses on the exchange of information might reduce the possibilities for tax avoidance, which could hamper FDI. But there are other ways to use DTAV rules in tax competition. Most importantly, tax haven governments make special offers to satisfy taxpayer demand for tax minimization. This is possible because DTAV rules set no limits on how governments can design their national tax legislation and because, in general, legal form and not economic substance is sufficient for establishing tax residence. Also, governments actively design solutions that can help taxpayers to play out the differences among different DTTs or different domestic laws and

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

70   M. Leibrecht and T. Rixen DTTs. This is most easily visible if one considers the role tax havens play in the tax competition game. Tax-­haven governments offer very low or zero tax rates in combination with tailored solutions, making it easy for businesses to establish legal tax residence under their jurisdiction. One way of doing this is to allow the incorporation of brass plate businesses which serve no substantive economic purpose, merely the tax-­privileged holding of assets for a multinational group. Tax havens actively search for market niches and try to specialize in different tax planning activities.18 This illustrates that tax havens are in competition with each other and with high-­tax countries for the assignment of so-­called paper profits. This also implies that high-­tax countries come under competitive pressure; they must take the threat of the erosion of their tax base into consideration in the design of their tax systems.19 Many high-­tax countries reacted to tax arbitrage with a combination of trying to curb it and efforts to remain competitive vis-­à-vis other countries. With respect to curbing tax arbitrage, governments have developed unilateral measures against tax avoidance. One example is the so-­called controlled foreign companies (CFC) legislation, which targets the use of foreign subsidiaries as base or conduit companies in tax havens. In the 1960s, the United States was the first country to introduce unilateral anti-­avoidance legislation. Resident shareholders, for example the parents of an MNE, controlling a subsidiary in a tax haven are taxable on the passive income (e.g. interest incurred from the lending of liquid funds) of the subsidiary in the current period, regardless of whether that income is actually distributed to them or not. Thus, CFC rules stretch the original DTAV principle of treating all parts of an MNE as separate national entities20 and in so doing pierce the “corporate veil” of the tax haven entity. In the meantime, such rules have diffused via the OECD to all major capital-­exporting nations (see, for example, Arnold 2000). Another example of an attempt to curb tax arbitrage is so-­called thin capitalization rules. These rules identify a threshold of excessive debt on the basis of the debt–equity ratio. The rules only apply to non-­resident lenders holding a significant percentage of the shares of the resident corporation.21 If the criteria are met, then the debt is treated as “hidden equity” and the interest payment is not deductible.22 A final example of an area in which efforts at curbing abuse have been made is transfer pricing. Again the United States was the first country to push for changes. In the 1990s, it adopted new transfer pricing guidelines; OECD recommendations were subsequently also reformed. Since then, the guidelines have been in a state of almost perpetual adaptation. Overall, the new guidelines move the actual rules closer to considering the consolidated profits of the MNE but take great care to formally reinforce the principle of separate entity accounting on a transactional basis. With the introduction of advanced pricing agreements (APAs) in many countries and their promotion by the OECD, this trend has become even more pronounced.23 However, the effectiveness of these measures is questionable. In the case of transfer prices, for instance, the manoeuvring room for taxpayers remains high

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Double tax avoidance   71 because the range of permissible prices is large. At the same time, the procedure imposes serious administrative costs on the taxpayer. As for the unilateral anti-­ avoidance measures, what can be observed is that they generally target certain circumscribed tax arbitrage schemes which taxpayers can simply circumvent by designing more sophisticated schemes. For example, CFC rules are often sidestepped by creating hybrid entities.24 What is observed in practice is a kind of proliferation spiral. Taxpayers react to anti-­avoidance rules by devising more sophisticated techniques of tax arbitrage; these in turn are then targeted by new anti-­avoidance rules by governments, and so on. In effect, governments continue to react to taxpayers’ arbitrage activities. A likely reason why governments do not design broad and effective rules against tax arbitrage is that they are under competitive pressure vis-­à-vis other high-­tax countries which are willing to leave loopholes open for their taxpayers.25 One recent example of this is the introduction of a so-­called Zinsschranke (i.e. a unilaterally binding boundary on debt financing of affiliates) in German corporation tax law. Originally, the finance minister planned to introduce a thin capitalization rule that foresaw the non-­deductibility of 50 per cent of interest expenses. However, this proposal met with opposition from business interests and parliamentarians of the Christian Democratic Union (CDU), one of the parties making up the coalition government. Eventually, the Zinsschranke passed by the government only foresees the non-­deductibility of 30 per cent of net interest expenses (those expenses above interest receipts), and this only if none of three exceptions (small company clause, group clause and escape clause (see Homburg 2007: 605) is applicable (Jarass 2007). Very recently, following continuous lobbying of business interests against the rule, the small company clause has been extended as part of the government’s stimulus package against the global recession: instead of the rule being applied to interest expenses above €1 million, the threshold has been increased to €3 million. To summarize, DTAV rules not only achieve the avoidance of double taxation but also provide an institutional foundation of tax arbitrage (for taxpayers) and, further, DTAV rules become instruments of tax competition (for home and host-­country governments).26 Some evidence for a role of DTAV rules in tax competition is presented in the next section.

4.3  Empirical evidence on the role of DTAV in tax competition In section 4.2, we outlined several channels through which DTAV rules can result in tax minimization by MNEs and thus can be used for tax competition by governments. This section tries to assess the empirical relevance of these channels. Unfortunately, no direct evidence is available, underscoring the role of DTAV rules in tax competition for mobile capital, so we must rely on indirect evidence here. Following Griffith and Klemm (2005), direct evidence for tax competition is based on empirical models relating a country’s tax rate on

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

72   M. Leibrecht and T. Rixen corporate income to the weighted tax rate of competitor countries (i.e. on reaction-­function approaches). The seminal paper in this respect is Devereux et al. (2004). A direct study would, thus, relate rules of DTAV by one country to that of competitor countries.27 However, the papers surveyed can provide indirect evidence for the role of DTAV rules in tax competition as they focus on a precondition for tax competition to set in, namely the sensitivity of mobile capital (especially FDI and paper profits) with respect to tax conditions. Specifically, we survey (i) empirical papers about the effect of DTTs on FDI, (ii) papers on the relevance of tax-­sparing rules for FDI to developing countries, and (iii) studies dealing with the empirical relevance of profit shifting, whereby a particular focus is put on transfer pricing and the impact of thin capitalization rules on debt financing of foreign affiliates. Finally, we present (iv) the main results of a study dealing explicitly with the role of “hybrid entities”. 4.3.1  Preliminary remarks The papers surveyed are structured along several dimensions (main outcomes, databases used, estimators applied, etc.).28 A particular focus is put on the derivation of the implied semi-­elasticities. Semi-­elasticities represent the percentage change in the endogenous variable (FDI, pre-­tax profit, etc.) when the exogenous variable of main interest (statutory tax rate on corporate income, dummy variable for DTT in force or not, etc.) changes by one unit (e.g. a change of 1 percentage point in the tax rate or the change of a dummy variable from 0 to 1). The derivation of semi-­elasticities is intended to make the results of the heterogeneous studies surveyed somewhat more comparable (see also De Mooij 2005). The empirical models used in the papers do not usually lead directly to semi-­ elasticities. Thus, the regression coefficients shown in the papers need to be suitably transformed in a first step. The empirical models applied in the various papers are usually based on “level-­level” or on “log-­level” specifications. Level-­ level specifications relate the endogenous variable in levels (e.g. FDI in millions of euros) to the independent variable of main interest, also measured in levels (e.g. the statutory tax rate on corporate income measured as a percentage or a dummy variable indicating the presence of a DTT). In contrast, log-­level specifications have the dependent variables in logarithmic form (e.g. log(FDI)). The advantage of the latter specification is that the implied semi-­elasticity is made directly visible by multiplying the regression coefficient by 100.29 In the case of level-­level specifications, the implied semi-­elasticity can be deduced by: ε(s) = â/(y_mean) × 100

(1)

with â being the regression coefficient and y_mean the sample mean value of the endogenous variable.

Double tax avoidance   73 Moreover, some empirical models are based on elasticities (ε) as regression coefficients (“log-­log specifications”). In these cases the implied semi-­elasticities are derived as:

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

ε(s) = ε/(x_mean) × 100

(2)

with ε = elasticity = regression coefficient and x_mean being the sample mean value of the independent variable of main interest Not all papers provide the information necessary (e.g. the mean values of y and x) to derive the semi-­elasticities from the regression coefficients. Therefore, only papers providing this information are included in the survey. 4.3.2  A brief summary of empirical studies 4.3.2.1  Double taxation treaties and foreign direct investment Empirical papers dealing with the impact of DTTs on FDI are quite scarce. This is even more so for studies analyzing the impact of tax-­sparing rules contained in DTTs on FDI to developing countries. Basically, signing a DTT might on the one hand lead to more FDI because double taxation of income is avoided and uncertainty is reduced. On the other hand, however, DTTs might also reduce the possibilities for tax evasion and avoidance (Egger et al. 2006). Thus, empirically speaking, the effect of DTTs on FDI is ambiguous a priori. Concerning studies dealing with FDI and DTT,30 the early paper of Blonigen and Davies (2000) concludes that the signing of DTTs does not have relevant effects on the volume of FDI because most regression coefficients lack statistical significance. Thus, a semi-­elasticity of zero is plausible. In a follow-­up study (Blonigen and Davies 2002), the same authors find that FDI is lower if a new treaty is signed. This suggests that the possibilities for evading taxes via FDI are fewer if a viable tax treaty is in force.31 The analyses of Blonigen and Davies are based on level-­level specifications. In more recent work (Blonigen and Davies 2004b), the same authors argue in favor of using a “log-­level” specification. Applying the latter leads the authors to conclude that new treaties do not seem to impact positively or negatively on FDI because the regression coefficients are not statistically different from zero. To sum up, the papers of Blonigen and Davies imply that DTTs do not impact on the level of FDI. However, two aspects of the Blonigen and Davies studies are worth mentioning. First, they rely largely on data for developed economies; and second, they consider the signing of a DTT to be an exogenous event. Concerning the first aspect, Neumayer (2007) provides evidence that the presence of a DTT increases FDI in middle-­income developing countries; thus, he finds a positive impact of DTTs on FDI. Egger et al. (2006) allow the signing of a DTT to be an endogenous event. Specifically, it is more likely that countries will sign a DTT in cases where bilateral FDI is already present. Using a rather sophisticated two-­step approach, they find that the signing of a new DTT reduces FDI significantly.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

74   M. Leibrecht and T. Rixen This result is consistent with the view that DTTs effectively reduce the possibilities to avoid taxes via FDI. If we turn to recent studies on the topic, Barthel et al. (2008) conclude that DTTs lead to substantial and significantly higher FDI stocks. The distinguishing feature of this study is the usage of a large dataset containing both developed and developing countries. Finally, Davies et al. (2009) again find little evidence for an effect of DTT on the level of total sales acting as proxy for foreign investment. However, they do find evidence supporting the contention that treaty formation has a positive impact on the probability of new investment in a country. Thus, tax treaties seem to have no effect on the volume of investment (given that a firm has already invested in a country), but they do seem to have an impact on the location of the investment.32 So what can we conclude from these papers with respect to our assertion that DTAV rules create an institutional foundation of tax competition? Since the more recent studies find evidence that DTTs positively impact FDI (especially to developing countries) and new investments, then one precondition for tax competition to set in – namely, the sensitivity of FDI to changes in the taxation environment – seems to be fulfilled. Put differently, without DTTs, less capital would move across borders, and tax competition for FDI between countries would be of less importance.33 However, one must bear in mind that the results of Egger et al. (2006) are evidence against our assertion. Their results imply that DTTs are effective in reducing the possibilities for MNEs to avoid taxes and that DTTs are not instruments of tax competition for governments. To sum up, the available literature supports our assertion if developing countries are included in the sample and if the impact of DTTs on new investments is considered. 4.3.2.2  Tax-­sparing clauses and foreign direct investment Tax-­sparing clauses are sometimes included in DTTs signed by developed countries, in particular if applying the credit system in international taxation (e.g. Japan or the United Kingdom), and developing countries. These rules have the aim of allowing foreign investors to partake of the special tax advantages provided by the host countries of FDI. Put differently, if a tax-­sparing clause is included in a DTT, then the home country of FDI calculates the residual tax liability of the investor as if the special tax advantage were not provided by the host country (see Hines 1998). Only two empirical papers on this issue are available so far. Both are based on Japanese FDI data and both find a huge impact of tax-­sparing rules on FDI (see the semi-­elasticities contained in Appendix A2, p. 86). This evidence is consistent with the results of Neumayer presented above, which show a significant impact of DTTs on FDI to developing countries. Thus, the studies surveyed provide fairly solid evidence for the positive impact of DTTs on FDI to developing countries, especially if they include tax-­sparing clauses. Hence, overall the evidence is in favor of DTTs making capital mobile across borders. Thus, DTAV rules are capable of spurring tax competition, which supports our assertion that these rules are an institutional foundation of tax competition.

Double tax avoidance   75

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

4.3.2.3  Tax-­motivated profit shifting As has already been outlined, various means to shift profits from high‑ to low-­ tax countries exist. These include transfer pricing, debt financing of affiliated companies or the location of intangible assets. Profit-­shifting incentives are exerted by differences in statutory tax rates on corporate income across countries (see, for example, Weichenrieder 2009). These differences lead to tax-­ motivated changes in the location of pre-­tax profits via the various means to shift profits. The papers surveyed here are separated into three categories: (i) papers dealing with profit shifting in general, (ii) papers dealing with profit shifting via transfer pricing or the location of intangible assets, and (iii) papers dealing with the effect of thin capitalization rules on debt financing of firms. The papers summarized in Appendix A3 relate a proxy for a firm’s declared profits (pre-­tax or net of tax, depending on the study) to the relevant (home or host country) statutory tax rate on corporate income.34 De Mooij’s paper (2005) also includes a survey of empirical studies. The author calculates a semi-­ elasticity of −2, meaning that a 1 percentage point shift in the relevant tax rate changes declared profits by about 2 percent. More recent studies (Weichenrieder 2009; Huizinga and Laeven 2007) broadly confirm this evidence even if the implied semi-­elasticities are somewhat lower (about +1 and −1.30).35 Concerning our assertion, these results imply that profit shifting is an empirically relevant phenomenon and that DTAV measures are not effective in curbing it. However, not curbing profit shifting might be in the interest of governments because this could lead to an inflow of tax revenues (in a low-­tax country) or because profit-­shifting opportunities probably make real capital less sensitive to differences in statutory tax rates across countries.36 The papers whose results are shown in Appendices A4 and A5 shed more light on the various channels of profit shifting (transfer pricing, location of intangibles and thin capitalization of foreign affiliates). Clausing (2003) was among the first to address the transfer pricing issue directly. Clausing relates the export and import prices of US multinational firms to the statutory tax rate on corporate income of the host country of a foreign affiliate. She finds substantial evidence for tax-­motivated transfer pricing in US intra-­firm trade prices. Specifically, the lower a host country’s tax rate, the lower are US intra-­firm export prices and the higher are US intra-­firm import prices relative to corresponding arm’s-length prices. Based on Clausing’s results, semi-­elasticities of about +4 for export prices and −4.7 for import prices are derived (cf. Appendix A4). Bernard et al. (2008) also give direct evidence for the importance of tax-­ motivated transfer pricing. Their paper shows that the wedge between the arm’slength export price and the intra-­firm export price of US-­based multinationals is negatively related to the taxation of corporate income in the foreign import country. This response indicates that tax-­motivated transfer pricing is present in the intra-­firm export prices of US-­based multinational companies. The mean semi-­elasticity derived is −1.25. This value implies that a 1 percentage point

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

76   M. Leibrecht and T. Rixen increase in the host country’s tax rate on corporate income leads to a decrease in the wedge between arm’s-length export prices and intra-­firm export prices of 1.25 percent. Dischinger and Riedel (2008) analyze the tax sensitivity of the location of intangible assets with respect to the statutory tax rate on corporate income. They show that the location of holdings and investments in intangible assets is indeed inversely related to the statutory tax rate on corporate income. This relationship can be considered as indirect evidence in favor of tax-­motivated transfer pricing in the intra-­firm trade of intangibles. The derived semi-­elasticity is −1.20.37 Concerning our assertion, these results collectively imply that profit shifting via transfer pricing is an empirically relevant phenomenon and that stipulations concerning transfer pricing contained in DTAV rules are by no means effective in curbing it. However, as in the case of profit shifting in general, it might be in the interest of governments not to inhibit profit shifting through transfer pricing. Turning to papers dealing with the impact that thin capitalization rules exert on the financing structure of affiliated companies, we see that such rules do indeed have a significant effect on the level of debt financing (cf. Appendix A5). The results indicate that the introduction of thin capitalization rules has a pronounced effect on the debt-­to-asset ratio of an affiliated company. This means that governments could reduce the amount of profit shifting via debt financing with relative ease.38 But even in the European Union, as of 2007 seven Member States had not applied thin capitalization regimes. Worthy of note here is that among those countries which have not applied such rules are many smaller nations that are known to be among the more aggressive tax competitors (see Haufler and Runkel 2008; Dourado and de la Feria 2008). Furthermore, the rules in operation in the other EU member countries vary according to the method adopted, the scope of application and their effects (see Dourado and de la Feria 2008). Not applying thin capitalization rules might be considered as an effort by governments to attract foreign real capital (see also Altshuler and Grubert 2005; Haufler and Runkel 2008), most likely coupled with a positive effect on the domestic capital stock and with positive technology spill-­over effects.39 Again, these results favor our conjecture. 4.3.2.4  The role of hybrid entities As is suggested above (in section 4.2), tax minimization using hybrid entities or hybrid securities is likely. But empirical evidence on the role of hybrid instruments is scarce. One notable exception is the paper of Altshuler and Grubert (2005), who analyze the effects of a specific amendment to CFC rules in the United States in 1997. In particular, new regulation in 1997 introduced “hybrid entities”. Altshuler and Grubert describe these as “business operations that are regarded as corporations by one country while being an unincorporated branch to another” (ibid.: 5). Hybrid entities allow US companies to

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Double tax avoidance   77 avoid the current tax under the CFC rules on intercompany payments (e.g. interest and royalties) made from an affiliate in a high-­tax country (H) to another affiliate in a low-­tax country (L). At the same time, country H also allows such payments to be deductible. The new regulations are usually referred to as “check-­the-box” rules because they give firms the possibility simply of checking a box on a tax form to either identify an entity as a separate corporation or classify it as the branch of another corporation (see Altshuler and Grubert 2005 for details). The introduction of these rules in 1997 provides sufficient varience to distill the empirical importance of hybrid entities. On the basis of this change in US national tax law, Altshuler and Grubert provide evidence that MNEs could save huge amounts of taxes by using hybrid entities. One piece of evidence in this respect is based on the comparison of the tax-­ rate sensitivity of declared pre-­tax profits by US foreign affiliates with respect to the statutory tax rate on corporate income in the host country before and after the introduction of check-­the-box. Altshuler and Grubert show that the tax-­rate sensitivity in 2000 (i.e. after the introduction of check-­the-box) was higher in absolute value than it was in 1996 (the year before the introduction). The corresponding semi-­elasticities given in Table A6 are −1.48 for 1996 and −2.07 for 2000, respectively. This difference in the magnitudes of the two mean values is suggestive evidence in favor of an increase in tax-­motivated profit shifting via hybrid entities after the 1997 rule change. Specifically, according to Altshuler and Grubert, it is possible that highly profitable subsidiaries in high-­tax countries became part of a consolidated hybrid entity based in a tax haven or that “high-­ tax host countries may have reacted to the increasing tax sensitivity of investment by easing up on their transfer pricing and thin capitalization rules in order to attract mobile corporations” (2005: 17). Another piece of evidence contained in Altshuler and Grubert which is consistent with the increasing importance of hybrid entities is that “there are less CFCs in the 2000 sample than in 1996” (2005: 16 footnote 3). Moreover, the authors find that the total tangible capital in all locations grew by 28 percent from 1996 to 2000. But in five low-­tax countries, tangible capital grew by almost 200 percent. According to Altshuler and Grubert, most of this growth “reflects the consolidation of the low-­tax CFC with the operations of a high-­tax affiliate” (ibid.: 18). Concerning our assertion that DTAV rules provide an institutional foundation of tax competition, the example of the 1997 check-­the-box change in US CFC rules suggests that unilaterally applied measures (in this case by the home country of FDI) make it easier for MNEs to avoid taxes. The introduction of check-­the-box can also be considered as a home-­country instrument to attract – or at least to avoid losing – firms because the ability to hide profits makes it more likely that a company will not opt to relocate its headquarters. To sum up, the indirect evidence provided in this section suggests that DTAV rules do indeed serve as an institutional foundation of tax competition: DTTs seem to increase the propensity of firms to become multinational via FDI. DTAV without

78   M. Leibrecht and T. Rixen

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

thin capitalization rules, despite their effectiveness in curbing profit shifting, can be seen as an example for host countries using not only statutory tax rates to compete for mobile capital but also more subtle channels. The presence of tax-­motivated transfer pricing via intra-­firm trade or through the location of intangibles, despite the existence of DTTs and transfer pricing regulations, is another example in this respect. The check-­the-box regulation represents an example for the role of unilateral DTAV rules introduced by a home country as instrument of tax competition (for the latter, see also Altshuler and Grubert 2005).

4.4  Summary and conclusions This chapter explores the role DTAV rules play in tax competition for mobile capital. We first sketched various channels, introduced by DTAV rules, through which taxpayers can minimize tax payments and which also constitute channels for governments to compete for mobile capital. Second, the overview of existing empirical research provides indirect evidence for our assertion that tax competition via subtle DTAV-­related channels exists. We may conclude that DTAV rules play a prominent role in the tax competition game. The deeper reason for this lies in the patchwork structure of DTAV rules. The fact that there are differences among the various domestic and bilateral rules creates possibilities for tax arbitrage, which in turn creates tax competition among countries and thus reinforces the differences among the tax rules. What policy conclusions follow from these findings? Specifically, since DTAV rules also comprise measures of tax coordination (e.g. DTTs), what conclusions can then be derived concerning the discussion on the virtues and pitfalls of tax coordination and tax harmonization? The answer depends on the welfare effects of tax competition. If one sees tax competition as a process to restrain selfish governments or to induce policy innovations, then the answer would probably be “do nothing because there are no virtues of tax coordination”. However, if one sees tax competition as a process leading to welfare decreases, then it is clear that the current sovereignty-­ preserving approach to tax coordination is insufficient to address the issue of tax competition effectively. As long as governments remain free to determine central aspects of their respective tax systems independently, they will always have an incentive to tailor their tax regimes to the specific needs of taxpayers wishing to optimize their international tax payments. Neither unilateral measures (which governments will often also use as strategic instruments in tax competition) nor bilateral tax treaties can be effective instruments to address the inherently multilateral problems of tax arbitrage and tax competition. What would be needed instead is tax coordination by delegation40 – that is, central parameters of tax systems would have to be determined and administered by an organization equipped with supranational taxation authority. Specifically in the European context, tax laws could be coordinated at the EU level. The focus of supranational tax policy should be on curbing tax competition for paper profits, since these are more mobile than real capital. Further, profit

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Double tax avoidance   79 shifting is a more obvious violation of minimal fairness norms than real tax competition.41 Under real tax competition, a taxpayer would have to pay the actual “price” demanded by the respective government, even though that price may also be under harmful competitive pressure (see, for example, Sinn 1997). This would not be the case, however, if paper profits could be shifted. In that case, taxpayers become partial free riders: they enjoy the services offered, but avoid paying the price for them by shifting their profits out of the country. Although the supranationalization of tax policy sounds utopian, in Europe discussions about just such a solution are already under way. Indeed, the Treaty of Nice has made the implementation of such measures easier. Specifically, some of the proposals for a common consolidated corporate tax base (CCCTB) foresee, inter alia, a solution to the problems linked to tax competition for paper profits:42 the tax base would be defined and administered by the European Union, whereas individual Member States would remain free to set their own respective tax rates. Thus, a central part of national tax sovereignty would be supranationalized. Alternative measures that could reduce tax competition for capital – paper profits in particular – include, for instance, the introduction of an “EU CIT” – that is, an EU-­wide corporate income tax law and tax administration with the tax and fiscal sovereignty based at EU level.43 Another measure would be the introduction of a minimum level for the statutory tax rate on corporate income as suggested in the Ruding Report (see Ruding Report 1992). But, as the current debate on the implementation of the CCCTB shows, the political will among the EU member countries to delegate even a part of their taxation rights to a higher level authority is low. Thus, given the current political environment in the European Union, supranational measures, in order to have any realistic chance of being adopted, will need to preserve member countries’ national sovereignty in taxation matters as far as possible. This, however, implies that tax competition will not be curbed effectively in the near future. Moreover, to increase the likelihood of adoption, supranational measures should be considered that target primarily tax-­induced distortions which, though not directly related to horizontal tax competition for mobile capital, nevertheless have side effects on it. The incentive to introduce measures of this type might be larger for governments. A notable example in this respect is the reduction of the distorting effects the accrual-­based classic corporate income tax system exerts, inter alia, on financing and investment decisions of firms. Several changes in the traditional corporate income tax system were suggested, discussed and in some cases even implemented (see OECD 2007 for an overview). The focus of those measures has been on eliminating the different tax treatment of equity and debt finance, for instance by introducing an allowance for the remuneration of equity finance (allowance for corporate equity, ACE) or by abolishing the deductibility of the remuneration for debt finance (CBIT, comprehensive business income tax). These measures result in increased efficiency in corporate income taxation because they achieve neutrality vis-­à-vis investment and/or financing decisions. One side effect of this is that tax planning and governmental tax competition

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

80   M. Leibrecht and T. Rixen aiming to exploit the differential treatment of debt and equity financing (e.g. the use of hybrid instruments) would become less effective under such systems. Providing members states with information concerning possible welfare impacts resulting from the coordinated introduction of a more neutral corporate income tax system by the European Commission is an important first step in this direction.44 In any case, the sorts of far-­reaching measures mentioned – in particular, a minimum statutory tax rate on corporate income, a more neutral corporate income tax system or an EU-­wide corporate income tax law and tax administration with tax and fiscal sovereignty at EU level – must be seen at present, in light of the current political environment, as long-­term options only. In the medium term, it is more likely that tax competition in the European Union will get fiercer (Genschel et al., 2010). To conclude, the given patchwork structure of DTAV rules is a good example of how tax coordination can lead to tax-­induced distortions instead of eliminating them. This is an example of a pitfall of tax coordination. The solution to such problems requires appropriate supranational action. Unfortunately, however, these are currently unlikely, given present political constraints.

Appendix A1: DTTs and foreign direct investment Blonigen and Davies (2000) Outcome(s): Bilateral double taxation treaties seem to exert a positive impact on FDI stock and affiliate sales with a time lag; US inbound and outbound FDI significantly increase with a treaty’s age Years and countries included: 1966–1992; the United States and (up to) 65 partner countries Specification and estimator applied: Level-­level specification; pooled ordinary least square and country-­pair fixed effects estimator Dependent variable: Affiliate sales in the host country, FDI stock, FDI flows Tax variable of main interest: Dummy variable equal to 1 if a treaty is in effect for the country pair in a given year (treaty dummy), Length of time a treaty has existed between the United States and a partner country in a given year (treaty age) Control variables: Gravity model and knowledge-­capital model type variables like real GDP of home and host country of FDI, real GDP per capita in home and host country of FDI, bilateral distance between countries, openness to FDI and trade, sum of home and host countries’ real GDPs, squared difference between the home and host countries’ real GDP, dummy variable for the level of a home country’s skill endowment, difference in relative skilled-­ labour abundance between home and host country of FDI

Double tax avoidance   81 Semi-­elasticities

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 1 Tax variable of interest

Semi-elasticity mean

Semi-elasticity min.

Semi-elasticity max.

Standard deviation

Treaty dummy

27.90a

−29.18

132.64

55.69

Treaty age

  2.44

−12.55

  11.62

  5.51

b

Notes a The positive sign implies that signing a treaty increases FDI, yet most coefficients (especially those derived on the basis of the gravity model) are not statistically significant; only results contained in tables 3 and 4 of the paper are used. b The positive sign implies that FDI increases with the treaty age; using only tables 5 and 6 of the paper results in semi-elasticities between 1.40 and 8.61, which correspond to the information given in the abstract of the paper by Blonigen and Davies (2000); in Blonigen and Davies (2004b), the authors argue against the validity of these results as they pick up the effect of old tax treaties on FDI which suffers from endogeneity problems.

Blonigen and Davies (2002) Outcome(s): Recent treaty formation does not promote FDI; it may even reduce investment as possibilities to evade taxes are reduced Years and countries included: 1982–1992; 24 OECD countries (not included: Mexico, the Central and Eastern European OECD Countries, South Korea) Specification and estimator applied: Level-­level specification; pooled OLS and country-­pair fixed effects estimator Dependent variable: FDI outbound stock and FDI outflows Tax variable of main interest: Dummy variable which takes the value of 1 when a bilateral tax treaty is in place between two countries. Dummy variables for “old” and “new” treaties (for example, the new treaty dummy has the value 1 if it comes into force after the starting year of the sample used and 0 otherwise) Control variables: Knowledge-­capital model type variables, openness to FDI and trade, bilateral distance between countries, various interaction terms Semi-­elasticities Table 2 Tax variable of interest

Semi-elasticity mean

Semi-elasticity min.

Semi-elasticity max.

Standard deviation

Treaty dummy (including old treaty dummy)

  82.01a

   72.43

90.84

  6.52

New treaty dummy

−56.21b

−123.61

−7.25

44.27

Notes a A positive sign implies that FDI is higher if a treaty is in force; yet this estimate probably suffers from endogeneity problems. b The negative sign implies that FDI is lower if a new treaty is signed, pointing towards reduced possibilities to evade taxes via FDI in the case of a tax treaty in force.

82   M. Leibrecht and T. Rixen

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Blonigen and Davies (2004b) Outcome(s): No systematic evidence is found that bilateral tax treaties increase FDI activity if a preferred log-­level specification is applied Years and countries included: 1980–1999; US inbound and outbound FDI from or to 88 partner countries Specification and estimator applied: Level-­level and log-­level specifications with preference for the log-­level specification; pooled OLS and country-­pairspecific fixed effects estimator Dependent variable: FDI stock and affiliate sales data in non-­financial sectors Tax variable of main interest: Dummy variable which takes the value of 1 when a bilateral tax treaty is in place between two countries, dummy variables for “old” and “new” treaties (e.g. the new treaty dummy has the value 1 if it comes into force after the starting year of the sample used and 0 otherwise) Control variables: Knowledge-­capital model type variables, openness to FDI and trade, bilateral distance between countries, various interaction terms, trade costs of both home and host country, costs of investing in the host country), log(exchange rate), log(country average tax rate on corporate income) Semi-­elasticities Table 3 Tax variable of interest

Semi-elasticity Semi-elasticity Semi-elasticity Standard mean min. max. deviation

Level-level specification: treaty dummy

  −23.93a

  −36.86

−10.99

12.94

Level-level specification: old   90.16b treaty dummy

  58.72

121.61

31.45

Level-level specification: new treaty dummy

−107.49c

−167.50

−46.90

50.23

Log-level specification: new   19.45d treaty dummy

   2.02

  47.70

16.68

Notes a The negative sign implies that FDI is lower if a treaty is in force. b The positive sign implies that FDI is higher if a treaty is already in force at the beginning of the sample period; yet the estimates are likely to suffer from endogeneity problems. c The negative sign implies that FDI is lower if a new treaty is signed, pointing towards reduced possibilities to evade taxes via FDI in case of a tax treaty in force. d The positive sign implies that FDI is higher if a new treaty is signed; yet no underlying coefficient is statistically significant; therefore new treaties seem not to positively/negatively affect FDI either positively or negatively; this is the preferred specification of Blonigen and Davies (2004b).

Double tax avoidance   83

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Neumayer (2007) Outcome(s): Developing countries (DCs) that have signed a tax treaty with the United States or with important capital exporters actually do receive more FDI from the United States and in total; however, tax treaties are effective only in the group of middle-­income DCs, not in low-­income DCs Years and countries included: 1970–2001 (bilateral US case and unilateral FDI flow case) and 1980–2001 (unilateral FDI stock case); US outbound FDI stocks in 114 host countries and FDI stocks and inflows to 120 DCs Specification and estimator applied: Log-­level specification; random and fixed effects estimation (country and country-­pair effects) Dependent variable: US outbound FDI stocks in less developed countries, total inbound FDI stocks in and FDI inflows to less developed countries, a country’s share in the total FDI stock (-inflow) in(to) all less developed countries considered Tax variable of main interest: Dummy variable for the existence of a tax treaty with the United States, FDI flow weighted cumulative number of tax treaties a DC has signed with OECD countries, where the weight is the OECD countries’ share of FDI outflows in total world FDI outflows Control variables: Dummy variable for the existence of a bilateral investment treaty (BIT) with the United States, FDI flows weighted cumulative number of BITs a DC has signed with OECD countries, log(GDP per capita), log(population), economic growth rate, dummy variable indicating whether a country is a member of the World Trade Organization (WTO), number of bilateral trade agreements a DC has concluded with the United States, the European Union or Japan weighted by their respective shares of world trade, inflation rate, measure of natural resource intensity, political constraints index Semi-­elasticities Table 4 Dependent variable used

Semi-elasticity Semi-elasticity Semi-elasticity Standard mean min. max. deviation

US outbound stock

  29.43a

  22.26

  40.92

8.21

Inward FDI stock and FDI inflows in DC

  0.54b

  0.20

  0.90

0.31

US outbound stock in lowincome DCs

−17.51c

−18.21

−16.81

0.70

Inward FDI stock and FDI inflows in low-income DCs

  0.43d

  0.00

  0.90

0.43

US outbound stock in middle-   24.22e income DCs

  16.65

  31.78

7.57

Inward FDI stock and FDI   0.68f inflows in middle-income DCs

  0.50

  0.80

0.13

Notes overleaf.

84   M. Leibrecht and T. Rixen

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Notes a The positive sign implies that the US outbound FDI stock is higher if a tax treaty exists. b The positive sign implies that the inward FDI stock and FDI inflows to DC is higher the greater the number of tax treaties signed. c The negative sign implies that the US outbound FDI stock in a low-income DC is lower if a tax treaty exists; yet none of the underlying coefficients is statistically significant. d The positive sign implies that the inward FDI stock and FDI inflows to low-income DCs are higher the greater the number of tax treaties signed; however, none of the underlying coefficients is statistically significant. e The positive sign implies that the US outbound FDI stock in a middle-income DC is higher if a tax treaty exists. f The positive sign implies that the inward FDI stock and FDI inflows to a middle-income DC is the higher the greater the number of tax treaties signed.

Egger et al. (2006) Outcome(s): Significant negative impact of newly implemented tax treaties on outward FDI stocks once the endogeneity of tax treaties is accounted for Years and countries included: 1985–2000; OECD countries Specification and estimator applied: Log-­level specification; two-­step approach with a difference-­in-difference approach in the second step Dependent variable: In the second step of the estimation the difference between bilateral outward FDI stock two years before and after a treaty is signed is used as the dependent variable Tax variable of main interest: Tax treaty dummy with entry one in the years after two countries have implemented a treaty Control variables: log(sum of GDPs), log(similarity index), absolute value of difference in log(capital to labor ratios), interaction terms, home and host countries’ government expenditures to GDP ratios Semi-­elasticities Table 5 Tax variable of interest

Semi-elasticity Semi-elasticity Semi-elasticity Standard mean min. max. deviation

Exogenous tax treaty

−10.16a

  −0.16

−0.62

6.78

Endogenous tax treaty

−19.97

−39.41

−9.90

5.91

b

Notes a The negative sign implies that new tax treaties reduce outward FDI stocks; yet none of the underlying coefficients is statistically significant. b The negative sign implies that new tax treaties reduce outward FDI stocks.

Barthel et al. (2008) Outcome(s): Tax treaties lead to substantial and significantly higher FDI stocks Years and countries included: 1978–2004; large dyadic panel data set from various sources (e.g. UNCTAD) with 30 FDI source countries (of which 10 are DCs), 105 FDI host countries (of which 84 are DCs) Specification and estimator applied: Log-­level specification; country-­pairspecific fixed effects estimator

Double tax avoidance   85

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Dependent variable: Bilateral FDI stock Tax variable of main interest: Dummy variable with entry 1 if a bilateral tax treaty is in force and 0 otherwise (treaty dummy), number of years since a tax treaty has been effective (treaty age) Control variables: log(GDP of host country), log(GDP per capita of host country), log(inflation rate), openness to trade, dummy variable taking the value of 1 if the home and the host country have signed a common regional trade agreement, dummy with entry one if a bilateral investment treaty is in force and 0 otherwise Semi-­elasticities Table 6 Tax variable of interest

Semi-elasticity Semi-elasticity Semi-elasticity Standard mean min. max. deviation

Treaty dummy

25.33a

5.34

66.53

18.80

Treaty age

  4.43

2.00

  9.00

  2.24

b

Notes a The positive sign implies that tax treaties increase FDI. b The positive sign implies that FDI increases with the treaty age.

Davies et al. (2009) Outcome(s): Little evidence for an effect of treaties on the level of total sales, which act as proxy for foreign investment; however, a positive impact of treaty formation on the probability of newly investing in a country is found – that is, a tax treaty has no impact on the volume of investment but does affect the location of the investment Years and countries included: 1965, 1970, 1974, 1978, 1986, 1990, 1994, 1998; Sweden as home country and 25 DCs as well as Australia and Luxembourg as partner countries Specification and estimator applied: Level-­level and log-­level specifications; panel probit estimator with various fixed effects (country, region, time and firm fixed effects) and fixed effects estimator Dependent variable: Micro-­data from the Research Institute of Industrial Economics is used for total affiliate sales in a host country, dummy variable indicating whether a Swedish multinational has an affiliate in a host country or not Tax variable of main interest: Dummy variable equal to 1 if a treaty is in effect for the country pair in a given year (treaty dummy) Control variables: Statutory tax rate on corporate income, GDP, GDP per capita, openness to trade, R&D intensity, dummy variable indicating the participation in a trade agreement, indicator of the size of the Swedish MNE and its experience abroad, fixed costs in terms of plant scale, age of the affiliate, dummy variable to proxy the purpose of the establishment (sales company or production unit)

86   M. Leibrecht and T. Rixen Semi-­elasticities

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 7 Dependent variable used

Semi-elasticity

Level of sales

−11.75a

Probability of investment

  16.70b

Notes a Information only for one specification is available to derive semi-elasticities; the negative sign implies that the existence of a tax treaty reduces affiliate sales; yet the coefficient is not statistically significant. b Only for one specification are the implied marginal effects given in the paper; the positive sign implies that having established a tax treaty increases the probability of having an affiliate in the partner country.

A2: Tax-­sparing rules and foreign direct investment Hines (1998) Outcome(s): Tax-­sparing rules have a substantial and significant positive influence on the level of Japanese FDI Years and countries included: 1989 and 1990; Japan and the United States with 67 partner countries Specification and estimator applied: Level-­level specification; OLS, two-­stage least squares (2SLS) Dependent variable: Japanese share of FDI (equity investment) in developing countries to total Japanese FDI (equity investment) minus the corresponding US FDI (equity investment) share Tax variable of main interest: Dummy variable for tax-­sparing rule in force (Yes/No) Control variables: log(GDP) Semi-­elasticities Table 8 Dependent variable used

Semi-elasticity mean

Semi-elasticity min.

Semi-elasticity max.

Standard deviation

FDI share

304.76a

138.23

479.01

127.33

Equity share

196.69

159.4

266.51

  37.890

b

Notes a Substantially higher values arise if the 2SLS estimator is used; the positive sign means that Japanese FDI is higher in countries with which Japan has tax-sparing agreements than with countries for which no such rules are in force. b Substantially higher values arise if the 2SLS estimator is used; the positive sign means that Japanese equity investment is higher in countries with which Japan has tax-sparing agreements than with countries for which no such rules are in force.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Double tax avoidance   87 Azémar et al. (2007) Outcome(s): Tax-­sparing provisions have a substantial and significant positive impact on the level of Japanese FDI Years and countries included: 1989–2000; Japan and 26 developing countries Specification and estimator applied: Log-­level specification; random effects estimator Dependent variable: Total Japanese FDI flows into host country Tax variable of main interest: Dummy variable for tax-­sparing rule in force (Yes/No) Control variables: GDP, labor costs, exchange rates, country risk, quality of infrastructures, natural resources, trade openness, geographical distance, regional dummy for East Asia and Pacific countries, proxy for agglomeration effects Semi-­elasticity: 180.11* * Only column 2 of table 4 in Azémar et al. (2007) is used to isolate the impact of tax sparing rules on FDI A3: Profit shifting: general evidence De Mooij (2005) Outcome(s): Includes survey of earlier papers; a mean semi-­elasticity of −2 is derived Weichenrieder (2009) Outcome(s): Empirical evidence is found for the net of taxes profitability of foreign affiliates in Germany being positively correlated with the home country statutory tax rate on corporate income; moreover, some evidence (i.e. when leverage is not considered as a regressor) is found for wholly owned German foreign affiliates being more sensitive to host-­country statutory tax rates on corporate income than non-­wholly owned affiliates abroad Years and countries included: 1996–2003; Germany and an unspecified range of host countries of German affiliates and foreign parent companies of inbound FDI in Germany Specification and estimator applied: Level-­level specification; two-­way fixed effects estimator (time- and firm-­specific effects) in inbound case, firm-­ specific effects estimator in outbound case Dependent variable: Yearly data on the net-­of-tax profit of an affiliate as a percentage of the affiliate’s total assets are used; items taken from the Midi database of the Deutsche Bundesbank, affiliates in the banking and insurance industries, holding companies as well as indirect participations are excluded (inbound case) Tax variable of main interest: Statutory tax rate on corporate income in the host country of a German affiliate, statutory tax rate on corporate income in the home country of a foreign affiliate in Germany

88   M. Leibrecht and T. Rixen Control variables: log(employment), log(sales), log(fixed assets), debt ratio, GDP growth of host country of German affiliates, domestic private credit to GDP ratio of the host country, dummy variable for ownership structure Semi-­elasticity

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 9 Case

Semi-elasticity mean

Semi-elasticity min.

Semi-elasticity max.

Standard deviation

Inbound case

  0.91a

  0.72

  1.23

0.17

Outbound case

−0.55

−0.67

−0.42

0.12

b

Notes a The positive sign means that the higher the tax rate in the country of location of the foreign parent company, the higher is the reported net-of-taxes profitability of the foreign affiliate in Germany. b The negative sign means that the tax rate responsiveness of net-of-taxes profitability for wholly owned German foreign affiliates is higher (in absolute value) than that for non-wholly owned German affiliates.

Huizinga and Laeven (2007) Outcome(s): For various European countries, empirical evidence on a substantial tax-­rate sensitivity of pre-­tax profits is found Years and countries included: Cross-­section of firms for 1999; EU-­27 less Cyprus, Malta, Latvia, Lithuania, Slovenia, Greece and Ireland but plus Norway Specification and estimator applied: Log-­level specification; 2SLS estimator with the statutory tax rate on corporate income of a country being instrumented by the difference between log(population host country of affiliate) and log(population of home country of parent), industry fixed effects included, simulation of aggregated country-­specific semi-­elasticities Dependent variable: The dependent variable is log(earnings before interest and taxes); data are taken from the Amadeus database of the Bureau Van Dijk, only manufacturing-­sector firms included Tax variable of main interest: Composite tax variable which summarizes all information about the tax rates and profits in all the countries a multinational enterprise is present in Control variables: log(fixed assets) of the affiliate, log(GDP per capita) of the country the affiliate is based in Semi-­elasticity Table 10 Semi-elasticity mean Semi-elasticity min.

Semi-elasticity max. Standard deviation

−1.31

−2.92

−0.28

0.81

Note Mean value is mean over the 21 country-specific semi-elasticities; the negative sign means that the tax base (i.e. earnings before interest and taxes) is reduced if the statutory tax rate is increased.

Double tax avoidance   89 A4: Profit-­shifting: transfer pricing evidence

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Clausing (2003) Outcome(s): Substantial evidence for tax-­motivated transfer pricing in US intra-­firm trade prices is present; specifically, the lower a host country’s tax rate, the lower are US intra-­firm export prices and the higher are US intra-­firm import prices Years and countries included: 1997–1999; the United States and 54 partner countries Specification and estimator applied: Log-­log specification; pooled OLS with industry dummies, robustness checks with time fixed effects, country fixed effects and country random effects Dependent variable: Monthly data on US international export and import prices (intra-­firm and non-­intra-firm) at the product level from the Bureau of Labor Statistics are used Tax variable of main interest: log(1 − average tax rate on corporate income), log(1 − statutory tax rate on corporate income) Control variables: log(GDP), log(GDP per capita), log(US dollar exchange rate), various dummy variables (e.g. intra-­firm dummy, dummy for dollar denomination of transactions, dummy for price imputation) Semi-­elasticities Table 11 Dependent variable

Semi-elasticity mean

Semi-elasticity min.

Semi-elasticity max.

Standard deviation

Export prices

  4.08a

  2.54

  5.51

1.35

Import prices

−4.73

−7.91

−1.99

2.38

b

Notes a The positive sign means that lower prices (relative to non-intra-firm prices) are charged on intrafirm exports to affiliated companies located in countries with lower tax rates. b The negative sign means that higher prices (relative to non-intra-firm prices) are charged for intrafirm imports from an affiliated company located in countries with lower tax rates.

Clausing (2006) Outcome(s): A substantial and significant impact of tax avoidance incentives on the pattern of US international intra-­firm trade is found which is consistent with tax-­motivated transfer pricing Years and countries included: 1982–2000; the United States and 51 partner countries Specification and estimator applied: Level-­level specification; pooled OLS with industry dummies, robustness checks with time fixed effects Dependent variable: Yearly intra-­firm trade data for US multinational firms and data from the US Bureau of Economic Analysis are used; for evidence on the “price effect”, the intra-­firm trade balance between the United States and the country hosting US affiliates is used as the endogenous variable

90   M. Leibrecht and T. Rixen Tax variable of main interest: The difference in statutory tax rate on corporate income in the host country of a US affiliate and the statutory tax rate on corporate income in the United States; the difference in average tax rate on corporate income in the host country of US affiliates and the average tax rate on corporate income in the United States Control variables: Share of sales, unaffiliated trade balance, dummy variables for Europe and Japan Semi-­elasticities

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 12 Semi-elasticity mean Semi-elasticity min.

Semi-elasticity max. Standard deviation

4.81

7.32

1.81

1.82

Notes Only tables 3 and A2 in Clausing (2006) on the price effect are explored; the positive sign implies that the intra-firm trade balance, measured as exports to foreign affiliates minus imports from foreign affiliates, increases with an increase in the foreign tax rate on corporate income in the host country.

Overesch (2006) Outcome(s): If an affiliate does not have a loss, carry forwards intra-­firm, then sales decrease with an increase in the host country’s statutory tax rate on corporate income; this is consistent with tax-­motivated transfer pricing in German intra-­firm sales Years and countries included: 1996–2003; Germany and 31 partner countries (the EU-­27 without Bulgaria, Cyprus, Germany and Romania, the United States, Canada, Mexico, New Zealand, Australia, Japan, Norway, Switzerland) Specification and estimator applied: Log-­level specification; 2SLS with parent, industry and time fixed effects Dependent variable: Yearly balance sheet data of German foreign affiliates; variables “accounts receivable from affiliated companies” and “accounts receivable from the parent company” made available by the Midi database of the Deutsche Bundesbank is used, majority-­owned direct participations outside the financial sector and without holding companies are included in the sample Tax variable of main interest: Statutory tax rate on the corporate income in the host country of a German affiliate in the case of “accounts receivable from affiliated companies”, difference in statutory tax rate on corporate income in Germany and that of the host country in the case of “accounts receivable from parent company” Control variables: log(total capital of the affiliate) instrumented by log(GDP of the host country), log(lending rate of the host country), various dummy vari­ ables (e.g. for ownership structure, loss carried forward), log(turnover of the affiliate), log(bilateral distance between Germany and the host country of the affiliate)

Double tax avoidance   91 Semi-­elasticities

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 13 Dependent variable Semi-elasticity mean

Semi-elasticity min.

Semi-elasticity max.

Standard deviation

Trade with parent

  0.51

−3.22a

2.84

1.70

Trade with affiliates −1.00

−3.38

1.80

1.31

b

Notes a Negative values are found in cases when loss carry-forwards are available to the affiliate. b Positive values are found in cases when loss carry-forwards are available to the affiliate.

Bernard et al. (2008) Outcome(s): The paper shows that the wedge between the arm’s-length export price and the intra-­firm export price of US-­based multinationals is negatively related to the taxation of corporate income in the foreign import country; this response indicates that tax-­motivated transfer pricing is present in intra-­firm export prices of US-­based multinational companies Years and countries included: 1993–2000; the United States and 140 partner countries Specification and estimator applied: Log-­level specification; pooled OLS with product fixed effects Dependent variable: The dependent variable is the wedge between the arm’slength export price and the intra-­firm export price of US-­based multinationals; data from the Linked/Longitudinal Firm Trade Transaction Database (LFTTD), which is based on data from the US Census Bureau and the US Customs Bureau, are used Tax variable of main interest: Statutory tax rate on corporate income and average tax rate on corporate income in the host country of US-­based multinationals’ affiliates Control variables: Tariff rate, log(employment of the affiliate), firm export share, log(real exchange rate with the US dollar), log(GDP per capita) Semi-­elasticities Table 14 Semi-elasticity mean Semi-elasticity min.

Semi-elasticity max. Standard deviation

−1.25

−0.39

−4.18

1.18

Notes The negative sign implies that an increase in the host country tax rate on corporate income leads to a decrease in the wedge between arm’s-length export prices and intra-firm export prices.

92   M. Leibrecht and T. Rixen

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Dischinger and Riedel (2008) Outcome(s): The paper shows that the location of holdings and investments in intangible assets is inversely related to the statutory tax rate on corporate income; this relationship can be considered as indirect evidence in favor of tax-­motivated transfer pricing in the intra-­firm trade of intangibles Years and countries included: 1995–2005; the EU-­27 less Cyprus and Malta Specification and estimator applied: Log-­level specification; fixed effects estimator, first difference estimator, Anderson–Hsiao estimator, Arellano–Bond estimator, fixed effects logit and linear probability model estimator Dependent variable: Balance sheet item “intangible assets” plus 1 taken from the Amadeus database; dummy variable with entry 1 if an affiliate owns intangible assets and 0 otherwise Tax variable of main interest: Mean difference between the statutory tax rate on the corporate income of an affiliate and the statutory tax rate on the corporate income of all other affiliates of a common parent company Control variables: log(GDP per capita of host country), log(population of host country), log(unemployment rate in host country), log(total assets of affiliate) Semi-­elasticities Table 15 Semi-elasticity mean Semi-elasticity min.

Semi-elasticity max. Standard deviation

−1.17a

−0.36

−2.04

0.45

Notes a The negative sign implies that an affiliate’s holdings of intangibles decrease, other things being equal, with an increase of the statutory tax rate on corporate income in the host country.

A5: Profit shifting: thin capitalization rules Büttner et al. (2006) Outcome(s): Thin capitalization rules are found to be effective in restricting debt finance Years and countries included: 1996–2004; Germany as home country of FDI and 24 partner countries Specification and estimator applied: Level-­level specification; fixed effects estimator Dependent variable: Debt-­to-asset ratio of German foreign subsidiaries (leverage); data are taken from the Midi database of the Deutsche Bundesbank Tax variable of main interest: Dummy for the presence of a thin capitalization rule, dummy variable for the presence of a thin capitalization rule interacted with the statutory tax rate on corporate income Control variables: log(lending rate for the private sector), log(turnover), statutory tax rate on corporate income, dummy variable indicating whether loss carry-­forward is reported by the affiliate

Double tax avoidance   93 Semi-­elasticities Table 16 Semi-elasticity mean Semi-elasticity min.

Semi-elasticity max. Standard deviation

−8.27a

−8.20

−8.34

0.06

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Notes a The negative sign implies that the debt to equity ratio is reduced when a thin capitalization rule is in force.

Overesch and Wamser (2006) Outcome(s): The German thin capitalization rule induces significantly lower intra-­firm debt levels of inbound investments; hence, tax planning via intra-­ firm finance is effectively limited Years and countries included: 1996–2004; Germany as host country of FDI and 31 investor countries Specification and estimator applied: Level-­level specification; difference-­indifferences approach with affiliate fixed effects and time dummies Dependent variable: Intercompany loans as share of total affiliate capital; data are taken from the Midi database of the Deutsche Bundesbank Tax variable of main interest: Interaction term between “treatment group dummy” and “post-­tax reform dummy” Control variables: log(turnover), dummy for loss carry-­forward of affiliate available or not, difference in statutory tax rate on corporate income of Germany and the home country of FDI, treatment group dummy Semi-­elasticities Table 17 Semi-elasticity mean Semi-elasticity min.

Semi-elasticity max. Standard deviation

−24.72a

−31.00

−31.86

6.49

Notes a The negative sign implies that the debt to equity ratio is reduced when a thin capitalization rule is in force; only results from tables 4 and 5 in Overesch and Wamser (2006) could be used, owing to lacking information on the mean of the endogenous variable.

Weichenrieder and Windischbauer (2008) Outcome(s): The change in German thin capitalization rules in 2001 induces significantly lower intra-­firm debt levels of inbound investments (especially for affiliates which are very highly leveraged prior to 2001) Years and countries included: 2000–2002; Germany as host country of FDI and several investor countries Specification and estimator applied: Level-­level specification; two-­way fixed effects estimator (affiliate and time fixed effects)

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

94   M. Leibrecht and T. Rixen Dependent variable: Intra-­firm debt as share of total affiliate assets one year lagged; data are taken from the Midi database of the Deutsche Bundesbank Tax variable of main interest: Interaction term capturing highly leveraged corporations in 2000 and interaction term capturing highly leveraged branches in 2000 Control variables: log(sales of the affiliate) Semi-­elasticity: −22.99* * The negative sign implies that the debt to equity ratio is reduced when a thin capitalization rule is in force; only results from Table 2, column 1 could be used, owing to lacking information; semi-­elasticity is calculated as ((1.624 − 0.417) × 100)/525 × 100 × (−1) A6: The role of hybrid entities Altshuler and Grubert (2005) Outcome(s): The introduction of hybrid entities into US CFC rules (“check the box”) substantially alters the tax planning opportunities of multinational enterprises, which results in a looser relationship between average tax rates of US foreign affiliates and the host countries’ statutory tax rate on corporate income, a looser relationship between the parent company’s R&D expenditures and the average tax rates of foreign affiliates, and a stronger negative impact of the statutory tax rate on corporate income on the pre-­tax profit declared in a host country Years and countries included: 1996 and 2000; the United States plus 60 partner countries Specification and estimator applied: Level-­level specification; cross-­sectional OLS Dependent variable: Pre-­tax profits as a share of sales; data from the US Treasury tax files for US multinational corporations are used Tax variable of main interest: Statutory tax rate on corporate income Control variables: Age of the affiliate, parent company R&D expenditures as share of sales, parent company advertising expenditures as share of sales Semi-­elasticities Table 18 Year

Semi-elasticity mean

Semi-elasticity min.

Semi-elasticity max.

Standard deviation

1996

−1.48

−1.61

−1.31

0.13

2000

−2.07

−2.3

−1.81

0.2

Notes The negative signs imply that the higher the statutory tax rate on corporate income is, the lower are the reported pre-tax profits of the foreign affiliate. As the ‘check-the-box’ possibility was introduced into US CFC rules in 1997, the difference in the magnitudes of the two mean values is suggestive evidence in favour of an increase in tax-motivated profit shifting via hybrid entities after the 1997 CFC rule change. Specifically, it is likely that highly profitable subsidiaries in high-tax countries became part of a consolidated hybrid entity based in a tax haven.

Double tax avoidance   95

Acknowledgments The part on hybrid instruments in section 4.2.2 was drafted by Martin Six. We are grateful to Julia Braun, who helped in the collection and the analysis of empirical papers.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Notes   1 NB: we emphasize tax avoidance by multinational firms and tax competition for investments by firms.   2 Note here that vis-­à-vis capital, the behavior of the home country as well as that of the host country has to be considered. Thus, there are three actors in the tax competition game, namely firms, the home-­country government and the host-­country government (see also Altshuler and Grubert 2005).   3 Among the seminal papers in this respect is Devereux et al. (2004).   4 In addition to concerns about fairness, there are also important concerns about efficiency. The basic question is whether the residence or the source principle is better suited to achieve tax neutrality. On the pros and cons of capital export neutrality (CEN), capital import neutrality (CIN), national neutrality (NN) and the concept of capital ownership neutrality (CON), see Frisch (1990) and Desai and Hines (2003).   5 While this describes the general pattern, there are many exceptions to this division of the tax base. The precise sharing rules are laid down in the so-­called source rules of the typical bilateral tax treaty and/or the MC (see OECD 2005: Articles 6–22).   6 This is the main reason why developing countries – generally net capital importers – are supporters of the UN MC, which was first developed in the 1970s and emphasizes the source principle (see Rixen 2008).   7 According to Avi-­Yonah (2006), the standard of single taxation has become part of customary international law.   8 Graetz (2003: 402) has observed that the “irony and . . . essential difficulty” is that MNEs exist because of the absence of, or imperfections in, an arm’s-length market, although the arm’s-length standard is used to determine transfer prices for tax purposes.   9 See, for example, Six (2007) for an analysis of the relevant provisions of national tax law in eight Central and Eastern European countries; see Wiedermann-­Ondrej (2007) for an analysis of the relevant provisions of US tax law. 10 For the features of hybrid instruments that can blur the distinction between equity and debt, see Duncan (2000). 11 A minimum of 25 percent interest in capital is required. On the question as to whether hybrid instruments have to be included in the calculation of the minimum holding requirement, see Six (2007). 12 Rixen (2010) provides empirical evidence that these rates depend on the degree of asymmetry of capital flows between treaty partners. The more asymmetric the capital flows, the higher the withholding rates. 13 In the case of interest payments, some DTTs even grant the source state no right to tax at all. See, for example, the DTT between Austria and Switzerland in force since 4 December 1974. 14 The same is of course true for cases where different levels of withholding tax are allowed for dividends and interest. 15 Similar problems clearly arise where the relevant articles permit different maximum levels of withholding tax. 16 For the legal reasoning behind this line of argument, see Six (2009), who also provides further references.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

96   M. Leibrecht and T. Rixen 17 For an analysis of the possible cases of such classification conflicts, see Eberhartinger and Six (2009). For specific examples of such situations, see Six (2009). 18 Tax havens have been subdivided into “production havens”, “headquarter havens” (which offer the incorporation in their territory regardless of where the shareholders reside), “sham havens” (which host financial intermediaries that are little more than an address for investment activity) and “secrecy havens” (which specialize in allowing personal income tax evasion by reinvesting funds that have been provided without the knowledge of authorities at home), depending on the particular features of their financial and tax systems. Most real-­world tax havens provide some mix of these functions. See Eden and Kudrle (2005). 19 In the economics literature, two distinct views on the role of tax havens have emerged. The “pessimistic” or “negative” view stresses that tax havens erode the tax base of high-­tax countries. They also lead to other types of welfare-­reducing costs. The “positive” view makes the point that – if one starts with the assumption that the first-­ best world of curbing tax competition entirely is not achievable, and the available choices are thus only between different kinds of tax competition – the presence of tax havens enables high-­tax countries to actually impose higher taxes on immobile (domestic) firms coupled with lighter taxation of multinational firms. This, in turn, can make tax competition for real capital less fierce and could actually increase the general welfare. See the example given by Dharmapala (2008: 14ff.). 20 There has been dispute among tax lawyers about the extent to which CFC rules violate international tax principles. The dominant view – and government practices reflect this – is that CFC rules are legitimate tools for fighting tax shelters (cf. Rixen 2008: 122–126). 21 There are surprisingly few theoretical tax competition models that consider thin capitalization rules as a device for governments to attract (or retain) capital. A notable example is contained in Haufler and Runkel (2008). The authors present a model with countries competing for MNEs via statutory tax rates on corporate income and applying thin capitalization rules. They find that, in a state of equilibrium, countries opt for inefficient and lax thin capitalization rules. 22 Some thin capitalization rules have been ruled by the European Court of Justice (ECJ) to be in violation of European law. In the Lankhorst-­Hohorst decision of 12 December 2002 (C-­324/00), the ECJ ruled that the German law treating some interest payments as hidden dividend payments, which would not reduce taxable profit in Germany, violated the freedom-­of-establishment clause under the Treaty of Maastricht of the European Union. 23 APAs are mechanisms under which MNEs and tax administrations can bargain over the appropriate method for determining reasonable transfer prices. Thus, the bargaining parties basically commit themselves to certain prices before the transactions actually occur. For a more detailed description of these policies, see Rixen (2008). 24 See section 4.3, pp. 76–78 on the “check-­the-box” regulation in US CFC law. 25 For a discussion of the underlying political rationale, see, for example, Roin (2008: 18–20). Engel (2001) illustrates this in his discussion of CFC rules in the United States. The empirical evidence for governments’ reluctant attempts to curb under-­ taxation suggests that such considerations are prevalent in all countries. Compare, for example, Rixen (2008: 117–151). 26 DTTs usually include a section on information sharing (see Aigner and Tumpel, this volume, Chapter 3). These agreements predominantly aim to threaten individuals who evade taxes illegally (Dharmapala 2008). We do not further elaborate on this aspect here as we are concerned with tax avoidance by MNEs. Moreover, empirical evidence on the effects of advanced information sharing is very scarce; see Dharmapala (ibid.) for two examples. 27 For thin capitalization rules, compare the analysis in Haufler and Runkel (2008: section 4).

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Double tax avoidance   97 28 See Appendices A1 to A6. 29 If the independent variable of main interest is a dummy variable, then the semi-­ elasticity can be represented by 100 × (exp(â) − 1), where â is the regression coefficient. 30 See Table A1 in the appendix for details. 31 A new treaty is a treaty that was signed after the first year in their sample. Old treaties are treaties that were already in force at the beginning of the sample period. For old treaties, Blonigen and Davies (2002) find a positive effect on FDI. Yet this coefficient suffers from severe endogeneity bias, according to the authors. It should therefore not be considered as valid evidence for a DTT’s effect on FDI. 32 Two further papers are Millimet and Kumas (2007) and Coupe et al. (2008). The former finds that the impact of a DTT on FDI is positive for initially low FDI levels and negative for initially higher levels. The latter finds no significant effect (see also Davies et al. 2009; Barthel et al. 2008). 33 As FDI is frequently seen as a motor for economic growth, DTTs might be welfare-­ improving in this respect. However, if DTTs induce welfare-­decreasing (inefficient) tax competition, then they lead to a situation in which all countries are worse off because the capital stock may remain unaltered even though tax revenues decline. 34 Additional surveys, also including early papers on the topic, are Hines (1997) and Devereux (2007). 35 NB: if the coefficient has a positive sign, this implies an increase in net-­of-tax profit of a host-­country affiliate if the home country’s statutory tax rate increases. This is evidence for profit shifting (see Table A3). 36 For the latter point, see also Desai et al. (2006b) and subsection 4.2.3 above on the “positive view” of tax havens. 37 Moreover, a study by Desai et al. (2006c) finds that US MNEs with a focus on R&D are more likely to establish affiliates in low-­tax countries (tax havens). 38 Note that there is an extensive literature showing that the degree of debt financing is closely related to differences in the statutory tax rates on corporate income across countries (see, for example, Weichenrieder 2007 for a survey). Thus, profit shifting via debt financing is a highly relevant phenomenon. 39 See Barba Navaretti and Venables (2004: ch. 7) on the effects of FDI on the host country. 40 DTTs are an example of tax coordination by explicit agreement (see Wildasin 2002). 41 Real tax competition is defined here as tax competition for real capital without profit-­ shifting opportunities. 42 See European Commission (2001) on the CCCTB. For good overviews of many issues concerning the choice between unitary taxation and separate accounting, see Sørensen (2004) and McLure and Weiner (2000). 43 See Lang and Zagler (this volume, Chapter 8). 44 See De Mooij and Devereux (2009) for a simulation exercise in this respect.

5 Intra-­firm dividend policies Evidence and explanation

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Christian Bellak and Nadine Wiedermann-­Ondrej1

Corporate dividend policy remains a topic on which researchers have failed to arrive at even a local sense of closure. (Marsh and Merton 1987)

5.1  Introduction Although considerable progress has been made in empirical research over the past 20 years, unexplored territory remains. One such area concerns intra-­firm dividend policies in a cross-­border setting of firms. In this study, the dividend policy of firms is defined as the decision made about the level of dividends (i.e. the payout ratio) and about the adjustment of dividends (i.e. dividend smoothing), with a focus on the latter. What is dividend smoothing? “Dividend smoothing implies a deliberate effort on the part of managers to adjust dividend payments in response to variations in the earnings stream” (Bhabra et al. 2002: 166). We are interested in particular in explaining intra-­firm dividend policies that take the form of intra-­firm dividend smoothing. In spite of the fact that these policies constitute an important element in multinational enterprises’ (MNEs) financial policies, empirical evidence is rather scarce (see section 5.2). The setting for the dividend policy decision chosen in this chapter is a German parent company – that is, a German MNE that owns subsidiaries abroad. An MNE may finance its foreign operations through various means (equity, debt, etc.). If the foreign subsidiary is profitable, then the parent company may wish to shift some of the profits home to finance new investment. Obviously, the parent company may also have other options. For example, it may reinvest the profits at the subsidiary where it emerged or at other subsidiaries. However, these options are not the focus of this chapter. Furthermore, there are also many options for shifting profits back home from a foreign subsidiary. They include share repurchases, dividend payments, profit shifting via transfer pricing (for example, see Leibrecht and Rixen, this volume, Chapter 4, especially in subsection 4.3.2.3, “Tax-­motivated profit shifting”), intra-­firm debt flows and equity-­related transactions. Some of these transactions

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Intra-firm dividend policies   99 are exclusively intra-­firm (the so-­called intra-­firm sphere) and some may be intra-­firm as well as between the firm and its personal shareholders (i.e. the socalled personal shareholder sphere). This chapter provides a survey of empirical evidence on dividend policies of MNEs and discusses the determinants of dividend smoothing in the intra-­firm case – that is, parent versus foreign subsidiary, with respect to the case of a German parent company. As a theoretical model, Lintner’s (1956) model has been used, as it has been shown to be relevant to the intra-­firm context by Bellak et al. (2009), inter alia. Our survey of empirical studies has revealed a large gap between dividend smoothing in the personal shareholder and intra-­firm spheres. With regard to shareholders, firms adjust dividends slowly, while intra-­firm dividends are adjusted much more quickly. However, the aim of this study is not to explain this difference between personal and intra-­firm sphere, but rather to analyze incentives and disincentives for the intra-­firm dividend smoothing behavior of firms. Thus, we are only concerned with firms actually paying dividends. Our findings point to the fact that signaling and agency cost arguments are of minor importance for the dividend smoothing decision in the intra-­firm context, but issues related to tax and business-­related considerations play some role in explaining smoothing. The chapter is organized as follows. We start from a survey of empirical studies on dividend policies in the intra-­firm and the personal shareholder spheres. Then we proceed to explain the smoothing behavior of firms by tackling its key determinants in the intra-­firm context. Lastly, a short section summarizes the results.

5.2  A review of empirical results on dividend smoothing This section presents a survey of 68 empirical results in total. Among them, 57 results were originally derived in the personal shareholder context2 and 11 results in the intra-­firm context. These results were taken from 32 studies, starting with Lintner’s study in 1956, that dealt with dividend payments to personal or institutional shareholders, and four studies that dealt with intra-­firm dividend payments. Empirical results on dividend smoothing were almost exclusively derived on the basis of the Lintner model (Lintner 1956) (for an exposition of the model and a survey, see, for example, Bellak et al. 2009). The results of the literature survey are summarized in Table 5.1. Grouping of the studies based on personal shareholder sphere and intra-­firm sphere, as well as based on various characteristics, revealed the following stylized facts: • •

The range of all four parameters shown in the table is enormous. Marked differences between OLS- (ordinary least squares) and non-­OLSbased studies (mostly panel econometric studies) can be observed: because of the left-­censored nature of the dividend data (that is, dividends can only

100   C. Bellak and N. Wiedermann-Ondrej

•

be zero or positive), the OLS technique is not sufficient to estimate the degree of smoothing. As regular regression techniques such as OLS do not account for data censoring, they yield inconsistent estimators. There are hardly any studies in the personal shareholder field that use Tobit (Tobin’s probit) models in their empirical specification, an exception being the study by Hayunga and Stephens (2008). Therefore, an obvious explanation for these gaps between the shareholder sphere and the intra-­firm sphere is the methodology applied in empirical studies.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 5.1  Summary of the results of earlier studies of dividend smoothing Indicator (number of studies)

Speed of Averaged Mean Median adjustment long-run adjustment adjustment payout ratio lag lag

Personal shareholder sphere Average of 40 OLS results Average of 17 non-OLS results Average of all results (57)

  0.43   0.61   0.49

  0.41   0.38   0.41

  2.80   1.14   2.36

  2.24   1.06   1.93

Intra-firm sphere Average of 5 OLS results Average of 6 non-OLS results Average of all results (11)

  0.70   0.70   0.68

  0.38   0.47   0.43

  0.46   0.51   0.53

  0.60   0.63   0.64

  0.48

  0.55

  2.61

  2.11

  0.46

  0.30

  2.03

  1.71

  0.57

  0.44

  0.75

  0.82

  0.63

  0.25

  1.18

  1.09

  0.64

  0.65

  2.50

  2.01

Personal shareholder sphere/intra-firm sphere × 100 All results 72.17 93.83

446.22

299.15

All studies Average of total studies (68)

  2.05

  1.71

Studies by other characteristics Average of credit countries (the US and the UK) (34) Average of exemption country (Germany) (13) Result of studies on multiple countries (EU-15) (1) Average of other countries (i.e. France, Portugal, the Netherlands, Oman, Tunisia, Malaysia, Turkey, Canada) (20) Results of the studies including any tax variable (14, with 9 in the intra-firm case)

  0.52

  0.41

Notes 1 Eight studies, including the estimates of the Lintner model, could not be used, owing to the lack of information on coefficients. 2 Only estimates of the Lintner model were included. When other control variables were used, the model that used the least additional controls was included. 3 The mean adjustment lag can be defined as (1 − c/c), where c = speed of adjustment coefficient. A higher value of c indicates a speedier adjustment or less smoothing. 4 The median adjustment lag can be defined as (ln 0.5/ln (1 − c)), where c = speed of adjustment coefficient (for example, see Pindyck and Rubinfeld 1998).

Intra-firm dividend policies   101 •

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

•

•

Marked differences in the results between studies dealing with the personal shareholder and the intra-­firm spheres could be observed. In particular, the intra-­firm dividends seem to be much less smoothed than dividends paid to personal shareholders. Specifically, while the payout ratio is almost the same as in the personal shareholder sphere, the median adjustment lag is only one-­third on average. Although tax considerations have been raised in many of the reviewed papers on a conceptual basis, only 14 results have been derived from studies that included a tax variable to reflect either the personal shareholder tax burden or the company tax burden (9 results). The speed of adjustment is quite similar between credit and exemption countries, but the payout ratio differs, and consequently the adjustment lags.

As the focus of this study is on intra-­firm dividend payments, the three3 empirical studies available are now described in greater detail as follows. On the basis of OLS regression, Desai et al. (2001) found a speed of adjustment parameter of 0.73 (in the range of 0.53–0.77) (Desai et al. 2006a). Based on Tobit specifications, they (Desai et al. 2001, 2006a) found a speed of adjustment parameters of 0.67/0.77. These results suggest that median adjustment lags are between 0.47 and 0.91 years. Bellak et al. (2009) examined the dividends repatriated from completely foreign-­owned manufacturing subsidiaries to their German parent company over the period 1999–2005. In line with the results obtained by Desai et al. (2001, 2006a), they showed that the target payout ratio and the degree of dividend smoothing are relatively low once the time-­invariant, unobserved heterogeneity is controlled. Depending on the specification, the smoothing parameter (speed of adjustment) was found to be about 0.88, implying a median adjustment lag of about 0.31 years. Thus, recent evidence on the dividend-­smoothing behavior of firms demonstrated marked differences between the dividend policies in the intra-­firm context and those in the personal shareholder sphere. In particular, dividend smoothing seemed to be used much less in the intra-­firm context. The following subsections provide possible explanations of the intra-­firm dividend-­smoothing behavior of German MNEs.

5.3  Dividend smoothing in the intra-­firm context: why parent companies do or do not smooth dividends of subsidiaries As was outlined earlier, explanations for the relatively low smoothing of dividends in the intra-­firm context are the main focus of this chapter. Though attractive, it would be too simple to apply the opposite explanations to those put forward in the personal shareholder context to the intra-­firm sphere. Rather, in the absence of evidence we can inquire whether some of the standard findings on dividend policies in the personal shareholder context can be applied usefully in

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

102   C. Bellak and N. Wiedermann-Ondrej the intra-­firm context to explain the low degree of smoothing (subsections 5.3.1 and 5.3.2) and to look for other reasons specific to the German institutional and economic environment (subsections 5.3.3 and 5.3.4). Some key determinants of the smoothing behavior of firms that have been outlined in the literature are: (i) the use of alternative forms of profit repatriation; (ii) provisions in the legal environment of the home country – in particular, tax issues (Rozycki 1997) – that may account for a certain dividend policy within the firm; and (iii) business-­related determinants (such as agency and signaling issues, financing issues). As the first item has been relatively well researched4 (albeit in the personal shareholder context; see, for example, Altshuler and Grubert 2002), the focus of this study is on the latter two. 5.3.1  The signaling approach (information asymmetry) Goddard et al. (2006) describe the signaling approach as an alternative approach to smoothing, whereby the dividend choices are linked to managers’ future expected earnings (rather than past and current earnings, as in the Lintner model), such that the dividend changes should indicate changes in future earnings. Therefore, dividend change announcements should depend on the expectations about the growth of future earnings. Unlike the Lintner model, this literature has revealed motives for changing the smoothing of dividends. When firms announce changes in the dividend policy, they convey information to markets. If asymmetric information is given, then better-­informed insiders use the dividend policy as a costly signal to convey their firm’s future prospect to less-­informed outsiders (see Bhattacharya 1979 and subsequent literature). Thus, dividends signal managers’ superior inside information concerning the firm’s future earnings, such that higher dividends may signal higher future earnings and a higher market value. By increasing the dividends, firms create a cost to themselves, as they commit to paying these dividends in the long term.5 This positive signal should therefore lead investors to reevaluate the cash flows and the firms’ values, and increase the stock price. When a firm decreases dividends, markets consider it as an indication that this firm is in substantial and long-­term financial trouble (a negative signal). Consequently, such actions lead to a fall in stock prices. However, how does the signaling argument for the individual shareholder sphere perform in practice? Aivazian et al. related the dividend decision to the debt decision of the firm, and maintained that “the use of bank debt reduces the need for signaling and mitigates agency costs, two of the main functions typically served by dividends that lead to a smoothed dividend policy” (2006: 442). Private bank debt is argued to reduce the value of the signaling and agency reduction roles typically fulfilled by dividend payments. Similarly, Vieira and Raposo (2007) argued that it is “probably because the reliance of civil law countries, like Portugal and France, on bank debt and their closely held nature reduce the information problems in the context of outside capital.” Furthermore,

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Intra-firm dividend policies   103 Goergen et al. (2005) concluded that in Germany, because of concentrated ownership, it may not be necessary for firms to use dividends as a signal. Besides these issues, Skinner (2007: 37f.) noted the general decline in the propensity to pay dividends over time, and observed that – in the United States – dividend payments are concentrated with the largest firms. This evidence points to the importance of information problems, but, at the same time, it is hard to reconcile the fact of larger, well-­known firms paying dividends, which have “reasonably good substitutes for dividend signaling” (Skinner 2007). This leads us to the conclusion that in the intra-­firm context the signaling argument is of limited relevance, as substitutes for signaling should be manifold, although, as Frankfurter and Wood (2002) pointed out, they may not be perfect substitutes. Thus, the information content of dividends is limited and dividend smoothing is less important. Moreover, the signaling argument is applicable in the intra-­firm context only if one assumes asymmetric information between the parent and the subsidiary, which – given the numerous instruments of control – is unlikely. This works similarly in private banks that “have a comparative information advantage over arms length bond markets” (Aivazian et al. 2006: 441). 5.3.2  The agency cost view (agency conflicts) The agency cost view of dividends was developed by Jensen and Meckling (1976), and has since been refined and empirically tested. Hayunga and Stephens (2008: 2) argued the following with respect to smoothing: “More recently, researchers have focused on information asymmetry and agency cost explanations of dividend policy as both are potential explanations of the market reactions observed around changes in dividends.” With regard to smoothing, the agency cost view maintains that dividends have information content, and thus higher dividends may mitigate agency cost problems. However, if we turn again to some empirical evidence, how important are agency costs as determinants of the smoothing behavior? Dewenter and Warther (1998) compared the dividend policies of Japanese and US firms. In Japan, in industrial groups, as opposed to independent firms, information asymmetries between managers and shareholders may be of little relevance. A parent–subsidiary relationship principally resembles the relationship between dependent firms in the group in Japan, where short-­term information transmission is not necessary. In this environment, dividend changes should be driven by changes in excess cash. Miller and Rock (1985) discuss dividend policy in the absence of information asymmetries, and they identify the optimal dividend policy as full payout of excess cash because that policy maintains the firm’s investment level at its Fisherian optimum. Therefore, in a full information setting dividend policy should be more responsive to earnings. (Dewenter and Warther 1998: 881)

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

104   C. Bellak and N. Wiedermann-Ondrej In other words, smoothing should only be given to a certain extent. In the context of an MNE, this argument suggests that the parent company, because of its information about the financing needs (see subsection 5.3.5) of other subsidiaries, uses dividends to distribute cash in the multinational group, rather than using them as a control device. Aivazian et al. (2006) argued that “Lintner type dividend smoothing is normally viewed as a solution to both agency and signaling problems.” However, as La Porta et al. noted, financial solutions to agency problems partly depend on the legal environment and characteristics of the investors. For example, “dividend smoothing may be optimal for firms with dispersed share ownership, but not for ‘internal’ shareholders who can observe earnings and management directly” (1999: 440). Khan (2006) used UK data on dividend payments and argued that his results are consistent with the agency models “in which dividends substitute for poor monitoring by a firm’s shareholders but can also be explained by the presence of powerful principals who are able to impose their preferred payout policy upon firms”. This latter notion is also of interest in the intra-­firm context, as many parent firms are seen as a “powerful principal”, because often the ownership share is high. The argument of concentrated ownership, which reduces the agency costs and thus the need to use dividends as a control device, was also put forward by Da Silva et al. (2004). Using data obtained from Germany, where concentrated bank ownership is common, they also examined the German companies held by “large foreign companies”, which resemble an intra-­firm dividend relationship. The payout ratio was 28 percent on average, and thus in line with that for other types of shareholders, except banks, where the payout was less. However, in their study on the smoothing behavior they excluded firms “that are subsidiaries of foreign companies” (ibid.: 135, table 8.4). 5.3.3  The free cash flow hypothesis Combining market information asymmetries with the agency approach (Frankfurter and Wood 2002) suggests that dividends mitigate investments of managers in unprofitable/wasteful investments/“pet investments” (e.g. Chetty and Saez 2007). By reducing the amount of free cash flow, dividends force managers to submit to the discipline of the financial markets. Unlike the above-­mentioned argument, this argument contributes somewhat to explaining why there is little smoothing as the parent company avoids the conflict arising from inefficient use of surplus funds between the principal (parent) and the agent (subsidiary) by siphoning off any excess cash. In summary, the above-­mentioned arguments suggest that the signaling of the subsidiary and the agency-­cost considerations of the parent firms have limited relevance in the intra-­firm case, and thus the almost complete absence of information asymmetries is in line with the low degree of dividend smoothing. The fact that there is some smoothing in the intra-­firm context, as shown by the survey of empirical studies, might thus be attributable to some low degree of

Intra-firm dividend policies   105 information asymmetries (incomplete control devices, imperfect substitutes, etc.), as some discretion seems to remain for subsidiaries’ managers (even with sophisticated devices other than dividends) and other factors.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

5.3.4  Tax implications Among the tax implications for corporate shareholders, issues such as avoidance of double taxation of parent and subsidiary (exemption method) firms as well as controlled foreign companies (dividends taxed even if no actual distribution has taken place) arise. Before we provide a detailed explanation, it should be mentioned that different views on the relevance of dividend taxes to financial decisions of firms have already been developed earlier, usually termed the “new view” and the “old view”. In a nutshell, the “new view” does not forecast any behavioral effects or efficiency costs of the dividend taxes, at least when imposed on firms currently paying the dividends. Though taxes on distributed profits are neutral, they still reduce the share values. On the other hand, the (business and personal) tax implications of the “old view” are distortions of the financing as well as the real investment decisions of the corporations. As no consensus has been reached so far between these two views of dividend taxation, rather than evaluating these two views this subsection lays out the various rules and regulations as well as the principal considerations that apply in cases when firms pay out dividends. Nevertheless, one has to keep in mind that the relevance of some of these arguments may differ greatly, depending on which of the two views of dividend taxation “better” describes the actual dividend policies of firms in the light of the various regulations. We have highlighted a few of these key differences here. The first aspect concerns the importance of new share issues (and alternative means) relative to retained earnings as a marginal source of finance. In other words, the absolute amount of subsidiary profits for distribution would be higher according to the “old” view, where the subsidiary would rely on new share issues, and thus the effects of dividend taxation on investment would be more severe even within the business sphere (intra-­firm, subsidiary–parent company). A related issue concerns the life cycle of a subsidiary (i.e. whether it is a new or a mature subsidiary), because the new view convincingly argues that the sources of equity finance – and thus the role of debt – change over time. Thus, for MNEs with several subsidiaries, different dividend policies may coexist at the same time, depending on the age mix of the subsidiaries. The third point to mention is the value of signaling information (as described in greater detail in subsection 5.3.1) and the reduction in the agency costs (cf. subsection 5.3.2), which are the two main reasons put forward to explain why firms pay dividends. From the viewpoint of the shareholders, the “old” view maintains that shareholders forgo higher net returns on shares for higher payout ratios. From the viewpoint of the subsidiary, dividends serve as a signal of profitable investment opportunities to the stock market. Both the arguments have to be qualified in the intra-­firm context, as the marginal shareholder is the parent

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

106   C. Bellak and N. Wiedermann-Ondrej firm and the subsidiary acts on an “internal stock market” vis-­à-vis its parent. As has been pointed out by several writers, in the personal shareholder sphere it remains unexplained why the firm does not reinvest profits to create earnings in the future, and rather forgo profitable investment opportunities (e.g. Sørensen 1995, vol. 2: 281; Zodrow 1991: 503, col. 1). This suggests that in the intra-­firm context, the parent firm will decide whether the subsidiary firm should invest in the profitable project or not. The “new” view argues that dividends “should fluctuate considerably in the face of changing investment opportunities” (Zodrow 1991: 503), and thus would describe such behavior accurately. Lastly, the low degree of dividend smoothing in the intra-­firm context presented in the above-­mentioned literature survey seems to reflect the new view more, as the dividends are considered as “residuals”. 5.3.4.1  Companies subject to taxation Corporations resident in Germany are subject to tax on their worldwide income. A corporation is considered to be a resident in Germany if it maintains either its registered office or its principal place of management in Germany. The principal place of management is where key decisions are regularly made and the place from which day-­to-day business operations are managed (Kaminski and Strunk 2006: 159; Brähler 2009: 9f.). 5.3.4.2  Corporate taxes Since 2001, profits have been subject to a corporate income tax rate of 25 percent at the level of the corporation, plus a solidarity surcharge of 5.5 percent (25 percent plus 5.5 percent of 25 percent = 26.375 percent). In 2008, a major reform of the corporate income taxation came into force, leading to a significant reduction in the statutory tax rate on corporate income. As of January 2008, the corporate income tax (CIT) rate stands at 15 percent, increased to 15.83 percent owing to the 5.5 percent solidarity surcharge. In addition, the local tax on trade and industry (average rate across Germany = 14.0 percent) leads to a combined CIT rate of 29.83 percent (European Commission 2008a: 146f.; Kessler and Eicke 2007: 1135; Kummer and Blome 2008: 470). Furthermore, in addition to the CIT, a corporation must pay trade tax. The trade tax (Gewerbeertragsteuer) is levied by local authorities. Every local authority is allowed to determine its own trade tax rate for its area. The trade tax burden on income usually ranges between 9 percent and 20 percent of the business income. The 2008 Business Tax Reform Act has repealed the deductibility of the trade tax, and therefore broadened the corporate tax base (Kessler and Eicke 2007: 1135; Ortmann-­Babel 2007: 266). 5.3.4.3  Taxation of dividends and capital gains To mitigate the impact of national and international double taxation, Germany generally exempts dividends and capital gains from its tax (exemption method). Dividends received by a German resident company from another German resi-

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Intra-firm dividend policies   107 dent or foreign company are generally exempt for corporation tax purposes, regardless of minimum shareholding, minimum holding period, activity requirements or double tax treaties. For trade tax purposes, the exemption is granted only when there is a minimum shareholding of 10 percent. Although dividends from domestic or foreign companies are exempted, according to § 8b Abs 1 KStG, 5 percent of such dividends is treated as a non-­deductible expense, triggering a liability to corporation tax. Effectively, this means that 95 percent of the dividends are tax-­free, but all related costs actually incurred are fully deductible (Schmidt et al. 2005: 170; Brähler 2009: 248; Prokisch 2005: 280; Einem and Tränkle 2004: 40; Maiterth 2002b: 171f.). Capital gains from the sale of shares in domestic or foreign subsidiaries held by a German corporation are also exempted from corporate income tax. As in the case of dividends, 5 percent of the capital gain is treated as a non-­deductible business expense. Furthermore, costs directly attributable to the sale of shares and capital losses on shares are not deductible (Brähler 2009: 254f.; Schmidt et al. 2005: 171; Maisto 2008: 60;6 Prokisch 2005: 280; Maiterth 2002a: 567). Gains on shares and dividends realized held for trading purposes by credit institutions, financial service institutions and other finance companies, as well as for certain insurance companies, are fully taxable as ordinary profits (Maisto 2008: 60; Prokisch 2005: 280f.). However, this rule does only apply to countries where the EU Parent–Subsidiary Directive is not applicable (§ 8b Abs 7 KStG). As a result, the distribution of profits from countries outside the European Union is taxable, at least at the level of the German corporate tax. German parent companies can therefore try to evade immediate taxation by avoiding dividend distributions and retaining those profits in the foreign country. A distribution would only be tax-­free in years when the German parent company made a loss. This might explain why dividend distributions are not smoothed, but adjusted to the parent’s needs. However, the distribution of dividends or the sale of shares generates a tax liability that can be avoided by retaining the profits in the host country of the foreign subsidiary. Particularly when there are investment opportunities for the subsidiary, a distribution of profits and a transfer of equity funds from the parent company to the subsidiary, at a time when the subsidiary is in need of those funds, would be tax-­penalized. This explains the decision of managers to smooth dividends in order to quickly react to investment opportunities that occur. Desai et al. found that subsidiaries in high-­tax-rate countries are most likely to pay dividends while receiving equity transfers from their parents (2002: 5). This strategy is consistent with the US tax regime that mitigates double taxation with the credit method. Under this method, dividends are taxed at least at the US level. If corporate taxes in the foreign country are lower, additional corporate taxes have to be paid. On the other hand, if they are higher, then no additional taxes are due. As a result, for the repatriation of profits from highly taxed countries, no additional tax is levied. The German system does not differentiate between dividends from low-­tax or high-­tax countries, and 5 percent of the dividends or capital gains are always taxed.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

108   C. Bellak and N. Wiedermann-Ondrej Desai et al. (2001) analyzed the effect of repatriation taxes on dividend payments of US subsidiaries abroad to their US parents, based on the Lintner model. They found that a 1 percentage point change in the dividend tax rate also changed the volume of the dividends repatriated by about 1 percent. Moreover, their results imply that the US adoption of an exemption system would increase the aggregate dividend payouts by 12.8 percent, with the effects, of course, varying sharply between the subsidiaries in different tax situations. Thus, they (Desai et al. 2001: 849) concluded that repatriation taxes “reduce the volume and efficiency of financial flows between subsidiaries and their American parents”, which was subsequently confirmed (Desai et al. 2006a). Results from the augmented Lintner model derived by Bellak et al. (2009) for German parent companies and their subsidiaries abroad imply that high dividend taxes do indeed have a statistically significant negative impact on the expected value of dividends repatriated: a 1 percentage point decrease in the dividend tax rate would increase dividends repatriated by about 3.5 percent. These studies imply that dividend taxes do matter for intra-­firm dividend payments. Moreover, this result applies to countries that use the credit system as well as those that use the exemption system. 5.3.4.4  Foreign withholding tax Generally, the distributing foreign corporation has to deduct withholding tax at the source. The German corporate shareholder cannot offset the amount withheld, because the dividend payments are tax-­exempt. As a result, dividends are taxed at the level of the foreign withholding tax (Andres et al. 2009; Brähler 2009: 248). For corporations, a withholding tax relief may be provided under the tax treaties. According to most of the German tax treaties, dividends paid to a German corporation are subject to a reduced withholding tax rate of up to 15 percent in the other country. If a German corporation holds more than 25 percent (some tax treaties provide a 10 percent threshold) of the foreign corporation, then the withholding tax can be further reduced to 5 percent (Brähler 2009: 261f.; Kaminski and Strunk 2006: 107; Maiterth 2002a: 567). In general, the withholding tax would be credited against the German tax liability. Because of the tax exemption for dividend income at the corporation level, the withholding tax is, in most cases, not creditable (Kaminski and Strunk 2006: 294ff.). If the foreign corporation is a resident in an EU Member State, then the principles of the EU Parent–Subsidiary Directive can be applied. According to this directive, dividends paid to its EU parent are not subject to withholding tax, provided the EU parent company has held a minimum of 10 percent of the shares in the subsidiary for at least one year immediately prior to the distribution (90/435/ EWG, 20 August 1990; Brähler 2009: 261; Maiterth 2002a: 567). As a result, those dividends can be distributed without any withholding taxation. The level of withholding tax in the foreign country can explain why corporations have an incentive not to distribute all of their profits. If the withholding

Intra-firm dividend policies   109 taxes are high, it is more favorable for a German corporation to retain the profits at the foreign subsidiary than to distribute profits and pay taxes.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

5.3.4.5  CFC rules Income of foreign subsidiaries is in general not subject to German tax, unless it is paid as a dividend to the German corporation. There is, however, anti-­ avoidance legislation (CFC legislation) to prevent the deferral of income. The CFC rules generally apply if companies are under the control of German residents. According to § 7 AStG, this is the case if more than 50 percent of the shares are owned directly or indirectly by German residents (Brähler 2009: 492; Kaminski and Strunk 2006: 238ff.). The CFC rules provide that a German-­resident company will be subject to German tax on certain passive income that was derived by a foreign company domiciled in a low-­tax jurisdiction but was not distributed to its German shareholders. Low-­level taxation is assumed if the passive income is subject to income taxes at a rate of less than 25 percent. In § 8 Abs 1 AStG, there is a list of activities considered to be active. All other items of income are considered to be passive. Passive income of a foreign company with a capital investment character is taxable in Germany, irrespective of profit distribution (Kaminski and Strunk 2006: 240; Brähler 2009: 497ff.). When a foreign company has decided to constitute a CFC according to German law, the income is taxed at the level of the shareholder. Normally, a distribution of dividends is subject to a tax exemption of 95 percent for corporations. However, this exemption does not apply to deemed dividends from CFCs. The reason for this is that an exemption will be granted only if dividends have already been subject to tax at the level of the corporation (Brähler 2009: 507; Kaminski and Strunk 2006: 253). According to § 8 Abs 2 AStG, the CFC rules do not apply if the taxpayer can prove that the controlled company is a resident in another EU or EEA Member State and is carrying on a genuine economic activity. Furthermore, it is necessary that Germany and the respective EU or EEA state in which the controlled company is a resident conclude an agreement that ensures the exchange of information necessary for taxation (Brähler 2009: 499). In summary, if the CFC requirements are met, then dividend smoothing with respect to taxes can be accomplished only when the parent company is a resident of an EU or EEA Member State. In all other cases, the German tax principles will deem a distribution of all taxable profits. 5.3.5  Business implications 5.3.5.1  Introduction Every year, corporations have the choice of distributing a part or all of the profits to the shareholders. However, a corporation does not have the complete authority

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

110   C. Bellak and N. Wiedermann-Ondrej to decide whether to distribute its profits or not. Because of legal provisions or its articles of association, a corporation can have an obligation to retain part of its profits. This is especially true for banks or insurance companies, as these corporations are required to fulfill certain capital resource requirements to reduce the risk of insolvency. In Germany, it is the management board that must prepare the annual financial statements of the company and immediately submit them to the supervisory board, together with a suitable proposal for the appropriation of profits. Subsequently, the supervisory board must review the financial statement and the appropriation of profits, and declare whether it has approved or made objections to the prepared financial statements (§ 171 Abs 3 AktG). It is the exclusive competence of the shareholders’ meeting to resolve the appropriation of profits (§ 174 Abs 1 AktG). However, the shareholders are bound with their decision on the approved financial statement. Still, the distribution of profits lies in the hands of shareholders. If these shareholders are powerful and united – as in the case of 100 percent of shareholding parent companies – then they can strongly influence the amount distributed. 5.3.5.2  Cash flow needs C ash F low N eeds of the F oreign S ubsidiar y

The distribution of dividends is always connected to an outflow of cash. It is possible for the subsidiary to have adequate earnings to declare profits but not to have sufficient cash to pay the distribution. As a result, the cash position of a corporation is an important factor with regard to dividend payments. As was described earlier, free cash flows can therefore be used to make a dividend distribution or, alternatively, the question of whether there will be long-­term investment needs can be considered. In the latter case, it might be better to retain those earnings than to distribute dividends, as the issuance of the new stock or bonds is associated with substantial expenses. Because of different investment needs, dividend distributions depend on the life cycle of a company. Start-­up companies with a strong need for investments tend to retain their earnings, whereas well-­established companies with larger cash flows and fewer projects tend to pay out more of their earnings as dividends (Damodaran 2005: 450ff.). C ash F low N eeds of the P arent Company

Another important factor for smoothing dividends is the cash flow needs of the parent company. This is especially true for holding companies that do not generate cash flows. To meet ongoing expenses, the parent company will ask its subsidiaries to contribute cash flows constantly. This demand for constant payments is also necessary if the parent company is interested in paying constant dividends. Any distributions above the needs of the parent company would

Intra-firm dividend policies   111 transfer cash to Germany and would – in the case of profit generation – lead to higher taxation. On the other hand, if the parent company has interesting investment opportunities domestically or in another country, it will ask its subsidiary to provide liquidity just in time and to the amount needed. 5.3.5.3  Perception of annual accounts

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

C onsolidated F inancial S tatements

The consolidated financial statement serves as a strong indication of the payout ratio of a subsidiary. Investors in a parent company tend to take the consolidated financial statement as a performance measure of their investment. Auer (2007) showed that the earnings of the consolidated financial statement have a great impact on dividend policy. As the consolidated financial statement indicates the overall profit of all the subsidiaries of the MNE, the profit of a subsidiary expected as being in line with the overall profit can only be an average. Outliers tend to distribute in accordance with the overall profits and thus smooth their dividend distributions. B alance S heet R atios

With the distribution of profits, it is possible to influence the ratios of the individual financial statement of the parent company and that of the subsidiary. For the subsidiary, a distribution leads to an outflow of equity; and for the parent company, dividends mean an increase in profits, and therefore an increase in equity. As a result, distributions can be used to influence annual statements and the corresponding ratios. Corporations can use dividends to change the debt ratios of individual financial statements of their foreign subsidiaries and that of the parent company. Increasing the dividends results in a higher equity ratio for the parent company and decreasing them reduces the same equity ratio. With the distribution of profits, it is also possible to change all the ratios connected with the profitability of the parent company’s individual financial statement. Although there are ratios that take profits from distributions into account (such as return on capital employed, ROCE), ratios such as earnings per share, return on equity and return on investment play an important role in the perception of a corporation standing. This can be of importance if the company is in need of further capital (either debt or equity) and has to prove solid creditworthiness (Desai et al. 2002: 16). This is even more important for banks and insurance companies, as these entities have to fulfill certain solvency capital requirements calculated not only at a group level but also at individual financial statement level. As the subsidiary contributes to those ratios of the parent company, along with the dividend distribution, it is in the interest of the parent company to be able to anticipate those contributions and receive them regularly and constantly.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

112   C. Bellak and N. Wiedermann-Ondrej Ratios can also be used as a basis for discretionary benefits of the management. Depending on the ratios used to calculate an additional part of the salary, they may greatly influence the decisions of the management. If dividends are taken into account, it is in the interest of management to receive those distributions rather than to reinvest abroad, even though there might be no interesting reinvestment opportunity in the home country. However, the effects of dividend distributions on the individual financial statement level are all eliminated if the ratios are calculated at the group level. In this case, all the involved companies are treated as one, and transactions between those companies are not taken into account. As a result, there remains no incentive to distribute profit above the needs of the parent company. 5.3.5.4  Shareholder structure C ontrolling Shareholders

In the case of a foreign subsidiary that is controlled by a controlling shareholder, the management of the foreign subsidiary usually knows the expectations of the parent company. If the body of shareholders is small and homogeneous, then the management of the issuing company can fulfill the needs of the shareholders. As dividend repatriations represent significant financial flows, repatriation policies may reflect the financing concerns of the parent firms that draw on subsidiary cash flows to finance domestic expenses. In particular, for firms that are already maintaining large amounts of external debt, dividend repatriations from foreign subsidiaries may offer an attractive source of financing to pay dividends to common shareholders and cover domestic investment expenditures (Desai et al. 2002: 5). I ncomplete O wnership

Incomplete ownership of foreign subsidiaries reduces the ability of the parent companies to monitor and control foreign managers. Partial ownership of foreign subsidiaries by local firms in the host countries increases the risk that a manager will pursue related-­party transactions that are not in the interest of the multinational parents. In such a setting, a rigid repatriation policy may help to control foreign management by limiting its financial discretion (Desai et al. 2002: 17) (see the agency argument earlier).

5.4  Summary This chapter has examined the dividend policies of MNEs. In particular, we have studied the relatively short dividend adjustment period – that is, the relatively minor smoothing after changes in earnings in the intra-­firm context. Our findings are as follows: •

The literature survey revealed marked differences between OLS- and non-­ OLS-based studies, as well as between studies dealing with the personal

Intra-firm dividend policies   113

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

•

shareholder and the intra-­firm spheres. In particular, intra-­firm dividends seem to be much less smoothed than dividends paid to personal shareholders. The signaling of the subsidiary and the agency cost considerations of the parent have limited relevance to the intra-­firm case, and therefore the almost total absence of information asymmetries is in line with the low degree of dividend smoothing: a Signaling issues seem to be of minor importance in the intra-­firm context, as the information content of dividends is limited for the parent company. b Agency arguments also suggest little smoothing, as less information asymmetry leads to fewer agency conflicts. This implies that in the intra-­ firm context, parents are less reluctant to adjust dividends.

Therefore, additional explanations were considered: •

•

Analysis of tax accounting considerations showed that, owing to the exemption method for dividends of foreign subsidiaries, the tax cost of those distributions is generally low and therefore plays a subordinate role. Only when the foreign withholding tax is high – which can be the case in non-­EU Member States – can taxes be an important factor for smoothing. Analysis of business-­related motives for the low importance of smoothing showed that in the case of intra-­firm distributions, the profit of the consolidated financial statement is an important determinant for dividend smoothing. Other factors are the present and future cash needs of the parent company and the subsidiary.

As Skinner (2007) writes, “In the end, then, dividends remain as much of a puzzle as always” – even more so with respect to dividend smoothing in the intra-­firm context. In particular, the lack of empirical evidence on a wider scale creates uncertainty as to more generalizable explanations for the low degree of smoothing in the parent–subsidiary context. Further research should therefore focus on using country­specific information to explain the differences in dividend smoothing between the countries. Also, the question of where the dividend decisions are actually made is relevant, because it seems that, contrary to the theoretical approach where “a firm” decides about its dividend payments, in the intra-­firm context it is often the parent company that decides about the dividend policy of the subsidiary, rather than the subsidiary itself. For example, in the case of Germany, in the personal shareholder context, frequently, the notion that large shareholders systematically co-­determine the dividend decision has been raised repeatedly (e.g. Behm and Zimmermann 1993; see also the result of Khan 2006 quoted on p. 000). The parent company can be considered as a “large shareholder”, as German firms prefer majority ownership of their subsidiaries in general. Therefore, the assumption that dividend policies are made in the interest of the overall MNE, and not just according to the needs of one single subsidiary, seems plausible. However, this decision might then deviate from the optimum dividend decision of a particular subsidiary.

114   C. Bellak and N. Wiedermann-Ondrej

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Notes 1 Christian Bellak is at the Department of Economics, Vienna University of Economics and Business, Vienna (Austria). Nadine Wiedermann-­Ondrej is at the Ministry of Finance, Vienna (Austria). The views expressed in this article represent the authors’ personal opinion and do not necessarily reflect the views of the Ministry of Finance. 2 An example of a study surveying some studies in the personal shareholder sphere is Andres et al. (2009). 3 We have discussed only three out of the four studies on intra-­firm dividends included in our survey, as the omitted study is based on aggregate data rather than on the firm-­ level data. 4 For the case of Germany, there is evidence that the (legal) tax environment of the home and host country is relevant not only to the dividend policy but also to the switch to other forms of profit shifting. This result has also been obtained by various studies of the US case (e.g. Altshuler and Grubert 2002). 5 The precondition – as outlined by Lintner in 1956 – is that the increase in the income is believed to be permanent. However, Benartzi et al. (1997) found only limited support. 6 Amended in 2003 according to Korb-­II-Gesetz, 2003.

6 Cross-­border hybrid finance and tax planning

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Does international tax coordination work? Ewald Aschauer, Eva Eberhartinger and Wolfgang Panny Hybrid financial instruments are neither typical equity nor typical debt and often lead to classification conflicts, especially in cross-­border transactions. The use of hybrid financial instruments for intra-­group financing offers the chance of possible double non-­taxation. However, a parent company that wishes to finance its foreign subsidiary via hybrid instruments faces uncertainties in multiple ways. The chance of double non-­taxation is connected to the risk of misclassification and double taxation. This chapter analyzes the influence of the existence or non-­ existence of a double tax convention on the benefits of using hybrid finance. We have therefore modeled the influence of uncertainty with respect to tax law to these financial decision using probability trees. We find that the existence of a double tax convention does not necessarily reduce the expected total tax burden. In many cases, the expected tax burden is even higher.

6.1  Introduction It is generally conceived that income taxation of cross-­border transactions should be designed such as to avoid double taxation in both countries involved, namely the source state and the residence state. Double taxation leads to excessive taxation that compromises cross-­border business, thereby precluding the economy from further development. Avoiding double taxation is therefore in the interest not only of the individual or the enterprise, but also of national states. To that effect, many states have, in their national tax laws, unilateral measures which avoid double taxation. In addition to that, bilateral double tax conventions (DTCs) are signed, which are based on generally accepted rules of the allocation of taxing rights, as can be found in the OECD Model Convention. As such, DTCs are the result of the tax coordination efforts of the countries involved. Within the European Union, tax coordination goes even further: in specific cases, income taxation is harmonized. DTCs normally do not go beyond the aim of avoiding double taxation. In particular, they do not seek to avoid non-­taxation of profits and do not ensure single taxation. However, more recently states have become increasingly worried about cases of double non-­taxation. Non-­taxation may arise from the other state failing

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

116   E. Aschauer et al. to use its right to tax, or because tax planning measures of enterprises (and individuals) take advantage of loopholes. As a result, more recent DTCs include subject-­to-tax rules that grant taxing rights to one state only if the transaction is actually taxed there; otherwise, the taxing right remains with the other state. However, generally states do not continually renegotiate the older treaties, and hence, subject-­to-tax rules are not as frequently found as one might suspect (Schilcher 2009). Some states use treaty-­overriding clauses in their national tax laws to ensure single taxation. When focusing on intra-­group finance via equity or debt, tax harmonization via DTCs usually works: normally, the yield (dividend or interest) is taxed in the residence state, and the source state has a right to levy withholding tax, which is credited by the residence state. However, the existing rules only cover typical cases of equity or debt and do not account for the large variety of financial transactions that can be found in reality. As soon as the form of finance or its taxation departs from the typical case, problems arise. One example is the case of thin capitalization, where the non-­deductibility of interest payments in the source state leads to double taxation of interest payments if the residence state does not reclassify accordingly. Another example – and one that will be the basis for this chapter – is the case of hybrid finance, which can lead to a conflict of qualification, and thus to double taxation or double non-­taxation. Previous research by Eberhartinger and Six (2009) demonstrated that from the legal point of view, neither DTCs nor the relevant EC directives (the parent– subsidiary directive, interest directive and royalties directive) achieve their aim (in short: to avoid double taxation and to eliminate withholding tax) in the case of hybrid finance. For a multinational group, engaging in cross-­border hybrid finance is connected not only to a risk of double taxation but also to a chance of non-­taxation, in spite of tax-­coordinating measures. Thus (in the absence of subject-­to-tax rules), a DTC does not ensure single taxation, and the question of whether untaxed income is supported by the existence or non-­existence of a DTC arises. This chapter attempts to identify the effect of the existence or non-­existence of a DTC on the expected total tax burden. Its focus lies on a group seeking to minimize its total tax burden using hybrid finance in a double non-­taxation scenario. It is assumed that the enterprise has limited knowledge about the correct tax treatment of the hybrid instrument in the foreign country, in the domestic country, and according to the DTC. Therefore, the enterprise in practice faces uncertainty with regard to the tax treatment, which is included in our model. Additionally, it faces uncertainty as to whether a tax audit will take place and a fine will be imposed. However, the enterprise’s managers are sufficiently educated to know about the probable classification. We have used probability trees that allow one to conveniently describe the situation and eventually derive the expected value of the total tax burden. The following three statements with regard to the total tax burden of probably untaxed yield resulting from intra-­group cross-­border hybrid finance can be hypothesized:

Cross-border hybrid finance   117 • • •

The existence of a DTC reduces the expected total tax burden. The existence of a DTC reduces the variance of the total tax burden. The existence of a DTC increases the probability of a total tax burden below the level of single taxation (or vice versa: decreases the probability of double or triple taxation).

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The following section will further explain the tax treatment of cross-­border hybrid finance. Section 6.3 explains the model, while section 6.4 presents the results and their analysis. Finally, section 6.5 discusses the limitations of our model and presents the conclusion.

6.2  Cross-­border hybrid finance and international taxation Hybrid financial instruments (mezzanine finance) are neither typical equity nor typical debt. The spectrum of hybrid instruments ranges from corporate shares with features typical of loans (such as certain preference shares) to loans with features usually associated with equity investments (such as profit participation). Such equity-­type loans include silent partnerships, participating bonds, convert­ ible bonds or warrant bonds, among the other profit-­participating loans.1 For tax purposes, hybrid instruments must usually be treated either as equity or as debt, and the yield is consequently treated either as non-­tax-deductible dividend or as tax-­deductible interest payment (Six 2007, 2009; Wiedermann-­ Ondrej 2007, 2008). In the hands of the recipient, the yield is normally treated as dividend received with preferential tax treatment or as taxable interest received. Only in rare cases is the fiscal treatment of an instrument split (the bifurcation method). For fiscal authorities, the wide variety of hybrid instruments in use and their rapid evolution make an unambiguous classification of any given hybrid instrument difficult, even in a purely national context. If two countries are involved, different classifications in the two countries – a conflict of qualification – cannot be excluded (Lang 2009). If the payment is deductible as interest in the source state and exempt as dividend in the residence state of the parent, then the result will be double non-­taxation. In the opposite case, where the hybrid instrument is treated as equity in the source state and as debt in the parent’s state, the difference in classification may result in double taxation. A parent company that wishes to finance its foreign subsidiary via hybrid finance, motivated by possible tax advantages from double non-­taxation, thus faces tax uncertainties of different degrees: • •

It will have to clarify the income tax treatment in the subsidiary’s state, i.e. the source state. Depending on how developed the tax system of the source state is, the degree of uncertainties may be very high. It will have to clarify whether the source state has a right to levy withholding tax and whether it actually levies withholding tax for the instrument in question.

118   E. Aschauer et al. •

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

•

It will have to clarify the income tax treatment in its own country of residence. Usually, the degree of uncertainty is lower there. Finally, it will have to clarify whether the foreign withholding tax (in the event that it was withheld) can be credited to its corporate tax burden.

For each of the questions, the company might be badly advised and could misjudge the legal situation. In that case, penalty payments could even arise. Thus, in contrast to plain debt (or plain equity), the chance of double non-­ taxation is connected to the risk of double taxation and the risk of misclassification. It can be assumed that under the existence of a DTC, the uncertainties are reduced. The legal classification of intra-­group, cross-­border hybrid finance, in the absence and in the presence of double tax conventions, has been summarized by Eberhartinger and Six (2009) as shown in Figures 6.1 and 6.2.

6.3  Model The model is based on the following situation. The parent company has decided to use a specific hybrid instrument for financing the foreign subsidiary. The reason for preferring such instrument to “normal” equity or debt finance lies in the probable tax advantages, indicating that the hybrid instrument offers the chance of double non-­taxation. However, the company faces uncertainty in several aspects: the appropriate tax treatment itself may be unclear in the absence of unambiguous legal provisions, case law or administrative guidelines, and the fisc may not share the company’s view about the appropriate tax treatment of the instrument. This situation can appropriately be modeled through a probability tree for a specific hybrid instrument. The tree covers all possible outcomes (and their probabilities): not only the preferred and probable outcome of double non-­taxation, but also all other possible and improbable outcomes as a result of differing legal qualification. 6.3.1  Assumptions On the basis of the previous chapters and the legal analysis of Eberhartinger and Six, we will calculate the expected value of the differential tax burden. The following assumptions apply: • • • • •

Both the parent and the subsidiary are companies subject to corporate tax. Parent and subsidiary are fully tax liable; they incur no losses. The parent holds a 100 percent interest in the subsidiary. The subsidiary is not within the European Union; the EC directives thus do not apply. The parent is risk-­neutral. Either it has a much diversified portfolio or it has signed a comfort letter, leading to full responsibility for the subsidiaries losses, independent of the form of the marginal finance.

cts ctr

TOTAL TAX BURDEN

Yes

cts wts ctr

No cts wts

No

No

Dividend

cts ctr

Yes cts wts ctr

No cts ctr

No

Yes

cts

No

No

Dividend

SS � Source state RS � Residence state WI � White income

Yes

Interest

No

ctr

Yes

Yes

wts ctr

No

wts

No

No

Dividend

Yes

ctr

Yes

Yes

wts ctr

No

Interest

No No

ctr

No

SS RS

No ctr

WI

Yes

Interest

No

No

Dividend Yes

Interest

cts � Corporate tax in source state ctr � Corporate tax in residence state wts � Withholding tax in source state

cts ctr

No

Yes

Interest

Figure 6.1  Taxation of cross-border hybrid finance without double tax conventions.

Yes

Tax credit in RS?

Corporate tax in RS?

Qualification in RS?

Yes

Yes

Corporate tax in SS?

Withholding tax in SS?

Dividend

Qualification in SS?

Yield

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Yes

cts ctr

Tax credit in RS?

TOTAL TAX BURDEN

cts wts

No

No

cts ctr

Yes4

Yes

cts wts ctr

No4

Interest

cts

No

No

cts ctr

No

Yes

ctr

Yes4

Yes

wts ctr

No4

wts

No

No

Dividend

Yes

ctr

Yes4

Yes

Interest

No

ctr

No

Yes

SS

ctr

WI

Yes

Interest

No

No

No

No

Dividend

Interest

cts � Corporate tax in source state ctr � Corporate tax in residence state wts � Withholding tax in source state

Interest

SS � Source state RS � Residence state WI � White income

cts ctr

No

Yes

Dividend

No

Figure 6.2  Taxation of cross-border hybrid finance under double tax conventions.

Yes

Dividend

Corporate tax in RS?

Qualification in RS?

Yes

Yes

Corporate tax in SS?

Withholding tax in SS?

Dividend

Qualification in SS?

Yield

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Cross-border hybrid finance   121 • • • •

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

• • • • • • • • •

Tax is decisive for the choice of the form of finance. Parent and subsidiary are taxed individually, and there is no form of consolidated or additive group taxation. The correct tax treatment is determined by the national foreign and domestic fisc. The group is not as fully informed about the tax treatment of the hybrid instrument as it should be. Costs of tax advice and costs of litigation are not included. Yield of hybrid instruments must be classified for tax purposes as either dividend or interest. Interest payments are fully tax-­deductible at the subsidiary’s level; however, they are fully taxable at the parent’s level. Withholding tax may or may not arise. Dividend payments are not deductible at the subsidiary’s level; however, they are fully exempted at the parent’s level. Withholding tax may or may not arise. Subject-­to-tax rules do not apply. Only corporate tax and withholding tax are considered. No other taxes on capital transactions are considered. Rules on thin capitalization as well as rules on minimum equity do not apply. The probability of a tax audit is higher in those cases where no tax is paid by the parent in the residence state or by the subsidiary in the source state. In the event that a tax audit reveals that the non-­taxation is unjustified and that the tax should have been paid, the difference is refunded to the fisc and a fine is levied.

6.3.2  Tree Figure 6.1, showing the possible tax treatments of a given hybrid instrument in a non-­DTC-case, is reflected in the probability tree of Figure 6.3. For reasons of clarity, Figure 6.3 assumes corporate tax rates of 25 percent in both countries and a withholding tax rate of 30 percent. Figure 6.4 translates Figure 6.2 (the DTC case) into the pertaining probability tree. As can easily be seen, the tree has the very same structure. All that changes is – let it be assumed – the withholding tax rates, which are reduced under DTC. Also, in both trees the path to the extreme right is the path that the enterprise intends to follow, as it leads to untaxed income. 6.3.3  Probabilities of legal classification The above-­mentioned trees also include the aspect of legal uncertainty on the classification of a hybrid instrument. In more colloquial language: we have tried to reflect the consultant’s opinion that the instrument will “probably” not be taxed, with the application of all caveats. At each branch node, the probability

122   E. Aschauer et al. for the left- or right-­hand branch is p or 1 − p, respectively. The sum of the probabilities for the two possibilities is 1. As has been shown earlier, the structure for the tree with and without a double tax convention is similar. However, the probabilities differ in general. It can reasonably be assumed that a tax credit for withholding tax is highly probable in the presence of a double tax treaty, and that the classification of a hybrid instrument in the residence state is more likely to be in line with its classification in the source state.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Hybrid financial instrument

Corporate tax in SS?

Withholding tax in SS?

Yes

No 20%

50%

50%

Yes

No

Yes

No

Yes 50%

60%

80%

20%

Corporate tax in RS?

40%

80%

20%

80%

No

Yes

No

50%

Tax credit in RS?

Yes

No

Outcome j

1

2

80%

Yes 50%

3

4

80%

20%

No

Yes

No

8

9

10

50%

Yes

No

6

7

5

20%

Figure 6.3  Decision tree cross-border hybrid finance without a double tax convention.

Hybrid financial instrument

Corporate tax in SS?

Withholding tax in SS?

Yes

No 20%

50%

50%

Yes

No

Yes

No

Yes 95%

60%

80%

10%

Corporate tax in RS?

40%

90%

10%

90%

No

Yes

No

5%

Tax credit in RS?

Yes

No

Outcome j

1

2

90%

Yes 95%

3

4

5

10%

90%

10%

No

Yes

No

8

9

10

5%

Yes

No

6

7

Figure 6.4  Decision tree cross-border hybrid finance with a double tax convention.

Cross-border hybrid finance   123 According to the trees, a probability pj can be assigned to each outcome j. Thus, pj denotes the probability that a sequence of branching results in the terminal node j. In general, pj for a certain node j is different in the trees with and without DTC.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

6.3.4  Tax audit However, if the company chooses to declare the yield in the source state as tax-­ deductible interest and tax-­exempt dividend in the residence state (node j = 10), then the probabilities for all the other nodes are relevant only to the extent that they can be “detected” in the case of a tax audit. In other words, irrespective of the appropriate tax treatment in the eyes of the fisc, the yield will remain untaxed (maybe incorrectly untaxed) unless a deviant tax treatment is ascertained during a tax audit. Thus, the probability pj for j = 1 . . . 9 depends on the probability of a tax audit. Figure 6.5 shows the possible outcomes in the event of a tax audit. We assume that the probabilities of a tax audit in the source state and the residence state are independent of each other. This is in line with the observation that coordination between domestic and foreign tax authorities, in spite of agreements, is rare. Possible outcomes Audit in SS?

Audit in RS?

Yes

No

Yes

No

Yes

No

1

3

9

10

2

5

10

3

8

4

10

5 6 7 8 9 10

Figure 6.5  Consequences in the event of a tax audit.

124   E. Aschauer et al. We further assume that in the source state, a tax audit can be carried out with respect to either corporate tax or withholding tax. However, in the course of the audit, possible errors not only in the original tax audited (corporate tax or withholding tax) but also in the “other” tax (withholding tax or corporate tax) will be corrected. This corresponds to current tax audit routines. The probability of a tax audit in the source state or resident state can therefore be described as:

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

aSS = ac , SS ⋅ aw, SS + ac , SS ⋅ (1 − aw, SS ) + (1 − ac , SS ) ⋅ aw, SS aRS = ac , RS ⋅ atc , RS + ac , RS ⋅ (1 − atc , RS ) + (1 − ac , RS ) ⋅ atc , RS



(1)

For outcome j = 10, for the case of double non-­taxation, aSS > aRS: in the source state, two auditable taxes are concerned, while in the resource state, only corporate tax is possibly due (atc,RS = 0). The combination of pj (the probability of classification as j) with the respective probabilities for a tax audit leads to the probability for each node j to actually occur, under the assumption that the enterprise decides to treat the hybrid instrument as fully untaxed in both countries. Pj =1 = aSS ⋅ aRS ⋅ p j =1 Pj = 2 = aSS ⋅ aRS ⋅ p j = 2 Pj =3 = aSS ⋅ aRS ⋅ p j =3 + aSS ⋅ (1 − aRS ) ⋅ p j =3 Pj = 4 = aSS ⋅ aRS ⋅ p j = 4 Pj =5 = aSS ⋅ aRS ⋅ p j =5 + aSS ⋅ (1 − aRS ) ⋅ p j =5 Pj = 6 = aSS ⋅ aRS ⋅ p j = 6



(2)

Pj = 7 = aSS ⋅ aRS ⋅ p j = 7 Pj =8 = aSS ⋅ aRS ⋅ p j =8 + aSS ⋅ (1 − aRS ) ⋅ p j =8 Pj =9 = aSS ⋅ aRS ⋅ p j =9 + aSS ⋅ (1 − aRS ) ⋅ p j =9 + (1 − aSS ) ⋅ aRS ⋅ p j =9 Pj =10 = aSS ⋅ aRS ⋅ p j =10 + aSS ⋅ (1 − aRS ) ⋅ p j =10 + (1 − aSS ) ⋅ aRS ⋅ p j =9 + (1 − aSS ) ⋅ (1 − aRS ) ⋅ p j =9 6.3.5  Payments in the case of audit In case of a tax audit, two possible outcomes arise if the enterprise chooses j = 10: 1 2

If the audit confirms the tax treatment that the enterprise has chosen, no further consequences arise. If the audit does not confirm the tax treatment, then the enterprise will have to remargin the tax and, as is common in many jurisdictions, will have to pay an additional fine.

Cross-border hybrid finance   125 The tax due (TB) as a result of an audit is defined as follows, in ten possible outcomes, j: 2 TB j =1 = τC , SS + (1 − τC , SS )τW , SS + (1 − τC , SS )τC , RS − tcRS TB j = 2 = τC , SS + (1 − τC , SS )τW , SS + (1 − τC , SS )τC , RS TB j =3 = τC , SS + (1 − τC , SS )τW , SS TB j = 4 = τC , SS + (1 − τC , SS )τC , RS

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

TB j =5 = τC , SS TB j = 6 = τW , SS + τC , RS − tcRS



(3)

TB j = 7 = τW , SS + τC , RS TB j =8 = τW , SS TB j =9 = τC , RS TB j =10 = 0 min((1 − τC , SS )τW , SS ; (1 − τC , SS )τC , RS ), if j = 1...5 tcRS =   min(τW , SS ;τC , RS ), if j = 6...10

(4)

In addition, the fine due is defined as a percentage of the too-­little tax paid. FSSj = TBSSj ⋅ γ SS FRSj = TBRSj ⋅ γ RS



(5)

6.3.6  Total tax burden As a result, the total expected tax burden (inclusive of the fine) is: TTBSSj = TBSSj + FSSj TTBRSj = TBRSj + FRSj



(6)

TTB j = TTBSSj + TTBRSj From this, the expected value (ETB) of the total tax burden – more precisely, the difference between the targeted zero taxation and the tax that should be paid – can be derived:

ETB =

10

∑ TTB ⋅ P j

j =1

j

(7)

126   E. Aschauer et al.

6.4  Results As indicated earlier, three measures will be used to clarify the influence of the (non-)existence of a DTC on the advantage of using hybrid finance that can reasonably be classified as being untaxed, but whose “correct” (in the eyes of the fisc) classification is not certain: 1

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

2 3

The expected total tax burden ETB is lower in the DTC case: ETBnoDTC − ETBDTC = ΔETB > 0. The variance of the total tax burden ETB is lower in the DTC case: varTBnoDTC − varTBDTC = ΔvarTB > 0. The probability of a total tax burden of less than 25 percent, P(TTB < 25%), is higher in the DTC case: P(TTB < 25%)DTC − P(TTB < 25%)noDTC = ΔP(TTB < 25%) > 0. The benchmark of 25 percent represents a level of taxation equivalent to single taxation, which is assumed to be the maximum tolerable burden for the enterprise.

In a first setting in Table 6.1, the variables of the no-­DTC case and the DTC case are fixed. The result is as expected: the existence of a DTC leads to considerably less expected tax burden, and the probability of a burden of less than 25 percent has risen by 9.24 percentage points to almost 60 percent. A DTC providing for reduced withholding tax is also advantageous for enterprises in complex cases such as hybrid finance. The reduction of withholding tax is an integral part of a DTC, but it is neither mandatory nor necessary; not all DTCs provide for such reduction. It is useful to refer to Table 6.2 to understand what the result would be if a DTC did not provide a reduction of withholding tax. The table enables us to reveal further effects of a DTC. Table 6.1  Variables, withholding tax reduced by DTC

τc,ss τc,rs τw,ss, dividend τw,ss, interest pj aSS aRS Γ ETB ΔETB varTB ΔvarTB P(TTB < 25%) ΔP(TTB < 25%)

No DTC

DTC

25% 25% 30% 30% cf. Figure 6.3 51% 30% 30% 23.16   7.14   6.14   3.40 48.87   9.24

25% 25%   5% 10% cf. Figure 6.4 51% 30% 30% 16.02   7.14   2.74   3.40 58.11   9.24

Cross-border hybrid finance   127

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 6.2  Variables, withholding tax not reduced by DTC

τc,ss τc,rs τw,ss, dividend τw,ss, interest pj aSS aRS Γ ΔETB ΔvarTB ΔP(TTB < 25%)

No DTC

DTC

25% 25% 30% 30% cf. Figure 6.3 51% 30% 30%   0.05   0.55 −1.93

25% 25% 30% 30% cf. Figure 6.4 51% 30% 30%   0.05   0.55 −1.93

It is evident from the table that the difference between the no-­DTC scenario and the DTC scenario can be found only in the different pj values. All other variables are equal. It is not surprising that the elimination of the effect of reduced withholding tax leads to practically the same expected tax burden, as all tax rates are left unchanged. Also, the variance is smaller in the DTC scenario, which indicates that the total tax burden for each outcome is closer to the expected tax burden when there is a DTC. However, it is surprising to note that the probability of a tax burden below the assumed level of withholding tax is smaller in the DTC case. In other words, it is more probable to end up with a total tax burden of more than 25 percent if there is a DTC. These exemplary results need to be tested for generalizability. The following variables will be varied: • • • • •

the probability of a tax audit in the source state, aSS; the probability of a tax audit in the residence state, aRS; the fine factor, γ; the probability of classification of the instrument as equity or debt for corporate tax in the source state, pj,c,SS; the corporate tax rate in the source state, τc,SS, and the corporate tax rate in the residence state, τc,RS.

All variations are made on a ceteris paribus assumption, based on Figure 6.3 and 6.4. For easy evaluation of the curves shown in Figure 6.6, it must be remembered that only a result above zero (for all the three results tested) supports the hypothesis. We hypothesized that the mere existence of a DTC (also without providing for reduced withholding tax) reduces the value of ETB, because cases of double or triple taxation are eliminated. Figure 6.6 shows that this is true only as long as the probability of a tax audit in the source state is high (>50 percent). If it is low, then the value of ETB becomes higher without a DTC. This indicates that the

128   E. Aschauer et al. 3% 2% 1%

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Probability

0%

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

�1% �2% �3% �4% �5% �6%

E P var

�7%

Figure 6.6  Results depending on the probability of a tax audit in the source state. Note For readability, all the following graphs have abbreviated denominations: E instead of ΔETB; p instead of ΔP(TTB < 25%); var instead of ΔvarTB.

existence of a DTC does not necessarily improve the group’s expected tax burden in a case where it seeks double non-­taxation and faces uncertainty. Also, in contrast to our hypothesis, the probability of a tax burden below 25 percent is higher without a DTC. As shown by the ascending graph in Figure 6.6, if the probability of a tax audit in the source state becomes higher, then the difference between the case with a DTC and that without a DTC becomes smaller. Only the variance of the total tax burden is constantly lower in the DTC case, as expected. The existence of a DTC leads to a reduction in the number of cases of triple taxation (in particular, via the highly probable tax credit), which in itself must reduce the variance. Figure 6.7 shows the effect of the existence of a DTC on the probabilities of the resulting non-, single, double or triple taxation, depending on the probability of a tax audit in the source state. The ordinate shows the difference in the respective probabilities between the case without and with a DTC. In the absence of a tax audit, the probability of ending up with non-­taxation becomes 3 percent higher without a DTC than with a DTC. With an increasing probability of a tax audit, even with a certain tax audit, the probability of non-­ taxation becomes further higher (but decreasingly) in the non-­DTC case when compared with the DTC case. On the contrary, the probabilities of ending up with single taxation but also with double or even triple taxation increase further in a non-­DTC case than in a DTC case. It is obvious that the existence of a DTC has a “concentrating” effect on the tax rates, and thus the variance of the tax rates is lower in the DTC case, as is shown in Figure 6.6. In Figure 6.8, the probability of a tax audit in the residence state is varied. It shows that the expected tax burden with or without a DTC differ only marginally. As in Figure 6.6, the probability of a low tax burden (up to 25 percent) is

Cross-border hybrid finance   129 5%

Triple Double Single Non

3% 2% 1% 0%

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Figure 6.7 Difference in the probability of non-, single, double, or triple taxation, depending on the probability of a tax audit in the source state.

higher in the absence of DTC. Again, only the variance of the total tax burden is lower in the case of DTC, as expected. Figure 6.9 shows that the variation of the fine factor has hardly any effect on the difference in the expected tax burden (which includes the possible fine factor) without and with a DTC. As is observed in the previous figures, the probability of low taxation is lower in the case of a DTC when compared with that in the absence of a DTC. However, the difference in the probability of low taxation with a DTC or without a DTC is insensitive to the fine factor (curve p is horizontal), because only the outcome j = 10 (double non-­taxation) leads to a tax burden below 25 percent, only the outcome j = 10 is covered, and, obviously, j = 10 does not involve fines. Again, the difference in variance is positive, as expected. 3% 2% 1% 0% Probability

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Probability

4%

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

�1% �2% �3% �4% �5% �6%

E P var

�7%

Figure 6.8  Results depending on the probability of a tax audit in the residence state.

1.0

130   E. Aschauer et al. 3% 2% 1% 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

�1% �2% �3% �4% E P var

�5% �6% �7%

Figure 6.9  Results depending on the fine factor, γ.

The previous results obviously depend strongly on the specific p for the specific taxes in both countries. The trees on which our analysis is based (Figures 6.3 and 6.4) show very similar probabilities. However, a question arises as to whether the results are still valid when the probability of corporate tax in either country varies. Figure 6.10 shows the sensitivity of the results to the degree of probability with which the group assumes that the yield of the hybrid instrument is liable to corporate tax. The left end of the curves indicates certainty regarding the deductibility in the source state. It is assumed that irrespective of the qualification in the source state, it has no influence on the qualification in the residence state, and that both states qualify independent of each other, and pj,ct,ss and pj,ct,rs are not related to each other. Thus, the previous results are again confirmed: the 3% 2% 1% 0% Probability

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Probability

0%

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

�1% �2% �3% �4% �5% �6%

E P var

�7%

Figure 6.10  Results depending on the probability of a corporate tax in the source state.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Cross-border hybrid finance   131 expected total tax burden hardly differs whether there is or is not a DTC. Only when the probability of corporate taxation in the source state is low (and thus, the probability of deductibility is high) is the existence of a DTC beneficial to the group. Again, the probability of low taxation is always higher in the absence of a DTC, and the variance of the total tax burden is, as expected, constantly lower in the DTC case. Finally, Tables 6.3–6.5 show that the results are also robust with regard to the respective corporate tax rates. The expected tax burden is decreased by the existence of a DTC only if the rates are low. The probability of a total tax burden below 25 percent is decreased if a DTC exists in almost all combinations. However, the variance of the tax burden is lower when a DTC exists. Table 6.3  Results for E, depending on the corporate tax rate in SS and RS ETB

CT SS

CT RS 5.00% 15.00% 25.00% 35.00% 45.00% 55.00%

5.00% 0.02% 0.05% 0.08% −0.02% −0.025% −0.49%

15.00% 0.01% 0.04% 0.06% −0.04% −0.29% −0.53%

25.00% 0.01% 0.03% 0.05% −0.07% −0.32% −0.57%

35.00% 0.01% 0.02% 0.03% −0.10% −0.36% −0.61%

45.00% 0.00% 0.00% 0.01% −0.12% −0.39% −0.66%

55.00% 0.00% −0.01% −0.01% −0.15% −0.43% −0.70%

Table 6.4  Results for p, depending on the corporate tax rate in SS and RS pTB

CT SS

CT RS 5.00% 15.00% 25.00% 35.00% 45.00% 55.00%

5.00% 0.00% 0.12% −1.81% −1.81% −1.81% −1.81%

15.00% 0.12% 0.12% −1.81% −1.81% −1.81% −1.81%

25.00% 0.00% 0.00% −1.93% –1.93% −1.93% −1.93%

35.00% 0.00% 0.00% −1.93% −1.93% −1.93% −1.93%

45.00% 0.00% 0.00% −1.93% −1.93% −1.93% −1.93%

55.00% 0.00% 0.00% −1.93% −1.93% −1.93% −1.93%

Table 6.5  Results for var, depending on the corporate tax rate in SS and RS Var CT RS 5.00% 15.00% 25.00% 35.00% 45.00% 55.00%

CT SS 5.00% 0.10% 0.32% 0.54% 0.70% 0.79% 0.84%

15.00% 0.11% 0.33% 0.55% 0.72% 0.81% 0.87%

25.00% 0.11% 0.33% 0.55% 0.73% 0.83% 0.89%

35.00% 0.11% 0.34% 0.55% 0.74% 0.84% 0.91%

45.00% 0.11% 0.34% 0.55% 0.74% 0.85% 0.91%

55.00% 0.11% 0.33% 0.55% 0.73% 0.85% 0.91%

132   E. Aschauer et al.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

6.5  Conclusion and future research This chapter attempted to identify whether the existence or non-­existence of a DTC supports untaxed income when intra-­group hybrid finance is employed with the aim of achieving double non-­taxation when facing uncertainty about the correct tax treatment. The results show that the existence of a DTC considerably reduces the expected tax burden if the DTC (as is usually the case) provides for a reduction of withholding tax in the source state. However, this is not surprising: lower tax rates lead to a lower tax burden (at least in this case). However, when one eliminates the withholding tax rate effect (that is, assuming that the DTC does not provide for lower rates), thus reducing our analysis to the effect of other DTC rules and to the tax laws of the countries involved, the result was different. On the one hand, the existence of a DTC reduced the probability of triple taxation, double taxation and non-­taxation considerably. However, on the other hand the existence of a DTC did not necessarily reduce the expected total tax burden. In many cases, the expected tax burden was even higher. This effect is considered to be the result of the reduced probability of double or triple taxation, which has less effect on the expected tax burden than does the reduced probability of non-­taxation. Thus, the total expected tax rate was found to be higher in the case of a DTC. This was particularly true when the probability of a tax audit was low; the differences decreased with an increasing probability of a  tax audit, but they generally did not reverse. Furthermore, the existence of a DTC did not increase the probability of a low total tax burden (<25 percent). On the contrary, we found that a DTC increased the probability of a total tax burden above 25 percent. This may be due to the fact that as soon as a minimum probability for a tax audit is assumed, there is always a detection risk and thus the risk of a fine, which adds to all possible cases of single taxation, and which raises all these cases above the 25 percent mark (if 25 percent is assumed as the corporate tax rate in both countries). Only the non-­taxation case remains below the 25 percent mark, with a decreasing probability. Our results are subject to some limitations. First, we assumed that both the parent and subsidiary are profitable and tax-­paying enterprises. We did not consider the losses, and thus the disadvantageous effect of loss carry-­forward (instead of immediate set-­off ) was not included in our study. However, earlier research showed that such an effect is crucial for the determination of effective tax rates of cross-­border equity finance, debt finance and hybrid finance (Eberhartinger and Pummerer 2009, 2010). Second, the specific ps in our model were not empirically determined, but were based on a skilled estimate by the authors. In addition, this chapter also justifies further approach to reality by including the cost of litigation as a variable in the possibility of challenging the fisc’s view on the “correct” taxation of the finance transaction. Furthermore, tax advice cost can also be included, which not only would cover the complexity of the legal question (the more complex, the more costly) but also would decrease uncertainty on the “correct” tax treatment.

Cross-border hybrid finance   133

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Appendix: list of abbreviations j τC,SS τW,SS τC,RS tcRS pj Pj aSS aRS ac aw atc γ F TBj TTB ETB var P(TTB < 25%)

outcome 1 . . . 10 nominal corporate tax rate in source state nominal withholding tax rate in source state nominal corporate tax rate in residence state tax credit in residence state probability of outcome j before tax audit probability of outcome j after tax audit probability of tax audit in source state probability of tax audit in residence state probability of tax audit of corporate tax probability of tax audit of withholding tax probability of tax audit of tax credit fine factor fine tax burden of outcome j total tax burden, inclusive of fine expected total tax burden variance of the total tax burden probability of a total tax burden of less than 25 percent

Notes 1 The term “hybrid finance”, as used in this chapter, does not include derivatives (options, futures, swaps). 2 We assume a limited tax credit, as prevalent in most countries worldwide.

7 Investigating the shift toward a value-­added-type destination-­ based cash flow capital income tax (VADCIT) Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Klaus Hirschler and Martin Zagler 7.1  Introduction The current system of corporate income taxation is strongly source oriented in terms of attribution of taxing power, which makes it vulnerable to tax planning measures (i.e. transfer pricing and systematic debt-­equity structuring). Firms certainly rely on transfer pricing to reduce profits. For example, a US company paid $3,050 per liter of mineral water imported from the Netherlands, another $8,252 per wristwatch imported from China, while a third sold airplane seats for only 10 cents to China (Pak and Zdanowicz 1994). Countries can fight the decline in the tax base by reducing the scope of tax planning, which is a tedious activity, and by reducing tax rates. The latter of course leads to a race to reduce tax rates among countries, which is empirically observable. It can be observed that, over time, statutory corporate income tax rates have declined from a median above 35 percent to a median around 25 percent. More importantly, the standard deviation has also declined, indicating that countries reduce their tax rates to set rates more in line with international competitors. While the decline in rates can be explained by efficiency gains due to a reduction in the excess burden of taxation, the decline in the standard deviation can only be interpreted by the race to the bottom in the international corporate income taxation. A similar pattern can also be found for effective average and marginal corporate income tax rates (Vondra 2008). These pieces of evidence imply that the current international system of corporate income taxation requires reform. The European Union has witnessed this trend for a long time, and several initiatives have been put in place to reform the current system of corporate income taxation. The initial proposals, the Neumark Report from 1962 and the Van den Tempel Report from 1970, call for a harmonization of corporate income tax rates across the Union. They were unsuccessful, just like the later proposals, the Draft Directive of 1975 and the Ruding Report of 1992. The Bolkenstein Report of 2001 was the first to propose a single common tax base for all EU-­wide operations. Recently, Devereux and Sørensen have discussed a number of proposals for future developments in corporate tax systems. One of these concepts is a

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The shift toward a VADCIT   135 briefly outlined value-­added-type destination-­based cash flow capital income tax (VADCIT) (Devereux and Sørensen 2006: 38ff.). These proposals, especially VADCIT, aim to circumvent the vulnerabilities of the current corporation income tax system. Thus, a reorientation of corporate income taxation is proposed to take place. The current system is strongly source oriented with respect to the attribution of taxing power, which makes it vulnerable to tax planning measures (i.e. transfer pricing and systematic debt–equity structuring). Devereux and Sørensen considered that instead of the current source-­based taxation of corporate profits, a destination-­based cash flow taxation could possibly prevent tax incentives of income shifting to low-­tax countries so that foreign investments would be treated in the same way as domestic investments (Meade 1978; Bradford 2000, 2004; Grubert and Newlon 1997; Bond and Devereux 2002; Bach 1993; Hiller 2003; Eberhartinger 2000). The issue of transfer pricing could be solved with the help of a corporate income tax, which is oriented to the system of value added taxes. Technically, taxation should result in cash flow taxation – that is, the tax base includes only cash flows from the firm’s “real” transactions and disregards financial transactions, including interest payments (Devereux and Sørensen 2006: 39). The taxable income for the proposed VADCIT is calculated as follows: gross receipts derived from sales to domestic customers in the particular country of destination of the company minus the initial or supplier costs in the particular country of destination minus accrued labor costs in the country of destination (Devereux and Sørensen 2005: 39). According to this system, sales to clients not based in the home country of the company would not be taxable as exports, and these payments would not be deductible in the foreign country if the purchased goods or services were the input factors for another company to provide services or produce and sell goods. As there is no tax levied on exports, and as imports are not deductible, the transfer price problem is solved because the transfer price has no impact on the tax base in the particular country. This chapter examines how far a VADCIT would be able to improve the current corporate income tax.

7.2  The VADCIT model 7.2.1  Calculating corporate tax by using VADCIT VADCIT makes transfer pricing less attractive. If a higher or lower transfer price has no effect on the taxable income in the exporting country or correspondingly in the importing country, then a transfer price organization with the aim of taking advantage of the difference in the tax rates between two countries cannot be executed to any further extent. To confirm this hypothesis, Table 7.1 shows the current tax system and Table 7.2 demonstrates the proposed VADCIT.

136   K. Hirschler and M. Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 7.1  Tax base and tax revenues according to the current legal situation Current legal situation

Transfer price at arm’s length

Transfer price too high

Corporation tax, country A Corporation tax, country B

25% 30%

25% 30%

Parent company (located in A, sells in B) Sales (country B) Expenses (country A) Wages Costs (country B) Economic profit (country A) Tax (country A)

50 30 10  0 10   2.5

60 30 10  0 20  5

Subsidiary (located in B, purchases from A, sells in B) Sales (final customer, country B) 100 Expenses (country B)   50 Wages    5 Economic profit (country B)   45 Tax (country B)   13.5 Effective tax rate subsidiary   30% Overall tax burden Overall effective tax rate (tax/total profit) Tax revenue, country A Tax revenue, country B

  16   29.091%    2.5   13.5

100 60 5 35 10.5 30% 15.5 28.182% 5 10.5

The current legal situation encourages firms to shift their tax burden with the help of transfer prices from high-­tax country B to low-­tax country A. We can illustrate this effect using a case study of a simple group structure with a parent company (State A) based in a different country and its subsidiary (State B). The producing parent sells the goods for a certain transfer price to its subsidiary, which then sells the goods to the final consumer. The subsidiary’s expenses for purchase from the parent company (import) are deductible as operating expenses in the subsidiary’s domestic country (B). Accordingly, the gross receipts of goods sold (export) are taxable in the parent company’s domestic state (A). A transfer price, which is higher than the one calculated according to the arm’s-length principle, reduces the taxable income in the high-­tax country B and raises the taxable income in the low-­tax country A. However, the opposite occurs if the transfer price was lower than that calculated at arm’s length. In the current system of corporate income taxes, the transfer price has an immediate impact on the group’s total tax burden. Table 7.2 illustrates that the internal transfer price has no effect on the group’s total tax burden if the export and import are not taxable and not deductible; hence, it does not affect net income. In the home country of the parent company, there are no taxes levied on the income gained through the goods sold to the subsidiary (export). Furthermore, payments from the subsidiary to the

The shift toward a VADCIT   137

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 7.2  Tax base and tax revenues with VADCIT (1) VADCIT – A and B in different countries

Transfer price at arm’s length

Transfer price too high

Corporation tax, country A Corporation tax, country B Sales price (final customer) Intra-group sales price Production costs (country A)

25% 30% 100 50 30

25% 30% 100 60 30

Parent company (located in A, sells in B) Sales (country B) Expenses (country A) Wages Costs (country B) Economic profit Profit according to VADCIT Tax (country A)

50 30 10 0 10 −40 −10

60 30 10 0 20 −40 −10

Subsidiary (located in B, purchases from A, sells in B) Sales (country B) 100 Expenses (country A) 50 Wages 5 Economic profit 45 Profit according to VADCIT 95 Tax (country B) 28.5 Effective tax rate subsidiary 30%

100 60 5 35 95 28.5 30%

Overall tax burden Overall effective tax rate (tax/total profit) Tax revenue, country A Tax revenue, country B

18.5 33.636% −10 28.5

18.5 33.636% −10 28.5

parent (import) are non-­deductible, and the amount of internal transfer prices does not affect the group’s total tax burden. Table 7.3 shows a variation of the previous case study shown in Table 7.2. Here, the producing parent company is now based in the high-­tax country and the marketing subsidiary is based in a low-­tax country. Though this does not change the impact of a VADCIT on the issue of transfer pricing, it can be noted that the effective tax rates vary much more dramatically. Under transfer pricing, the effective tax rate was around 28 percent and 29 percent, respectively, and thus within the limits of the two statutory rates of 25 percent and 30 percent. However, this is no longer the case under a VADCIT. Here, the effective tax rates are 33 percent and 21 percent, respectively. If the exporting country has a lower statutory rate, then the firm has a high tax burden (Table 7.2). If the exporting country has a lower rate, then the firm has a low overall tax burden (Table 7.3). We can therefore interpret a low statutory tax as an export subsidy. The exporting country pays for that export subsidy, both directly to the firm in terms of higher negative taxes and because of lower overall tax revenues from importing firms.

138   K. Hirschler and M. Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Table 7.3  Tax base and tax revenues with VADCIT (2) VADCIT – A and B in different countries

Transfer price at arm’s length

Transfer price too high

Corporation tax, country A Corporation tax, country B Sales price (final customer) Intra-group sales price Production costs (country A)

30% 25% 100 50 30

30% 25% 100 60 30

Parent company (located in A, sells in B) Sales (country B) Expenses (country A) Wages Costs (country B) Economic profit Profit according to VADCIT Tax (country A)

50 30 10 0 10 −40 −12

60 30 10 0 20 −40 −12

Subsidiary (located in B, purchases from A, sells in B) Sales (country B) 100 Expenses (country A) 50 Wages 5 Economic profit 45 Profit according to VADCIT 95 Tax (country B) 23.75 Effective tax rate subsidiary 25%

100 60 5 35 95 23.75 25%

Overall tax burden Overall effective tax rate (tax/total profit) Tax revenue, country A Tax revenue, country B

11.75 21.364% −12 23.75

11.75 21.364% −12 23.75

7.2.2  Comparison of VADCIT with VAT and a common consolidated corporate tax base (CCCTB) In its general structure, a VADCIT is similar to the existing VAT system. Exports are not taxable within the VAT system when there is an “intra-­ community delivery”. Within the VAT system, the purchaser has to declare an “intra-­community purchase” for taxation, and can deduct the “intra-­community sale” as an input tax. If VAT and input tax are the same, then there is no taxation of imports. Taxation within the VAT system will arise when a sale is made to a customer, and VAT will be payable in that country, where the sold good is used. This principal idea of VAT taxation in the country of use of the good is the main problem with a VADCIT, as will be shown later. When one compares VADCIT with the concept of a common consolidated corporate tax base (CCCTB) (“Company taxation in the Single Market” – Commission Staff Working Paper SEC 2001; Lang et al. 2008), there are a lot of differences, but one great common feature can be observed. Both systems have the idea of removing the influence of transfer pricing on the tax base. Within a

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The shift toward a VADCIT   139 CCCTB system, the removal of the influence of transfer pricing would be done by consolidating the profits and losses of all companies that are a part of the consolidation tax group. Within VADCIT, removal would be done by non-­deduction of the cost of imports. Except for this common feature, there are many differences. The CCCTB system is still a traditional corporate tax system, which calculates profits within the separate system of corporate tax and divides profits among companies that are part of the consolidation tax group. A VADCIT is a little different from a common consolidated corporate income tax base, where tax revenues are distributed solely on the basis of destination-­based revenues. The major difference is that the former is a cash flow tax, whereas the latter is not. The advantage of a cash flow tax is its investment neutrality; while the disadvantage is a high absolute flow of taxes between the firms and the government if the tax rate is not lowered. Both systems also subscribe to the idea of removing the influence of transfer pricing on the tax base. Within a CCCTB system, the removal of the influence of transfer pricing is done by consolidating the profits and losses of all companies that are part of the consolidation tax group. Within VADCIT, removal is accomplished by non-­deduction of the import price. However, the main difference lies in the implementation. A CCCTB has the advantage of allocating revenues according to other criteria as well. The existing systems of CCCTB, as well as the EU-­wide discussed system of CCCTB, are still (or would be) the traditional corporate tax system, which calculates the profit within the separate system of corporate tax and divides the profit among all companies that are part of the consolidation tax group.

7.3  Legal questions When introducing VADCIT, there is the question of compatibility with Article 33 of the EC VAT directive. According to Article 33, taxes that have the character of VAT are forbidden. Because of the fact that there is no deductibility of import as an input tax, it should not be difficult to introduce a VADCIT. Apart from the above-­mentioned discussion, there are other issues concerning the treaty of the European Community. Article 92, together with Article 87 (common subsidy prohibition) as well as Articles 90 and 91 (non-­discrimination through indirect tax rule), provide for a licensing requirement and an ex ante approval by the commission for levies other than indirect taxes if the levies result in either a relief or a refund for exported goods. The proposed VADCIT would necessarily need such an approval, as it results in a means mentioned in Article 92. The corporate income tax exemption for exports leads to the tax relief for services for non-­domestic customers. This non-­deductible expense for input-­ related expenditure paid in or to another country functions as a compensatory levy in terms of an economic subsidization of export prices. The problem of VADCIT can be directly deduced from Article 87. Furthermore, the ECJ decided in the case Commission/Greece that a tax refund concerning exports is a prohibited subsidy (ECJ, C-­183/91, Commission/Greece, Tz 17ff.). This judgment

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

140   K. Hirschler and M. Zagler must be taken into account for the future organization of VADCIT, because a tax-­free export is comparable to a tax refund concerning exports. In addition, the treatment of imported goods as non-­deductible and domestically purchased goods as deductible is incompatible with the current ECJ case law on discriminative tax legislature. At last it should be mentioned that a pure cash flow tax could create problems within the system of double tax treaties. Furthermore, the question of whether a system such as VADCIT is an income tax, especially when import costs are not deductible, could be raised. If this arises, then there could be a double taxation because there would not be a tax credit for the corporate tax of the permanent establishment in country A as well as in country B where the company is based.

7.4  Analysis of VADCIT 7.4.1  Scope of VADCIT: strategies to evade the new tax VADCIT is intended to replace corporate income taxation. If one of the companies involved in a cross-­border transaction does not have the legal form of a corporation, then VADCIT could no longer be applied if it is really planned only as a corporation income tax. To continue transfer-­pricing schemes, enterprises could switch to transparent legal forms (e.g. private limited companies) and disregarded entities. Likewise, if VADCIT is not implemented worldwide, then the transfer price problem could occur at the water’s edge of countries that apply it and in countries that have a different and incompatible concept of taxing corporate income. As a result, the success of the model in the scope of transfer price problems depends on the degree of international participation in the application of VADCIT. In this respect, VADCIT proves to have the same problem as that observed in the discussion on the implementation of a CCCTB in the European Union (see http://ec.europa.eu/taxation_customs/taxation/company_ tax/common_tax_base/index_en.htm). It is necessary to draw a clear line between the taxable sales in the corporation’s domestic country to domestic customers and the non-­taxable exports to non-­domestic customers. Therefore, the definition of the various items of VADCIT, such as “customers”, “domestic” or “sale”, is critical to its outcomes. In this context, one can orient oneself, at least concerning the supply of goods, on the system of intra-­community purchase and intra-­community supply (see EC Directive 77/388/EWG modified by Directive 91/680/EWG). It should be clarified whether the same restrictions concerning the VAT of intra-­community purchase or supply apply to a VADCIT, or whether there is a separate definition of the terms “export” and “import”. As far as there is a consonance with the VAT system, export shipments could be documented through VAT-­required audit trails and the system of “recapitulatory notification” (see Article 22 of the EC Directive as amended by Directive 91/680/EWG) The buyer can determine whether an intra-­community purchase (i.e. an import) has taken place from the

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The shift toward a VADCIT   141 seller’s invoice documents. For the seller, it is more difficult to detect whether the final destination of a given sale is within the border of his/her country of residence or not. On the one hand, the seller basically can make this demarcation concerning the consignment of goods, with regard to the shipment’s destination. On the other hand, distinguishing is not easy if the purchase takes place at the seller’s premises and the buyer himself or herself ships the goods to a foreign destination. If a transaction is arranged between two companies, and the purchaser transports the item to another state, then it is an intra-­community supply in terms of the VAT directive. Furthermore, for the selling entrepreneur it is an export for VAT purposes (Ruppe, UStG 1994 [2005] Art 1 BMR Rz 6 et seq.). It is questionable whether the same principle applies within the proposed VADCIT: this would mean that the gross receipts derived from this sale would not be taxable income for the selling entrepreneur. At the same time, the cost of the import of the foreign company would not be deductible, and therefore would not reduce the buyer’s tax base. The solution for VAT purposes would be entirely different if the goods were sold to a non-­firm customer who passes the goods to another country. In the sense of the EU VAT directive, this case would be traded as supply of goods to a “domestic customer”. This means that, although the good crosses borders within the European Union, no export or import occurs (see Article 1 in conjunction with Article 3 of the VAT Act). Again, it would be different if a non-­firm customer was the buyer and took the goods to a third country (a non-­EU member or a non-­VADCIT country, respectively). In such a case, it would be possible to assess it as an export, based on the export documents of the third country. If both cases were treated in the same way for corporate income tax purposes, a mismatch would occur between a non-­firm purchase of a company in a third country and a company in the European Union. The export to a third country could be proven, and, as a consequence, would affect the tax burden, whereas between the EU members a treatment as export would not be possible, owing to administrative constraints. For administrative purposes, every buyer is therefore considered as a domestic customer as long as the good is purchased in a registered office in the country and it is not possible to record and trace the good’s final destination of usage. As the European Union has abolished internal borders, records of intra­community movements of goods of non-­firm customers are no longer available. As a result, the purchase in country A and the transport to country B is a sale to a domestic customer. By all means one can recognize the new playing field for the fiscal administrations of the countries, as well as for the tax departments of delivering (and acquiring) companies to distinguish accurately between a taxable domestic delivery and a non-­taxable export delivery. For VAT purposes, the provision of services is employed, which is closely connected with the sale of goods. In this respect, the following question emerges: Under what circumstances can a customer be deemed a domestic customer? If a person is not a domestic customer, can the service made to this person be taxed? The same applies if a good is delivered from the company in country A to the final consumer in country B. If this delivery is seen as an export, then country A

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

142   K. Hirschler and M. Zagler is not allowed to tax it. However, does any taxation actually take place? On the one hand, it can be presumed that a taxation of the purchasing price in country B is similar to an import turnover tax. On the other hand, the buyer in country B could be obliged to withhold a part of the purchasing price, which then has to be transferred to country B’s tax authorities. The company in country A could also be charged with the responsibility of paying the import turnover tax. The tax liability to country B could also be fulfilled with the help of the legal obligation to keep records as long as there are administrative and execution cooperation treaties. Furthermore, the non-­firm purchaser in country B would be obliged to levy a withholding tax and transfer it to the other country’s fiscal authorities. This could solely be possible for the companies, as is actually the case in the context of withholding taxes. However, does it make a difference whether the final consumer in country B buys the good physically in country A and him- or herself delivers it to country A? Depending on the answer and the country in which the profit is taxable, there could arise new tax-­driven strategies to sell goods or provide services, especially by making contracts between the companies in a country that is not the home country of the customer, and especially when the corporate tax rate in country B is lower than that in country A (within an affiliated group, company A in country A sells goods to company B in country B, which sells the goods to customers in country A, who are taxed in country B). All these above-­mentioned questions, together with the issue of the different countries’ tax rates, play an important role in the practical implementation of a VADCIT. Though some questions regarding the determination of taxable income are eased with the help of a strong cash flow-­oriented tax base, it still makes a difference with respect to the state in which the performance is realized, as long as the tax rates are not harmonized. Accordingly, an accurate definition of the term “domestic customer” would be needed to avoid double taxation or double non-­taxation. 7.4.2  Allocation of taxable income The proposed VADCIT could lead to significant changes in the tax revenue of the countries involved. Taxable income is calculated by subtracting initial costs, including labor costs, from gross receipts derived from sales to domestic customers. In the case of an export, no taxes are levied in the exporting country of the goods. To ensure the fiscal consideration of paid expenses, a tax credit must be granted and paid to the taxpayer in the case of a negative tax base (see Figure 7.1 and Table 7.4). Within the VAT system, this would match the refund of an input tax balance (see Art. 21, para. 3 of the VAT Act; Ruppe 1994). As a consequence, countries with a high export rate may lose tax revenues. On the other hand, taxes are levied on the performance in the domestic customer’s home country. As a result of the import character of input-­related expenditures, no deduction is granted. Thus, the taxation of profits would economically result in a taxation of the sales proceeds devoid of domestic wages. Hence, countries with a high import rate might obtain a higher tax return.

The shift toward a VADCIT   143 P Ltd.

Debit Expenses (domestic) 30 Wages 10

Credit

Sales (export) 50

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Import 50 Wages 5

Credit

Sales 100

Profit 45

Profit 10 50

S Ltd.

Debit

50

100

100

Figure 7.1  Group’s cross-border transactions (1).

Notably, this issue becomes apparent if the profit margin is relatively low, because in such a case the company’s strain on liquidity proportional to the high tax burden could be crucial for the company’s existence from a cash flow perspective (see Figure 7.2 and Table 7.5). To avoid such a financial outcome, the export purchase price has to be reduced to prevent the importing company’s liquidity falling, with a risk of bankruptcy and corporate default, as the taxation of corporate profit could imperil the company’s liquidity. Irrespective of what up to now have been solely tax-­based thoughts, the proportion of income defined in financial accounting and the taxable income in each company may differ dramatically (book-­to-tax difference). In financial accounting, import or export has no effect on the treatment of the sale or purchase as revenue or expenses. This leads to a major difference concerning the term “income” in income tax law and accounting law. The corporate tax could become 100 percent of the economic profit, as shown in Figure 7.2 and Table 7.5, and might lead to a loss in the profit and loss account (Table 7.6). The corporate income tax as a personal tax, which at least should tax the earned income, would transform itself strongly into a transaction tax. Table 7.4  Tax bases of P Ltd and S Ltd Tax base P Ltd Sales (export) Expenses (domestic) Wages Taxable profit

   0 (not taxable) −30 −10 −40

Tax base S Ltd Sales (export) Import Wages Taxable profit

100 −0 (not deductible) −5 95

144   K. Hirschler and M. Zagler P Ltd.

Debit Expenses (domestic) 30 Wages 10

Credit

Import 50

Sales (export) 50

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Credit

Sales 60

Wages 5 Profit 5

Profit 10 50

S Ltd.

Debit

50

60

60

Figure 7.2  Group’s cross-border transactions (2).

7.4.3  Intra-­company transfer of assets Though the sale of goods between company A and company B, which are independent legal entities, is clarified within VADCIT, the question arises as to whether the same logic holds true for an intra-­company transfer of goods between permanent establishments in country A and country B. To satisfy the fiscal idea of the equal treatment of independent legal companies and permanent establishments, the answer to this question has to be affirmative. This indicates that an intra-­ company transfer is a tax-­free export in country A and is a non-­deductible expense Table 7.5  Tax bases of P Ltd and S Ltd (if the profit margin is low) Tax base P Ltd Sales (export) Expenses (domestic) Wages Taxable profit

0 (not taxable) −30 −10 −40

Tax base S Ltd Sales (export) Import Wages Taxable profit

60 −0 (not deductible) −5 55

Table 7.6  Profit and loss account of S Ltd Profit and loss account S Ltd (continuation of Figure 7.2) Sales (export) Import Wages Profit before corporate tax Corporate tax (20% of 55) Loss after tax

60 −50 −5 5 −11 −6

The shift toward a VADCIT   145 in country B. This makes it equivalent to VAT, which, at least in the European Union, is deemed an intra-community transfer – that is, either an intra-­community purchase or an intra-­community supply (see Article 1, para. 3 Z 1 of the VAT Act). Precisely, the exemptions regarding the treatment of intra-­company transfer as intra-­community purchase or supply show the difficulty in distinguishing it from the fiscally disregarded intra-­company service.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

7.5  Statistical effects The introduction of a VADCIT would have a series of significant effects. The first is merely statistical. If we consider the example from Figure 7.2 and assume a tax rate of 25 percent, then the host country of P Ltd should pay a negative tax of 10 to the company, as the profits are −40. In contrast, the country that hosts S Ltd would levy a tax of 13.75 on the calculated profits of 55. As actual profits are only 5, the company would make a loss. In the short run, the change of the tax regime might indeed lead to bankruptcies among import companies and the flourishing of exporters. As long as firms do not react, we might observe these transition effects, which are certainly significant. However, this situation would not be sustainable. Importers would force exporters to accept lower prices, and if they wished to sell anything, they would also comply. Certainly, export prices would drop by 10, which is the tax credit paid by the government of the exporting countries. This would set actual pre-­tax profits of P Ltd to zero. However, the company would still receive a negative tax of 10. On the other hand, S Ltd would witness its pre-­tax profits increase to 15 and would be taxed 13.75, leading to a meager net profit of 1.25. Thus, taxes are unequally distributed. In our case, P Ltd moves from a profit before taxation of 10 to a profit after taxation of 10, and remains untouched. However, S. Ltd witnesses its profits fall from 5 to 1.25, and faces an effective tax rate of 75 percent. Given that both firms are now making profits, there is no market mechanism to ensure another distribution of the tax burden. Only when the export price drops to 37.50 would the profits of the exporter drop by 25 percent to 7.50, and the profits of the importer drop by 25 percent to 3.75. Note that this would imply a fall in export prices of 25 percent. The introduction of this tax would give the exporter the power to fall short of this price cut, unless the importers had not made any profits. Changes in the export and import prices would have significant consequences for official statistics. We might observe dramatic changes among the indicators of competitiveness, such as the terms of trade and exchange rate indices. The introduction of a VADCIT would also have an impact on the reported import inflation. Because of the deep impact, we are skeptical as to whether all the effects could be accounted for in official statistics, and hence we might indeed find structural breaks in many economic time series, which could undermine our understanding of the economic phenomena and potential policy interventions, and this certainly would not be without a cost.

146   K. Hirschler and M. Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

7.6  Fiscal effects The fiscal motive of taxation is to raise revenues for government expenditures. With the current corporate income tax revenues bound to decline, a VADCIT may be an interesting substitute to secure revenue flows. The most significant change brought about by such a tax would be a shift of taxing rights from the source country to the destination country. At the firm level, this would imply that all exporting firms would receive huge input tax credits, whereas import companies would be faced with enormous tax bills. As opposed to the current system, the volume moved by tax authorities can be manifold, with significant problems. In particular, tax carousels, already known in Europe from VATs, would have another easy target. As with VAT carousels, the exporting country might have little interest in auditing its exporters with care, as it would not receive any VADCIT from these companies. This would cause serious audit issues. One solution would be to give the country of final sales the right to audit the entire corporation with all its national affiliates. With different languages and legal systems, this certainly would not be a feasible strategy. When sales were made in more than one country, firms would have a choice of their preferred auditor. Clearly, this would open doors to headquarter shopping, and tax competition would enter through the back door. For identical tax rates, countries that are net exporters of goods would, by and large, lose from a shift to such a tax, whereas countries that are net importers would gain. Hence, there is little chance that net exporting countries will agree to such a move, unless it allows them to set higher tax rates and thus increase their revenues. Another issue is how to treat direct sales to consumers abroad. According to the logic, the tax rate applied should be the tax right of the country of residence of the final consumer. This is already complicated in the case of internet sales with VAT, and might even be more problematic with a VADCIT, as the latter would differ between firms according to their profitability. It should be noted at this point that a VADCIT is a little different from a common consolidated corporate income tax base, where tax revenues are distributed solely on the basis of destination-­based revenues. The major difference is that the former is a cash flow tax, whereas the latter is not. The advantage of a cash flow tax is its investment neutrality, and the disadvantage is the high absolute flows of taxes between firms and the government. However, the main difference lies in the implementation. A CCCTB has the advantage that revenues can be allocated using other criteria as well. A VADCIT would help to solve the transfer price problem, but has a huge cost of implementation. Whether it is worth pursuing the risk depends entirely on the relative significance of the international intra-­firm trade, where transfer pricing issues prevail, and international inter-­firm trade, where the implementation costs of the VADCIT are the bigger problem.

The shift toward a VADCIT   147

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

7.7  Economic distortions The introduction of VADCIT is not neutral with respect to economic decisions in the case of different tax rates. In Figure 7.1, suppose that it is possible to transfer wage costs of 8 from the country of P Ltd to the country of S Ltd. This can be achieved either by means of creative accounting or, more simply, by moving part of the production. This will increase calculated profits from −40 to −32 for P Ltd, and will lead to an increase in taxation of 2. In contrast, the increase in wage costs in the country of S Ltd will reduce its profits from 45 to 37. Now, suppose that this country levies a tax of 50 percent. Thus, the tax advantage from moving production to the high-­tax country will be 4. The mere transfer of wage costs to the high-­tax country will lead to a net tax gain of 2, and will provide ample scope for tax planning. However, in cases when labor actually needs to be moved across borders, there would also be distortionary economic effects. It might not be socially desirable for workers to move just for tax purposes. By the same argument, firms might wish to transfer sales to low-­tax jurisdictions. The reason is that all the taxes are raised at the country of the final sale, because of the destination principle, whereas the wage costs are deducted from the tax of the source country. Apart from cross-­border transactions, this will not be easily feasible. Finally, if all countries worldwide did not introduce a VADCIT, then additional problems could arise. A country with a low tax rate outside the VADCIT system might manage to attract profits from inside the entire zone with transfer prices. Thus, the problem of transfer pricing would not be reduced but potentially aggravated.

7.8  Conclusions This study has discussed the introduction of VADCIT, a tax levied on the difference between domestic revenues and domestic costs for intermediate products and labor. The consequence of such a tax would be that all the profits of the firm in all its international subsidiaries would be taxed at the tax rate of the final products’ destination country at that country’s tax rate. VADCIT changes the corporate income tax from the source to the destination principle, and this eliminates the transfer pricing problem. We analyzed a business case to demonstrate the mechanisms of VADCIT, and were able to prove the indifference of the tax to different transfer prices. This case study demonstrated several problems with VADCIT. First, we found that gross tax flows are much larger than those under the current system, with net importing countries benefiting and net export countries losing tax revenues. Second, we observed that firms may consequently transfer production to high-­ tax countries and move sales to low-­tax jurisdictions. Third, the introduction of a VADCIT would distort prices. Finally, VADCIT is in conflict with current EU legislation, which makes it an unsatisfactory candidate for bringing about reform of the corporate income tax system.

8 The case for and against an EU tax

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Michael Lang and Martin Zagler

8.1  Introduction This chapter analyzes the advantages and disadvantages of several types of taxes that could be levied at the European level. The European Commission has suggested a modulated VAT, EU corporate income tax, energy tax, excise duties on tobacco and alcohol, personal income tax, and a climate charge on aviation (Cattoir 2004). The analysis attempts to identify which type of tax – if any – is most logical and desirable to introduce at the EU level. For every tax, we will examine its virtues and pitfalls from an economic and legal perspective, following a rigorous methodological approach. From an economic perspective, we argue that an EU tax is efficient, and a move from a national tax to European tax reduces the deadweight loss of taxation. From a legal and policy perspective, we examine the interaction of these taxes with other taxes, whether such taxes would be easy to manipulate, whether their introduction would hurt some Member States more than others, and whether they would be easy to collect and their levy be visible for the European citizen. We find that neither economic nor legal and policy arguments alone are sufficient to rank all proposed EU taxes. The combination of the two disciplines allows us to completely evaluate the alternatives and suggest possible candidates. 8.1.1  Legal and policy issues The introduction of an EU-­wide tax requires action by the Community legislator. From a legal perspective, it could therefore be interesting to know whether such legal measures under secondary law can be based on the EC Treaty. Changes in the EC Treaty in that respect are, of course, not impossible, and even the introduction of new secondary Community law requires, in most cases, a broad consensus among Member States. Although such questions are legally challenging, they are less relevant from a political point of view. Therefore, for all the potential taxes the question of whether there is a legal basis for their introduction under the existing primary law framework can be excluded. An issue of legal relevance, in the case of already-­existing taxes, might be their degree of harmonization. It is much easier to draft a concept for an EU tax if such a

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   149 tax already exists in all or most Member States and is harmonized under European law, or if other relevant factors are essentially similar in most Member States. Moreover, some taxes are closely interrelated either with other taxes or with other legal instruments, such as tax treaties. Therefore, it is necessary to analyze the conditions under which it seems realistic to isolate one of these taxes from other taxes and shift the competence to levy these taxes to the European level. From a legal perspective, the relevant question is how the other domestic taxes, or yet other domestic instruments, would interact with such a tax levied all over Europe. Finally, the possible difficulty of collecting the tax will also be taken into consideration. The possible number of taxpayers could be relevant, as well as the possible strategies available to taxpayers to avoid being covered by the tax or to avoid paying it. 8.1.2  Relevant criteria under the economic approach Primarily, taxes have a fiscal motive and serve the government in raising revenues to finance its expenditures (Nowotny and Zagler 2009: 245ff.). By reducing disposable income, taxes have an income effect on households and firms, and may reduce private spending. It has long been acknowledged that taxes can also have a distortionary effect on economic activity (Marshall 1890: 343ff.). Taxes change relative prices, and therefore lead to a typically unwanted substitution effect away from the taxed goods and toward untaxed goods.1 The negative impact of the substitution effect can be measured by the deadweight loss of taxation.2 The theory of optimal taxation postulates that a tax system should minimize this deadweight loss of taxation (Ramsey 1927; Mirrlees 1971), and this criterion has been used to evaluate costs and benefits of alternative tax systems. We will apply this methodology to evaluate the introduction of EU taxes. Two aspects are of particular importance in this context, and therefore render the analysis non-­trivial. First, in moving from the national to the European level, the degree of openness may change. A single Member State may be more open than the European Union as a whole with regard to the goods, services, labor and capital moving within and beyond the borders of the Union. However, only the latter will be reflected in the statistics of openness of the European Union as a whole, and is typically the smaller part. Second, market concentration changes. As markets may be concentrated in a single Member State owing to the presence of a national champion, this may be less likely for the Union as a whole. We will introduce the aspects of openness and market power in a simple model of taxation. Consider that in a given country, the inverse demand for product X (which can be a good, a service, a unit of labor, or a unit of capital) is given by p = f (X ) where p is the gross market price and we have a conventionally falling demand, f  ′(X ) < 0. A single firm supplies xi of aggregate supply, X = ∑xi. If the firm is

150   M. Lang and M. Zagler sufficiently large, then it may influence the price by increasing or reducing quantities. In this case, the firm has some market power. The firm will maximize profits π, which are given by the net of tax revenues minus costs,

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

π = (1 − τ)pxi − c(xi) where τ is the tax rate and c(xi) is a cost function that is increasing in output, mi = c′(xi) > 0. Marginal cost mi is the additional cost incurred for supplying an additional unit of output. Firms know about their potential influence on the market price, and may therefore maximize profits with respect to the quantity supplied, which is subject to the inverse demand function introduced earlier.3 This yields a conventional Amoroso–Robinson mark-­up pricing rule (Amoroso 1930): (1 − si/ε)(1 − τ)p = mi where si is the market share of the firm and ε is the price elasticity of demand.4 On the left-­hand side, we have marginal revenues or the additional revenue gained by the firm for supplying an additional unit of output. In the absence of market distortions, a firm may obtain the (gross) price, p, for an additional unit of output. With a tax distortion, the firm may only receive the net price, (1 − τ)p. Market power works similarly to a tax by further reducing marginal revenues. If a firm is very small, then its market share is negligible, i.e. si = 0, and there is no distortion from market power. A similar effect can be observed when the demand is very elastic, where ε goes to infinity. These effects are multiplicative, and hence an optimal tax policy has to consider market power. Figure 8.1 will make the argument clearer. The downward-­sloping solid line represents the demand function introduced earlier. The upward-­sloping solid line is the marginal cost curve, and represents the supply function in perfect competition, si = 0. The equilibrium market outcome under perfect competition and in the absence of taxation is represented by point A. The introduction of a tax would shift the supply curve upward to the upward-­sloping dashed line. The equilibrium is given by point B, which exhibits higher prices and lower quantities. We can measure the deadweight loss of taxation using the triangle ABC. If we have no taxation but market power, the quantity supplied would be identified by point C at the intersection of marginal cost and marginal revenue. The latter is given by the dashed downward-­sloping line, which by definition is steeper than the demand function. The equilibrium is given by Cournot’s point B, with higher prices and lower quantities as in perfect competition once again. It is a coincidence chosen only for the ease of exposition that these two equilibria are identical. Therefore, the deadweight loss of market power is also the triangle ABC. However, the combined effect of taxation and market power is not the sum of these two effects, or twice the triangle ABC. We can identify the quantities supplied under market imperfection with taxation at the intersection of marginal

The case for and against an EU tax   151 p mi /(1��)

mi �c (xi)

E �

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

B

D

A

İ C

F

(1�si /İ)

p �f (x) X

Figure 8.1  Monopoly taxation in a closed economy.

revenues (the downward-­sloping dashed line) and the supply function with taxation (the upward-­sloping dashed line), point D. The gross price can be read at point E. The total deadweight loss is more than twice the individual effect and is represented by the triangle AEF. Both market imperfections and taxes drive a wedge between the supply and demand function, which in isolation equals the distance BC, and roughly equals the tax rate, τ, or the market distortion, si  /ε. Note that the tax wedge is larger under perfect competition (BC) than under imperfect competition (DE), which can be determined by multiplying the two terms in parentheses in the Amoroso– Robinson rule, τ + si  /ε − τsi  /ε. Taxes are therefore less distortionary in an environment of imperfect competition, as shown by the negative last term, and it is thus socially less expensive to raise taxes under these circumstances. We can use this finding to identify our first evaluation criterion: ECO1: If moving taxing rights to the European Union reduces (increases) the market power in segmented markets, then the European Union would set a lower (higher) tax rate than individual Member States.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

152   M. Lang and M. Zagler We have not yet introduced openness, which is particularly important in segmented national markets. If large economies within Europe have more competitive markets and the European Union chooses to set an optimal average tax rate, then small countries would see their tax rates decline, whereas large, competitive economies would see their tax rates increase. This would lead to the redistribution of the costs of taxation (measured by the deadweight loss) toward large Member States. In the next step, we open the economy and allow imports and exports of goods, services, labor and capital by assuming that foreign supply is completely elastic at some given world market price, W. Domestic firms can supply goods, services, labor and capital to world markets at this price, but may not be able to supply domestically or internationally at a price above the world market price. Figure 8.2, which is initiated from where we have left earlier, depicts this situation starting from balanced trade, in the absence of market power. Initially, domestic supply and demand coincide at the world market price, W, at point A. Domestic firms provide exactly the amount of goods requested by domestic costumers, and there are neither exports nor imports. The introduction p

mi /(1��) mi �c (xi) K

E

G

B

D H A

W

C F J

(1�si /İ)p

p �f (x) X

Figure 8.2  Monopoly taxation in an open economy.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   153 of a source-­based tax increases the costs of domestic producers and shifts the domestic supply function upwards. It will therefore intersect earlier with foreign supply, W, at point H. From this point onwards, the total supply in the domestic economy is horizontal and coincides with the world supply, W. While there is still a demand from domestic consumers at point A, domestic suppliers will only supply at point H. The difference, AH, will be covered by imports. The deadweight loss of the source-­based tax equals AHJ. This loss is fully borne by producers, as consumers are fully compensated by world markets. The introduction of a destination-­based tax, in contrast, shifts both the domestic and the foreign supply curves upwards, as the tax is paid not only by domestic firms but also by foreign firms supplying the domestic market. In this case, all domestic demand will be supplied by domestic firms at point B. Domestic firms continue to be cheaper than foreign firms until point K. The difference in the quantities between the domestic supply B and the total supply of domestic firms K will be sold abroad, and will be considered as exports. The deadweight loss of the destination tax equals the triangle ABC, and it is not obvious whether it is bigger or smaller than the deadweight loss of the source tax. Despite the fact that a destination-­based tax boosts exports, it is not sufficient to identify which type of tax is socially preferable. Introduction of an EU-­wide tax may shift both the domestic and the world supply upwards, irrespective of whether it is a source- or a destination-­based tax. As long as the European Union levies a tax below domestic tax rates and allows Member States to set individual tax rates above EU rates, moving taxing rights may not change anything. This leads to the second criterion: ECO2: In open markets under perfect competition and for a destination (source)-based tax, transferring taxing rights to the European Union does not have any (has) economic consequences. Therefore, destination-­based taxes are the prime candidates for EU-­wide taxes in competitive markets. As a last step, we now reintroduce market imperfections in the open markets, which are achieved by introducing the marginal revenue curve again. In the absence of taxation, Cournot’s point would be (by coincidence) at point C. However, firms cannot set supply at point B, which would be optimal in segmented markets, as they are underbid by foreign competition. Thus, their best choice is to supply at point A, without market imperfections. Indeed, opening to trade is known to be detrimental to local market power, and is one of the predominant reasons for the introduction of the European common market. The same argument holds in the case of a source-­based tax. In the case of a destination-­based tax, the optimal choice of quantity for the domestic firm would be at point D, with supply set at point E. This is below the world market price and is therefore feasible. Domestic demand will be fully met by domestic firms at world market prices. The combined deadweight loss of

154   M. Lang and M. Zagler taxation and market power equals the triangle ADF. Foreign firms will not enter the domestic market, and domestic firms will not supply to the world markets. Thus, the introduction of a destination-­based tax protects domestic firms with market power, similarly to import duties. This leads us to the third criterion of taxation:

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

ECO3: In open markets under imperfect competition and for a destination (source)-based tax, transferring taxing rights to the European Union will protect (destroy) the market power of domestic firms. It is therefore worthwhile to pursue a source-­based EU tax in open markets with market power. Table 8.1 summarizes these results.

8.2  Corporate income tax (CCCTB) 8.2.1  Description Developing an EU corporate income tax (EUCIT) would require first a definition of a common (consolidated) tax base. This base would be applied in a compulsory manner to all companies liable to corporation tax or to a precisely defined group of companies subject to this tax, for example those listed on a stock exchange, multinational companies with a turnover above a given threshold, or European companies. For the companies concerned, the EU tax would have to replace the existing national corporate taxes. The applicable tax rate would need to be defined at the EU level. 8.2.2  Legal and policy issues Corporate tax law has not yet been harmonized. However, this idea could be seen together with the project of a CCCTB (European Commission 2007). Although the introduction of a CCCTB does not seem to be within reach right now, the project is still on the agenda of the Commission. There are speculations that this idea may be pushed forward again now that the new Commission is installed and the constitutional issues of the Treaty of Lisbon are settled. However, even optimists do not see a consensus of all 27 Member States within reach. The CCCTB is viewed as a candidate for enhanced cooperation between Table 8.1  Economic evaluation criteria Market power

Segmented markets:

Yes No

Yes

No

Leave tax domestic Source tax to EU

Move tax to EU Destination tax to EU

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   155 Member States, which requires the commitment of nine countries now that the Treaty of Lisbon has come into effect (Kubik and Massoner 2009). Many Member States consider the harmonization of corporate taxes to be a sensitive issue. Even if the harmonization covers just the tax base and not the tax rates, which is the idea under the CCCTB approach, some governments fear that such a move would be a door-­opener for further harmonization. Therefore, they oppose such ideas from the beginning. Many governments are careful to maintain sufficient room to implement their tax policy. They want to be free to lower the corporate tax burden to make their country a more attractive business location. However, one cannot ignore the fact that those countries that have benefited from tax competition so far have come under serious pressure recently. More countries, and bigger countries, are no longer willing to accept that they are becoming vulnerable as a business location owing to tax incentives granted by smaller countries. However, so far only smaller countries have been forced to give up favorable tax regimes. Countries like the United Kingdom have been successful in protecting the tax havens that are under their influence, such as Jersey and Guernsey, from similar strong pressure. As the United Kingdom is still among those countries that are strongly against corporate income tax harmonization, it seems unlikely that the United Kingdom, for one, would abandon its resistance toward a CCCTB. Furthermore, one may doubt whether further harmonization is realistic. However, representatives of the Commission and the Member States have already devoted much time and work to the CCCTB project (Aujean 2008). Different rules in different Member States have been compared and possible compromises for harmonized rules have also been explored. If the Commission proceeds with the CCCTB project, it needs to present a draft for a harmonized tax base. Such procedure could also be used for a EUCIT. Thus, an attempt to introduce the EUCIT would benefit from the efforts made in connection with a CCCTB. Evidently, the introduction of a EUCIT would be a much more ambitious project than the introduction of a CCCTB. First, a CCCTB might only provide for a harmonized tax base, leaving it within the competence of Member States to determine the tax rate. By contrast, a EUCIT might require a harmonized tax rate, at least insofar as such a tax generates revenues for Europe. Furthermore, the Commission still favors the idea that a CCCTB should be optional for the taxpayers (European Commission 2007; Hey 2008). The taxpayers should decide whether they want to be taxed under their domestic tax system or under the CCCTB rules. The Commission presumes that the CCCTB rules would become very attractive owing to low compliance costs, and that most taxpayers who are active in cross-­border business might opt for the CCCTB. However, for a EUCIT such a system would not be feasible. A EUCIT would work only if the application of harmonized rules were mandatory for all or certain corporations. However, the tax authorities of some countries have already raised concerns about whether optionality is acceptable. They fear that the CCCTB might be viewed as an additional tax planning tool, and that compliance

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

156   M. Lang and M. Zagler costs might be even increased because companies would prepare calculations regarding whether they were better off under their domestic tax base or under the CCCTB rules (Hey 2008). If a CCCTB becomes mandatory, the Community legislator would have to face challenges similar to those that would apply in the case of a EUCIT. Unless all corporations were covered under the EUCIT, the scope of such a tax could not be determined. However, if only certain types of corporations were covered, then the determination of the scope might also lead to additional tax planning opportunities. Taxpayers could maneuver around the tax and integrate all their activities in one corporation or split them up between different corporations, depending on whichever was more beneficial to them. Certainly, one could also consider applying the EUCIT rules on all kinds of corporations. In that case, a EUCIT could be levied as either an alternative or an addition to a domestic corporate tax. If the existing corporate income taxes were replaced with a EUCIT, then the effect on domestic tax policy in Member States would be dramatic. Currently, most of the tax policy deliberations in Member States are concerned with the tax base and the tax rate of the corporate income tax. If Member States need to give up their right to levy a domestic corporate income tax, then such a move would touch the core of tax policy. However, the EUCIT might be levied in addition to existing corporate income taxes. Evidently the compliance costs would increase dramatically if every corporation had to apply two completely different rules of corporate income tax. Such a reform would require either mandatory or voluntary harmonization of the corporate income tax base. Again, the room for domestic tax policy would be diminished dramatically. Even if a EUCIT were made mandatory for all corporations, taxpayers would still have the opportunity to change their legal form. As long as there are other entities existing which are treated as transparent and whose income is therefore taxed at the hands of members or shareholders, taxpayers will always be able to switch from one legal form to another. It would be quite rational to calculate whether it would be more beneficial to be taxed under the EUCIT, under partnership or as a sole entrepreneur under the domestic individual income tax. Thus, a EUCIT might not make sense unless it was accompanied by a general business tax that was applicable to sole entrepreneurs and partnerships as well, and that led to the same tax burden. However, such a reform might lead to an even more dramatic change in the tax system. In addition, one has to take into account that corporate income tax and individual income tax cannot be viewed separately from each other. Many domestic tax systems may want to achieve neutrality insofar as the tax burden of an individual taxpayer who directly invests in certain activities and that of taxpayers who invest through a corporation remain similar. Undoubtedly a Europe-­wide harmonized business tax could solve some of these problems. However, quite often neutrality is not achieved merely with respect to business income. In some countries, it is viewed as a question of fairness that the tax burden of employees, business people and a person receiving rental income or other private income

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   157 should correspond to each other. This would be difficult to achieve if only the corporate income tax or the taxation of businesses were harmonized. Obviously, countries might apply an imputation system and treat the corporate income tax as a prepayment of their (domestically levied) individual income tax. However, such imputation systems have themselves been under pressure by European law, as they do not function properly when dividends are distributed across borders (ECJ C-­319/02; ECJ C-­292/04). Furthermore, if neutrality at the level of the individual taxpayer requires that corporate income tax that cannot be credited, owing to the lack of sufficient income or the existence of losses at the level of the shareholder, has to be reimbursed, then Member States would have to reimburse the European tax. Thus, depending on the individual situation, Member States would have to subsidize European taxes in a not very transparent way. Hence, it does not seem realistic to harmonize a EUCIT without harmonizing the individual income taxes. Although it is not explicitly stated, one may assume that a EUCIT would be collected by the tax administrations of Member States and that legal protection would be granted under the laws of that Member State, even if there were to be a minimum level of harmonization in procedure and, maybe, special European remedies. If for these reasons such a tax could be considered as being levied by Member States, even if they collected the tax for the European institution, then one might argue that the existing tax treaties are applicable to them (Traversa 2008; Baker and Mitroyanni 2008). If the existing tax treaties are not applicable, serious additional problems have to be solved. It would depend on whether the EUCIT were levied only on a territorial basis. Therefore, EU rules would have to be developed, granting “unilateral” relief from double taxation in relation to third countries. Either third-­country income could be exempt or a credit could be granted for third countries’ taxes levied on third-­country income. In addition to the concerns just raised, one has to take into account the fact that corporate income tax cannot be seen as a stable source of revenues. Companies that have made sufficient losses might not pay European taxes at all, even though they might benefit from the common market. Furthermore, during difficult economic periods most companies will suffer, and thus such a tax would generate very little revenue, although during such periods revenues might be extremely important for the EU Commission, to enable it could react to policy needs. 8.2.3  Economic issues It is common wisdom in economics that pure profits (everything that exceeds the normal return to equity capital) can be taxed without distortionary effects (Nowotny and Zagler 2009: 299). However, a tax on corporate income is not a tax on pure profit; it also taxes returns to equity, as firms cannot rely on external finance alone. Therefore, a corporate income tax taxes both pure profits and the return on capital. A corporate income tax is a tax on the rental price of the capital.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

158   M. Lang and M. Zagler Therefore, the relevant market for investigation of the effects of a transfer of corporate income taxing rights to the European Union is the market for loanable funds, where the capital can be rented. A thorough analysis of financial markets in Europe is certainly beyond the scope of this chapter,5 and is also not necessary for the sake of our analysis. To analyze the advantages and disadvantages of transferring taxing rights to the European Union, we need to take a closer look at the segmentation of markets and market power. For this purpose, it makes sense to distinguish between small and medium-­sized enterprises, on the one hand, and large firms and corporations on the other. Large firms certainly have access to European equity capital markets. There is no evidence that such firms face problems in acquiring funds on these markets. There is a home bias in investor behavior (Lewis 1999), which may lead to a certain degree of market segmentation. However, owing to high elasticities, this is not reflected in the prices, so that firms could pay a lower dividend to a domestic asset holder than to a foreign asset holder. Therefore, we can conclude that market segmentation is not an issue of concern for large enterprises. Furthermore, given that the European financial markets are highly liquid, there is no reason to assume that either party has any substantial market power. Thus, by observing the upper left corner of Figure 8.1 we can conclude that corporate income taxes should be left at the domestic level. If the EU were allowed to set national tax rates, then it might not choose tax rates different from those set by the individual Member States. In the more likely case that the EU needed to select a single tax rate, it would be an average of the national tax rates, which would be sub-­optimally high for some countries and sub-­optimally low for others. We have been assuming that EU Member States can set their corporate income tax rates optimally (Zodrow and Mieskowski 1986). However, we are also aware that due to a fiscal externality, countries engage in tax competition. This is a strong argument in favor of coordination of corporate income taxes. It does not necessarily imply that taxing should be transferred to the EU level. Nevertheless, things look very different for small and medium-­sized enterprises. Here, financial markets are much less accessible for equity finance. In many cases, these firms rely on single-­family ownership with debt finance from the local banking sector (Cressy and Olofsson 1997). There is very little evidence that small enterprise have access to the European financial markets, despite the efforts by the European Union to reduce the problem. Thus, we certainly find segmented markets with single firm owners typically being the sole provider of equity, with a high degree of market power (the lower right cell in Table 8.1). In this case, it would be desirable to move a destination-­based tax to the EU level. Despite recent debates,6 but thoroughly in line with the OECD model tax convention, corporate income tax is a source-­based tax, and this will very likely change. Therefore, we can also rule out a transfer of taxing rights to the EU level for small and medium-­sized enterprises.

The case for and against an EU tax   159

8.3  Personal income tax 8.3.1  Description In the Commission’s working paper, three main options for a personal income tax are discussed:

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

•

•

•

The first consists of setting a per capita tax on all EU citizens. The annual amount would be about €260 per person in the EU-­15 (estimate based on the EU budget in 2003). It would have the advantage of being extremely visible, simple and efficient. On the other hand, it would probably be unacceptable to many people for equity reasons. However, bering in mind unsuccessful past experiences, such as that of the “poll tax” in the United Kingdom, this option is not further discussed. A second option consists of setting a surcharge on the Member States’ personal income tax. This surcharge would be a percentage of the national tax. This could have the advantage of leaving the progressiveness of the personal income tax unchanged in the various Member States. However, as Member States’ systems are quite different, it is likely that simply setting an equal surcharge across Member States would deliver quite inequitable results – that is, with regard to horizontal equity. Hence, the surcharge would have to be fine-­tuned to take national differences into account. In a concrete proposal based on work by Biehl (1985, 1990, 1992), a two-­stage procedure is employed to derive a progressive surcharge on income tax. In the first stage, an overall tax burden is determined for each Member State, based on its income per capita or other relevant variables. A political decision would have to be taken on the degree of progressiveness of the national contributions to the EU budget. In the second stage, the overall tax burden of each Member State is transformed into a uniform percentage surcharge on national personal income tax payments. The surcharge could be shown on each tax declaration (and each national tax invoice), so that each taxpayer would know his or her contribution toward the financing of EU expenditure. A third option would be to create a separate EU personal income tax, with a specific progressiveness, rebates, etc. Citizens would then have to fill in two tax returns, one for the Member State and one for the European Union. This situation would be similar to that in Quebec (Canada), where citizens pay separate taxes to the province and to the federal government. However, the administrative and compliance costs in this case would be significant. In this scenario, the tax base and the tax rates would be determined at the EU level. This tax would be as simple as possible. This would presumably imply a broad base, in combination with low rates and no exemptions (other than a basic personal allowance). Thus, there would be no need to consider other aims of national systems, of a kind that usually result in complex income tax systems. In view of the otherwise enormous administrative (and political) problems, a harmonized EU income tax would need to be managed by

160   M. Lang and M. Zagler the tax authorities of Member States. The tax law would be introduced via a Council regulation for the core elements, and via directives for the administrative elements.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

8.3.2  Legal and policy issues As is correctly pointed out by the Commission paper, personal taxes constitute one of the most direct and visible links between taxpayers and citizens on the one hand, and elected authorities on the other hand: “The attraction of an EU personal income tax partly rests on the opportunity to exploit this direct link in order to enhance accountability.” Obviously, this effect largely depends on the extent to which a taxpayer has the impression that he or she is “suffering” from such a tax. As personal income tax law does not always require taxpayers to file tax returns and have their income assessed and levied subsequently, a tax that is often levied as a withholding tax, many taxpayers do not realize how much income tax they pay. Although theoretically it would be possible to have all taxpayers assessed on their individual income tax, the administrative burden would be enormous. The idea of an EU-­wide personal income tax is not completely new, and there is already a prototype: employees of the European Commission and other European institutions are already covered under a special income tax with the employment income that they receive from these institutions (Endfellner and Dornhofer 2005). This tax derives from the fact that this income must not be taxed under domestic tax systems, either in the country where the activities are carried out or in the country where the employee resides (or used to reside before starting their employment). Although this tax could be considered a nucleus for a Europe-­wide personal income tax, the number of taxpayers who are covered under these rules is relatively small, and this tax is not a comprehensive type of personal income tax but is limited to the remuneration that these employees derive from the European institutions. However, one may also doubt whether a personal income tax is the perfect candidate for the EU-­wide tax, as the revenues raised by Member States are far from stable. The times when personal income taxes were considered to be “queens” of taxation are over. On the other hand, the fact that personal income taxes have lost importance among Member States in recent years and decades could make it easier for them to refrain completely from levying such a tax and to allow an EU tax to be levied. However, income tax is not always regarded as a mere source of revenues; it also acts as an instrument for the redistribution of wealth. Ideological and cultural approaches are very different throughout the Community. In some countries, a progressive tax rate is seen as a constitutional requirement, or at least as a necessity of a fair tax system. Other countries have introduced a flat tax rate. In those countries, concerns related to fairness, equality and ability to pay are not raised at all, are not considered to be contradicted by the introduction of a flat tax rate, or are not seen as relevant by the leading political forces.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   161 Furthermore, countries that favor progressive rates differ in their approaches. In some countries, it is crucial that the progressive tax rates be applied on the complete amount of income, whereas others accept that certain items of income are excluded, and these items fall under a special tax rate.7 All these differences illustrate that it would be quite burdensome to reach a political agreement regarding whether a personal income tax levied all over Europe should be based on a flat tax rate or on a progressive tax rate, and, in the latter case, how progressive such a rate should be. The issue is even more complicated in that personal income taxation and social security contributions are interrelated to a certain extent, but treated differently among Member States. In some Member States, social security contributions are levied completely separately from income taxes and quite often capped at a certain amount, thus having regressive effects, whereas in other Member States the social security system is financed by tax revenues, and no specific social security contributions are levied (European Commission 2008b). Even in Member States where separate social security contributions are levied, the social security system is, to a certain (sometimes large) extent, financed by tax revenues. These interdependencies and differences would make it even more difficult to replace domestic personal income tax systems by a personal income tax levied at the European level. It has already been mentioned that corporate income tax and personal income tax are closely related to each other. In the case of dividend distributions, countries that are concerned with avoiding double taxation either grant a credit for the underlying corporate income tax or reduce the personal income tax burden for dividends significantly as compared with the remaining personal income tax base. For a tax levied all over Europe, such as a harmonized personal income tax, it would be difficult, if not impossible, to adjust the rules so that they would be able to react flexibly to the different corporate income tax systems unless these rules were also harmonized. With the exception of the interdependence of a personal income tax with the social security system on the one hand, and corporate income tax on the other hand, all the problems mentioned earlier are relevant for both a personal income tax that is levied as an alternative and one that is levied as an addition to domestic personal income taxes at the European level. However, one should not assume that if such a tax were levied in addition to the domestic taxes, then domestic personal income taxes would remain completely unchanged. At least with regard to the tax burden, it would not be politically acceptable if the European tax burden came on top of the domestic burden. Member States would have to reduce domestic tax rates, which again raises concerns about the impact of these revenue reductions on the social security system. Another consequence of having two types of personal income levied in parallel to each other is the increase in compliance costs. Such a system would be politically acceptable only if the rules about the tax bases of the two types of taxes were largely or even completely harmonized. Even if such a harmonization is not seen as a requirement for the introduction of an EU-­wide personal income tax, it would still have

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

162   M. Lang and M. Zagler to take place in the long run. Furthermore, the political pressure toward such harmonization would be tremendous. From that point of view, the option of introducing a European surcharge on top of domestic personal income tax is not very different, and again might require harmonization of the tax base. A surcharge would require that the tax base be identical throughout the European Union. Therefore, all the different versions of a Europe-­wide personal income tax discussed so far would involve the difficult task of harmonizing the tax base. However, the type of surcharge described by the Commission paper is different. Obviously, the idea is to determine a certain tax burden up front and to convert this overall tax burden into a percentage of the personal income tax of the individual taxpayer and “flag” that part of the domestic personal income tax as European personal income tax, and transfer it to the European level. However, one might question whether such a “surcharge” would be a real European tax, as such a system would not be very different from how the European Union’s own resources are calculated today. Such a system comes close to a tax levied by the local government, a certain percentage of the revenues from which have to be transferred to the central government. Although it has not been explicitly stated, one may assume that not only such a surcharge, but also all other types of personal income tax, would be collected by the tax administrations of Member States and that legal protection would be granted under the laws of that Member State, even if there might be a minimum level of harmonization in the procedure, and some special European remedies. If, for these reasons, such taxes could be considered as being levied on behalf of Member States, even if they were EU taxes, one might argue that the existing tax treaties would be applicable.8 If the existing tax treaties are not applicable, then serious additional problems have to be solved. Personal income taxes are usually levied on a worldwide basis. Although one solution would be to cover only sources of income within the European Union, this would create an incentive to relocate business and other activities outside the Union. Therefore, EU rules granting “unilateral” relief from double taxation in relation to third countries would have to be developed. Either third-­country income could be exempt or a credit could be granted for third countries’ taxes levied on third-­country income. Within the European Union, it should not really matter where the EU tax was levied, as the revenues would have to be transferred to the European institutions anyway. However, the question of which country’s tax authority would be competent to collect the tax is not completely irrelevant. As long as there is also a domestic personal income tax, the collection of the domestic and the EU personal income tax, for reasons of practicability, should go hand in hand and the tax authority of a given Member State should be responsible for the same portion of the tax base. In other words, rules similar to the ones in the tax treaties should be introduced for dividing the competence. However, not all Member States have concluded tax treaties with each other. Furthermore, the treaties that have been concluded are not identical, and even if they are completely in line with the

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   163 OECD Model, under some tax treaties the exemption method is applicable and under others the credit method applies. Therefore, if one prefers to develop general rules to provide relief from double taxation within the European Union, one has to accept that these rules are not completely in line with the tax treaties concluded between Member States. The alternative is to copy, if possible, the tax treaty rules on the allocation of taxing rights and, as a consequence, to accept that the allocation of the tax base for the collection of the income tax would follow very different rules. Revenues generated by a personal income tax certainly depend on the economic situation as well. However, they are relatively stable. For example, wage taxes are levied in more economically difficult situations even if the employer suffers a loss, and thus does not deliver any financial means. 8.3.3  Economic issues A personal income tax is a direct tax on production factors. It includes taxation of labor income and taxation of profits but excludes taxes on corporations, which fall under corporate income taxes, discussed earlier. For the latter, therefore, we can carry over the analysis in the previous chapter, which indicated that an EU tax would be a bad idea. Despite many efforts by the European Union, labor markets remain highly segmented in Europe. This may be due to language barriers, high reallocation costs or other factors.9 Labor markets in Europe are certainly not competitive. With trade unions on the one side and employers’ associations on the other, continental European labor markets certainly fail one of the fundamental characteristics of competitive buyers, where many buyers face many sellers. Labor is also a very heterogeneous factor of production, which leads to search frictions. These search frictions are another source of market imperfections (Layard et al. 2005). Labor markets are segmented and imperfect (the upper left cell of Table 8.1). Proposition ECO 1 states that in this case, it is preferable to leave the tax domestic, as having differentiated taxes on labor leads to a lower tax burden. From an economic perspective, personal income taxes on both dependent and self-­dependent labor favor national taxation, instead of an EU tax.

8.4  Modulated VAT 8.4.1  Description The European Parliament, in the Langes Report (European Parliament 1994a), advocated a modulated VAT. The Langes Report stated that “a proportion of a largely harmonized VAT, imposed on the basis of tax declarations and clearly denoted on each individual invoice as EU taxation would at the present be the most convincing form of own revenue”. The amount of the EU tax would be clearly differentiated from the amount levied for the national VAT on the

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

164   M. Lang and M. Zagler invoices. Both the national parliaments and the European Union would be granted the power to separately determine which rate would be imposed for purposes of the national budget and which for the EU budget. There would be a combined VAT rate consisting of the national and the EU rates. For the EU rate, a figure of 2 percent has been suggested as a starting point, but the figure could subsequently be increased if the commitments of the European Union were extended. To contribute to equity across individuals within Member States, the report also proposed that there would be two VAT rates, for example 1.5 percent for basic necessities and 3 percent for other goods and services. The total combined tax rate should not increase following the introduction of EU rates, as the “national” tax rates could be correspondingly decreased by Member States. This would be possible because national (VAT and GNI) contributions to the EU budget would be significantly reduced, or even eliminated. 8.4.2  Legal and policy issues VAT is largely harmonized within the European Union. Community legislation in this area almost entirely determines the tax base, and Member States are only competent to determine the tax rate. However, even in that respect Community law does not give complete leeway to Member States. As far as certain ranges and the number of reduced tax rates are concerned, Community law also provides guidance. From that perspective, VAT is an ideal candidate for a tax whose revenue could partly go directly to the European Union. However, one must not overlook that the tax base is completely harmonized, as there are still some rules in place that permit Member States to deviate from the general rules, and the Community legislators need to work hard to harmonize these rules. Nevertheless, when compared with other candidates for an EU tax, the degree of harmonization is extremely high. Therefore, it is more realistic to achieve a further degree of harmonization in the area of VAT than for other taxes where there is much less, or even no, harmonization. Furthermore, VAT directives also contain some administrative rules.10 One may already be aware of the interaction between the domestic procedural rules and the Community rules in that area. To date, the fact that the systems of legal protection vary between the different Member States has not constituted an obstacle. The administration of the VAT rules is largely outsourced to the business sector, which might be considered an advantage for the governments at both the domestic and the European levels, as the compliance costs are rather high at their side. On the other hand, business communities complain about the high burden that the levy of VAT represents. However, if an additional surcharge to VAT were levied on behalf of the European Union and was directly or via the national tax administration transferred to the European Commission, then the compliance costs for the business community would not necessarily increase, when compared with today’s compliance costs. Recently, tax evasion in the area of VAT has become a serious concern for many tax authorities. Member States are exploring ways to make the VAT

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   165 system immune to manipulation by the taxpayers. Therefore, one may raise the question of whether it would be wise to introduce a system at the EU level which is open to manipulation by taxpayers and could subsequently endanger the supply of stable revenues to the European Commission. Nevertheless, the VAT system will have to be changed to prevent manipulation. Otherwise, Member States will become too vulnerable. The concept of a modulated VAT will automatically benefit from all these efforts (European Commission 2009a). The fact that at least parts of the VAT revenues would still contribute to the budget of Member States can also be seen as an advantage from another perspective. A tax whose revenues were completely transferred to the European level might not receive the complete attention of the tax authorities of Member States, as they might be more interested in levying their domestic taxes properly. They might be tempted to apply the rules too generously, because their own revenues are not at stake. However, if at least a certain part of the revenues were kept locally, then domestic authorities might have an incentive to administer the levy of such a tax as carefully as they do in other cases. Current VAT rules prevent Member States from levying additional domestic VAT. Therefore, besides the already-­existing VAT, which is harmonized under the directives, no specific domestic VAT exists. Therefore, Member States would not have to abolish an existing tax in the event that such a modulated VAT were introduced at the Community level. However, a surcharge on the existing VAT rates would still have an impact on the tax systems of Member States if the tax rates for the regular VAT were reduced. As has been suggested by the Commission paper, to keep the overall VAT burden of the taxpayers at the same level, Member States might lose revenues for their domestic budget. As nowadays VAT is one of the most relevant and stable sources of revenue, such changes would have a serious impact on the budget of Member States. Member States would either have to shift competences to the European level to introduce new taxes, or have to increase the rates of other taxes that are currently applied. However, if a modulated VAT were levied in addition to the existing VAT rates, the overall VAT burden in Europe would increase, and one cannot ignore the fact that the ratio between the VAT and the income tax burden in a country usually reflects certain – albeit very subjective – ideas of social justice and fairness. Thus, the introduction of a surcharge on current VAT rates could affect the delicate balance between the two taxes in favor of VAT, and therefore lead to a difficult ideological debate within Member States, which could endanger the realization of such a project. 8.4.3  Economic issues A VAT is a tax on the final consumption of goods and services. These are very heterogeneous products, and consequently are very different markets. While consumer services are hardly tradable, and thus reside in the local market segments, other consumer products are certainly more tradable; however, the trade is predominantly intra-­firm cross-­border trade. Nevertheless, as the tax is only

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

166   M. Lang and M. Zagler charged between the firm and the consumer at the final sale, this aspect is irrelevant. Evidence of cross-­border shopping showed very little activity; if it occurred at all, it was typically only relevant for border regions less than 50 kilometers from a national border (Cnossen 2001). However, with regard to the market power the picture is extremely heterogeneous. There are many sectors in which a very small number of firms dominate the market, whereas in other areas a vast number of suppliers are confronted with a typically large number of potential buyers. Sectors that are less competitive include the automobile industry, certain areas of consumer electronics, energy and transportation. By comparing this analysis with the data presented in Table 8.1 (the upper right cell), we can observe that a move of taxing rights to the European Union is clearly favored, because moving taxing rights to the Union would allow the Union to set a different tax rate that would offset some of the negative effects of market segmentation. It may be the case that consumers do not shop across borders because prices are equalized anyway. While casual empirical evidence is against this proposition, the idea that markets for the consumption products are contestable (Baumol et al. 1982) and therefore less segmented than is observed cannot be immediately ruled out. In this case, we would be in the lower right cell of Table 8.1, which would also lead to the conclusion that VAT, which is a destination-­based tax, should be transferred to the European Union. The only argument against an EU VAT is: what if the markets are not competitive? In this case, we would be in the upper left cell, and could conclude that taxes should remain local, as there are advantages in setting different tax rates. As was mentioned earlier, this may be the case for some products, but certainly not all. The products most likely to be affected are from the automobile industry, certain areas of consumer electronics, energy and transportation. For most of these sectors, special excise taxes are in place and will be discussed in the following paragraphs. The only exceptional case may be consumer electronics, for which only very few EU member countries have a special excise tax, Italy’s cell phone tax being the prime example. In this case, differentiation of tax rates could be achieved by setting the special excise taxes accordingly, so that nothing would be lost if VAT taxing rights were transferred to the EU level. In summary, it makes very good sense to transfer VAT to the European Union, given that harmonization has already been promoted for a very long time, and that both legal and economic arguments favor such a measure.

8.5  Excise duties on tobacco and alcohol 8.5.1  Description Excise duties on tobacco and alcohol are taxes on the consumption of alcohol and tobacco and are widely used across Europe, albeit with significant differences in rates.

The case for and against an EU tax   167

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

8.5.2  Legal and policy issues The idea of using excise duties on these goods as an EU tax has been examined in a European Parliament report (1994b) and in Agenda 2000 (European Commission 1997). Tobacco and alcohol excise duties are already covered by directives, although the degree of harmonization is not very high. In particular, the determination of tax rates is still largely within the competence of Member States. Even minimum tax rates are not yet agreed in all the areas. However, experience has taught us that politically it seems to be easier to develop areas further where certain harmonization measures are already in place than to agree upon harmonization in areas so far not harmonized. It took ECOFIN many years of negotiation to pass the “tax package” in 1990, which included the parent subsidiary directive and the merger directive. However, after a certain degree of harmonization had been achieved it was possible to agree upon further changes, including an extension of the scope of the directive. This is even more evident in the area of VAT, where amendments are agreed upon more frequently. From that perspective, it seems to be a smaller step to harmonize these taxes completely, even at the level of tax rates. The Commission has recently taken some initiatives, partly successfully, to develop harmonization further. This included an update of the definitions of different types of tobacco products to remove loopholes that allow certain cigarettes or fine-­cut tobacco to be presented as cigars, cigarillos or pipe tobacco, which benefit from a lower tax rate. As differences on tobacco taxation levels within Member States are being narrowed down, taxation rules will also become more transparent, thereby creating a level playing field for manufacturers and giving Member States the flexibility to set minimum taxes. Furthermore, the amount of smuggling and fraud, which is a serious concern in that area of taxation, can also be reduced. However, the fact should not be underestimated that for Member States it makes a difference whether the revenues of such taxes are theirs or are assigned to the European Union. In that regard, the tax revenue could be shared between Member States and the Union. Under a harmonized tax base, the tax could be levied at a certain tax rate on behalf of the European Union. In addition, Member States could decide whether they also want to levy such a tax for the purpose of increasing their own tax revenues. Such a system would enable Member States to maintain their policies, and, at least as far as their own share in tax revenues is concerned, maintain zero taxation for certain goods as well. However, another difficulty is the excise taxes on alcohol, which have some impact on agricultural policies. Many Member States are concerned that their wine farmers would suffer from excise taxes on wine. However, not all Member States would be equally affected. Moreover, the taxation of alcohol products strongly reflects social and cultural habits, which vary throughout the European Union. On the one hand, decisions could be easily achieved if for a certain number of Member States it is politically easier to agree upon harmonization measures. On the other hand, owing to the lack of “horizontal justice” among all Member States, if such a harmonized tax were to be introduced at the EU level, the resistance of the affected states could be much more severe.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

168   M. Lang and M. Zagler One could question whether the revenue raised by such a tax would be stable enough. Today, such taxes are levied not only to generate revenues for the government, but also to change the behavior of citizens. Consumption of tobacco and alcohol may harm the health of the population, and the tax burden imposed on such consumers should motivate citizens to change their behavior. Thus, one can assume that in an ideal world, the levy of such excise taxes would lead to a significant reduction in tobacco and alcohol consumption. However, evidence demonstrates that this risk should not be overestimated. First, one should distinguish between alcohol and tobacco. While Commission papers indicate that tobacco consumption leads to serious health problems and even death for a considerable proportion of the European population, the same studies assume that reasonable consumption of alcohol could even be healthy (European Commission 2004: 19). In most Member States, excise taxes on alcohol are levied exclusively for financial reasons, and not for health policy reasons. Although in the case of tobacco, excise taxes are different, evidence shows that even significant increases in alcohol excise taxes do not change consumer behavior dramatically. According to World Bank studies, an increase in the tobacco excise tax burden of 10 percent would lead to a mere 4 percent fall in consumption. Within the European Union, the average tobacco excise tax burden increased by approximately 30 percent from 2002 to 2006, whereas consumption dropped only by about 10 percent (European Commission 2008c: 7). These figures indicate that excise taxes on tobacco could still provide a stable revenue basis even if the amount of the tax rate were largely driven by health policy considerations. The collection of such a tax could be more difficult for some goods and less difficult for others. As far as tobacco is concerned, there are, as will be explained subsequently when we discuss the economic arguments, only a few relevant players concerning the production of cigarettes and similar products. However, in the area of alcohol the number of producers is rather high, as is the number of different types of producers. Nevertheless, whether such a tax would be visible to the individual consumer is a different question. Today, most consumers do not know how much of the price that they pay for a product is due to the tax burden. Even if the tax part were “flagged up”, it is doubtful whether the consumer would pay attention. Although it is consumers who have to bear the taxes, they are not the ones who have to deliver the tax to the authorities. 8.5.3  Economic issues Excise duties on tobacco and alcohol are taxes on particular products, but otherwise have many things in common with other indirect taxes – in particular, the VAT described in the previous section. Both products are sold quite widely across Europe. There is limited market segmentation, although the industries have tried to institute a domestic bias,11 but only with limited success. There is a high degree of concentration in the tobacco industry, with the biggest five firms typically supplying more than 90 percent of the market

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   169 (Delipalla and O’Donnell 2001). Proposition ECO3 states that under these conditions, and for a destination-­based tax such as tobacco duties, the transfer of taxing rights to the European Union might reduce the market power of domestic firms, and thus lead to an improvement. However, typically the same five companies dominate not only every single market within Europe, but also the European market as a whole, so that there is limited scope for competition under the transfer of taxing rights to the European Union. Alcohol, on the other hand, is supplied under conditions of more perfect competition. A transfer of taxing rights to the European Union in the case of open markets and little market power would make sense only for source taxation, which is not the case here. Thus, there may be no gain from transferring taxing rights for alcohol to the European Union.

8.6  Energy taxation 8.6.1  Description Many European economies tax electricity and fuels, typically as a surcharge on quantities. The European Union is considering either using a similar tax scheme or taxing CO2 emissions. Considering the content of the Energy Directive 2003/96/EC,12 two main possibilities can be envisaged: a broad-­based energy tax and an energy tax on motor fuel used for transport. In the first option, the tax base would encompass all the energy sources covered by the directive, including mineral oils, electricity, coal and natural gas. A differentiation of excise duties according to the use or quality of the product as well as exemptions and tax refunds might be foreseen in a number of circumstances, as laid down in the new directive. This could, for example, be the case for energy-­ intensive companies, for which competitiveness issues with third countries are very sensitive. The tax would be raised when the taxable products are delivered for consumption and not at the level of the final consumer, to simplify tax collection. In the second option, the tax base would include only motor fuel used for transport – that is, only a part of the energy sources covered by the directive. 8.6.2  Legal and policy issues In the area of energy taxation, a certain degree of harmonization has already been achieved. This could be viewed as an advantage if one believes that agreeing on further steps toward harmonization is always easier than starting to harmonize an area. For example, Directive 2003/96/EC widened the scope of the European Union’s minimum rate system, previously limited to mineral oils, to take in all energy products, including coal, natural gas and electricity. In particular, the directive reduced distortions of competition that existed between Member States because of divergent tax rates and reduced distortions of competition between mineral oils and other energy products. Because of these developments,

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

170   M. Lang and M. Zagler Member States are aware that this area of taxation is no longer exclusively under their control and that their room for maneuver regarding tax policy is limited. The fact that some types of energy taxes, particularly energy taxes on fuels, have a huge impact on Member States’ budgets has its pros and cons. On the one hand, such taxes could serve as a serious source of revenue for the European Union, once the Union gets the revenue, or a share of it. On the other hand, the sensitivity of Member States in that area is even higher, as they know that shifting the revenues of such a tax to the EU level could have a severe budgetary impact on them. As has already been indicated, a surcharge collected by Member States could be levied on top of an EU-­wide energy tax, which might leave room for domestic policy. However, such a surcharge could lead to additional compliance costs for both taxpayers and governments if the tax base was not identical and if a Member State preferred to exempt a part of the European tax base when levying its own surcharge. This problem could be avoided only if the prospect for Member States’ tax policy were limited to deciding on the tax rate and did not include the tax base. Although some types of energy taxes may raise substantial revenues, their intention is to influence human behavior and investment decisions. However, unless such taxes are not so high as to be considered prohibitive, their levying will not prevent business and consumers from using energy. Thus, there may not be any significant risk with regard to such a tax not being a revenue raiser. However, this depends on the tax base as well. If the tax did not cover all sources of energy that can easily be substituted by each other, a tax levied on certain energy sources might only lead to an increase in the costs of that type of energy, and could change consumer behavior as well, which finally could also have the effect of reducing tax revenues. The question of whether energy taxes provide for stable revenues depends on other effects as well. The market price of certain types of energy also depends on political factors. In times of war, oil prices tend to go up. Energy-­exporting countries could put pressure on European countries and threaten not to deliver energy, or to deliver less. Thus, a shortage in the energy supply would increase the price. On the one hand, if the tax were levied as a percentage of the price of the product, then this could increase the tax revenues as well. By contrast, if there were a real shortage so that less energy was purchased by private consumers or business, or if the price increase led to a change in consumer behavior, then the tax revenues might go down as well. Energy taxes are easy to collect if the number of suppliers is small. However, the downside is that consumers are usually not aware of the amount by which the tax burden has increased their price. Visibility to citizens is not a strength of energy taxation. 8.6.3  Economic issues Energy is sold in various forms, particularly as electricity and fuels, both to consumers and producers. As there are very few suppliers of both electricity and

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   171 fossil fuels, we can observe a lot of market concentration. Energy is easily transportable across Europe, and hence we find little market segmentation. When energy is sold to producers, the tax is levied at the place of use, and is therefore a source tax; and if energy is sold to consumers, it is a destination tax. Given these market conditions, we can find ourselves in the lower left cell of Table 8.1, which indicates that a transfer of taxing rights to the EU level is beneficial in the case of source taxation, as it will destroy some of the market power of domestic firms. This will lead to a more efficient market outcome at the EU level. Given that the effect is neutral in the case of a destination tax, a transfer of taxing rights is beneficial in the case of energy taxation for firms, while it leaves households unaffected. The only caveat to mention is that energy taxes have a very limited control effect, and hence neither national nor European taxation can change much about the use of energy through taxes (Zagler et al. 2008). However, this issue is beyond the scope of this chapter, which is concerned only with improvements in market conditions.

8.7  Climate charge on aviation 8.7.1  Description A climate charge on aviation is similar to an energy tax for air transport. The tax can be levied either as a CO2 charge or as a tax on airline fuels. 8.7.2  Legal and policy issues A climate charge on aviation could help to achieve Community environmental objectives while contributing to the financing of the European Union. For these purposes, the consumption of aviation kerosene could be taxed. A tax on aircraft fuel would decrease CO2 emissions by decreasing the demand for air travel, and by giving airline companies a greater incentive to improve aircraft fuel efficiency. Aviation kerosene can be taxed when delivered for consumption, as in the case of other mineral oils (European Parliament 2002). The current legal framework (Directive 2003/96/EC) provides a tax exemption for kerosene for international aviation, which allows Member States to tax kerosene only on domestic flights or via bilateral agreements between Member States.13 To make a climate charge on aviation possible, an amendment of the current legal situation would be necessary. Initially, this may seem to be a disadvantage, as no harmonized rules exist that could be used as a basis for the development of an EU-­wide aviation tax. On the other hand, except on domestic flights, Member States do not have any competence to levy a tax immediately. Thus, the introduction of an EU-­wide tax would not hurt their tax revenues. Moreover, as they are not used to exercising any tax policy in that area, from their point of view an aviation tax might not require them to give up their existing competence.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

172   M. Lang and M. Zagler However, it should be taken into account that many aircraft are also operated for extra-­EU flights, following intra-­EU flights. Thus, airlines could avoid such a tax by getting their fuel delivered outside the European Union. Even if the Union tried to negotiate treaties with neighboring countries to have this type of tax levied in those countries as well, airlines could still minimize their tax burden by using airports in other countries that are not included. An extensive geographical coverage might be needed to secure the effectiveness of a tax on aviation fuel. To create well-­balanced taxation, an international approach with third countries would be required. The Commission presented a Communication on the taxation of aircraft fuel in 2000 (European Commission 2000) and brought up this issue for discussion at the Thirty-­third Assembly of the International Civil Aviation Organization (ICAO) in September 2001: “It appeared from the discussions that it will be very difficult, if not impossible, to reach an agreement on this issue at ICAO level.” However, unlike in the case of financial transactions, it would not be necessary to include the whole world, as it is not practical to consider that airplanes might evade such a tax by fueling their planes in other continents. It is questionable whether it would make sense to tax kerosene but not to include, say, petrol. If petrol is not included in such a tax or if it is not ensured that the consumption of petrol triggers a tax liability that is at least identical, then consumers might use cars and other means of transportation more frequently. From an environmental point of view, such a tax could create an incentive to use cars instead, which could harm the environment. On the other hand, administrative and compliance costs would be expected to remain limited. The number of taxpayers will be small if the airlines are forced to levy the tax. The collection of the tax should not raise too many difficulties. However, certainly airlines might try to shift their tax burden, as a part of their costs, to the consumers. Nevertheless, the amount of the tax burden would not be visible to consumers. 8.7.3  Economic issues Air transport is, by definition, one of the most integrated markets in Europe and the world. However, there are two limitations. First, the endpoints of a voyage are typically defined and cannot easily be altered. In particular, in most cases it is not possible to change the country of origin and the country of destination, with very few exceptions, as when two or more airports lie within a short distance of each other across a frontier, which is the case for Basel, Luxembourg and Vienna, for example. However, the position of hubs, where passengers change flights between their point of origin and destination, is highly mobile. As the existence of a hub is of significant importance for an economy, economies compete with each other by offering low charges for aviation, and thus no European country taxes airline fuel. The general idea is that a European tax on airline fuel could eliminate the harmful competition in the international taxation of aviation. The other dimension of analysis is market power. While we see a great deal of competition in point-­to-point flights through discount airlines, the network

The case for and against an EU tax   173 and business market is dominated by three large alliances. In the framework of our analysis, this puts us in the lower left cell of Table 8.1, which suggests transferring taxing rights to the European Union, as doing so would help destroy the market power of carriers.

8.8  Communication tax

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

8.8.1  Description A communication tax is based on the very large concept of communication, including telecommunications, data transfer, broadcasting, as well as transport. As we have already discussed transportation taxation, we will now focus on voice and data communication. 8.8.2  Legal and policy issues A tax on voice and data communication could raise some tensions with other policy goals of Member States and even human rights questions. Many Member States grant incentives to stimulate communication. In particular, the internet is used widely, and it should be easy and inexpensive for everyone to gain access to the internet. The underlying assumption is that the use of the internet and other means of communication ensures that people living in different social conditions and different regions have access to all kind of information. These measures help to establish a knowledge-­based society – and levying taxes on communication could endanger that goal. Furthermore, there could be a human rights aspect as well, as freedom of speech requires access to information. At the technical level, one has to take into account that if services regarding voice and data communication simultaneously came under the VAT regime, then economic double taxation would result. No specific EC directive covers a communication tax; however, there are many provisions that relate to the VAT treatment of telecommunications services, broadcasting services and electronically supplied services. Therefore, the introduction of an EU communication tax could be encompassed by possible interpretation conflicts with the existing VAT provisions. The introduction of such a tax would have to be accompanied by an adjustment of VAT rules. The collection of such a tax does not seem to be too difficult. The number of providers of these services is limited. The providers could be obliged to collect the tax and flag its amount on their invoices. However, it is doubtful whether the tax burden would be sufficiently visible for the final consumers. 8.8.3  Economic issues In contrast to transport taxes or tobacco and alcohol taxes, there is no obvious negative externality associated with voice or data communication. Thus, the

174   M. Lang and M. Zagler

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

introduction of a special excise tax is per se problematic from a public finance perspective. Hence, very few countries actually have a communication tax. Communication services are local by definition. For example, buying internet access in a foreign country will not help domestic residents to satisfy their communication needs. There are typically very few local suppliers – usually the former state monopolist and a handful of other firms. With high market power and market segmentation, proposition ECO1 suggests keeping the tax, if it exists, local.

8.9  A tax on financial transactions 8.9.1  Description A tax on financial transactions (known as a Tobin tax) would be levied whenever a financial asset changed ownership. This raises two problems. The first question is how to differentiate financial transactions from other transactions. Selling a company would lead to taxation, while selling all the assets (buildings, machines and equipment, the client stock, patents, etc.) individually to the same buyer might not. Second, assets would have to be evaluated, which might create substantial problems in the case of derivatives. 8.9.2  Legal and policy issues It is difficult to analyze a tax on financial transactions, in general, from a legal perspective, because such a tax could be structured in different ways. However, if such a tax were brought in, it would need to cover all kinds of financial transactions that can be substituted by each other. Otherwise, such a tax would be easy to manipulate and would not lead to any significant revenues, only to a change in the behavior of actors in the financial market. Moreover, the introduction of such a tax would make sense only if its scope were not limited to the European Union and it did not cover only the financial transactions that take place within the Union or made by parties located within the Union; otherwise, such a tax would merely lead to financial transactions being conducted outside Europe. As a result, tax planning opportunities would increase, along with compliance costs. European financial centers might suffer and the tax might not generate significant tax revenues. Therefore, it seems to be necessary to develop a global concept as regards such a tax, and to consider at least the important financial centers outside Europe. Because the future of financial centers in Europe would be at risk if taxpayers were successful in maneuvering around such a tax, it is likely that those Member States where such centers are located will be particularly sensitive toward the introduction of such a tax. Even if it could be ensured that such a tax could not be circumvented, one has to assume that such a tax might impede the dynamic of financial markets. Therefore, those Member States that have benefited most from that dynamic in the past might be more resistant to the introduction of such a tax.

The case for and against an EU tax   175

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

If all or most of the revenues of such a tax were delivered to the European Union and not retained within the Member States, then these states might even show less support. Whether such a tax would be easy to collect would depend on its structure. It has to be anticipated that such a tax would not be restricted to financial transactions taking place at stock exchanges; otherwise, it would be easy to circumvent it. If all transactions between enterprises, or even between private persons, were covered, then serious collection problems might arise. 8.9.3  Economic issues In the recent economic turmoil, a tax on financial markets has always been mentioned as a possibility to reduce forces of instability stemming from financial markets. Typically, the argument based on mobile capital is that no single country can raise a tax on financial transactions without the loss of a large volume of business and available capital in the economy.14 The product that would be taxed is not the asset itself or the provision of funds to the issuer of a bond, but the service that the financial market provides in transferring assets from one owner to another. This service is source based, and the charge would fall not on the final consumer (the buyer of the asset), but at the point of sale or at the source. Financial markets are certainly highly competitive and there is no segmentation, which is in agreement with proposition ECO2, which states that there is no efficiency gain in transferring taxing rights to the European level. National governments could individually agree on the same level of taxation without any benefits of a European tax. However, it may be politically easier to transfer taxing rights and let the European Union decide the rate, instead of agreeing among Member States on the level of taxes and any potential changes.

8.10  Transfer of seignorage revenue 8.10.1  Description Seignorage enables central banks to issue money printed at a much lower cost than its face value. The issue of new money devalues existing money holdings through the process of inflation. Thus, it is current holders of money who pay the seignorage, or inflation tax. 8.10.2  Legal and policy issues Within the Eurosystem, new currency is issued by national central banks on behalf of the European Central Bank (ECB). When the central bank issues currency, it purchases something else for that amount – typically, bonds. While the money issued to the public bears no interest, the ECB will receive a return on bonds, which will increase its profits.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

176   M. Lang and M. Zagler Currently, these profits are redistributed to the national central banks on the basis of a complex weighting scheme that includes GDP and population weights. Seignorage is then distributed to the shareholders of the national central banks, which are mostly governments or public or semi-­public institutions. In the end, national governments are the beneficiaries of the seignorage tax. The actual revenue depends on national legislation on the division of central bank’s profits. Neither the ECB nor its member banks publish official figures, but ­seignorage revenues are estimated to be around €10 billion. This tax is therefore certainly not sufficient to finance the EU budget. Very few people who hold money that bears no interest are aware of the fact that the central bank receives these interest payments instead. Money holdings are therefore indirectly taxed by the central bank. Thus, a seignorage tax has very limited visibility. A final problem is posed by countries that are members of the European Union but not of the Eurozone. Currently, these countries collect seignorage individually. As seignorage revenue increases with looser monetary policy, some countries might have to give up a larger share of revenues if national seignorage revenues were also allocated to the EU level. 8.10.3  Economic issues Seignorage is a tax on the holdings of cash. Thus, the use of the methods of analysis developed in subsection 8.1.2 is far-­fetched. Moreover, for all members of the Eurozone, the tax is already collected at the European level and then distributed among Member States, on the basis of relatively arbitrary weights. Hence, this is a prime candidate for an EU tax. The only serious issue is how to treat economies that are not members of the Eurozone. For consistency, we will briefly sketch the analysis pursued in this chapter. Demand comes from all economic agents with a need to make a transaction based on money, while the supplier of cash is the central bank. The central bank is a monopoly by definition. However, it is not in the business of maximizing profits; rather, it has a very different mission, which is to achieve the social optimum with its policy. Therefore, as regards money supply a good central bank would do what a market under perfect competition would do for consumer products. To the holder of the bills, it makes no difference where these have been issued, whether in his or her home country or abroad. Hence, the tax is levied at the destination and not at the source. With no exploitation of market power and no market segmentation, we would be in the lower right cell of Table 8.1, which indicates a transfer of taxing rights to the European Union in the case of a destination tax. This is therefore in line with the above-­mentioned concept.

8.11  Summary This chapter has analyzed the advantages and disadvantages of several types of taxes that could be levied at the European level from a legal and economic

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

The case for and against an EU tax   177 perspective. We have investigated a modulated VAT, an EU corporate income tax, an energy tax, excise duties on tobacco and alcohol, a personal income tax and a climate charge on aviation. The analysis has identified which type of tax – if any – it is most logical and desirable to introduce at the EU level. We found that neither an economic analysis alone nor a legal analysis by itself could identify candidates for a European tax. Furthermore, we observed a consensus among the disciplines in favor of a modulated VAT and ecological taxation. The former observation is not surprising, because EU finances are already partly based on national VAT revenues, while the latter observation coincides with a recent EU proposal to introduce environmental taxes. The environment constitutes a global public good, and can therefore hardly be controlled at the national level. Thus, the transfer of taxing rights to the European level seems a natural move. Both disciplines also agree on the rejection of a personal income tax at the European level. This is noteworthy, as the European Union’s own resources are also based on GNP, which is closely related to national income. However, because of the complexity of national income tax systems and its adoption as an instrument for redistribution, a transfer of taxing rights to the European Union is undesirable. We found that in some cases, economics would favor a tax, while a legal analysis would object. The strongest example here is a seignorage tax, which is already levied at a European level for members of the Eurozone, but, owing to a lack of transparency, has been rejected as an EU tax on legal grounds. Several taxes receive mixed results, such as corporate income tax, excise duties on tobacco and alcohol, and taxes on communications and financial transactions. Simplicity of the tax system would certainly be a virtue for the introduction of EU taxes. The preceding analysis suggests that a modulated VAT and energy taxes should be considered as the main instruments for a potential EU tax system.

Notes   1 Only lump-­sum or poll taxes are non-­distortionary. The distortionary effect can be desired in the case of Pigovian taxes, for example for environmental protection, but undesired otherwise.   2 We will explain this measure below.   3 The first-­order condition reads ∂∂xπi = (1 − τ) f ( X ) + (1 − τ) xi f '( X ) − c '( xi ) = 0 . Reformulating this expression and substituting the inverse demand function yields (1 − τ) p[1 + xXi Xff ('(XX)) ] = c '( xi ) . By defining the demand elasticity ε = − Xff ('(XX)) and the market share si = xXi , we obtain the equation that follows in the main text.   4 It measures the change in the percentage of prices when the demand changes by 1 percent.   5 For an overview, see Tilly et al. (2007).   6 See the contribution by Hirschler and Zagler in this volume (Chapter 7).   7 See, for example, the treatment of certain dividend and interest payments under the Austrian Personal Income Tax Code.   8 Contrary to Traversa (2008).

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

178   M. Lang and M. Zagler   9 For a more thorough analysis, see Zimmermann (2005). 10 Council Directive 2008/9/EC of 12 February 2008, laying down detailed rules for the refund of value added tax. 11 For example, we typically imagine that the French smoke only Gauloises cigarettes and Germans drink only German beer. 12 Directive 2003/96/EC is on restructuring the Community framework for the taxation of energy products and electricity. 13 Article 14 (Directive 2003/96/EC): “1) In addition to the general provisions set out in Directive 92/12/EEC on exempt uses of taxable products, and without prejudice to other Community provisions, Member States shall exempt the following from taxation under conditions which they shall lay down for the purpose of ensuring the correct and straightforward application of such exemptions and of preventing any evasion, avoidance or abuse: . . . (b) energy products supplied for use as fuel for the purpose of air navigation other than in private pleasure-­flying. For the purposes of this Directive ‘private pleasure-­flying’ shall mean the use of an aircraft by its owner or the natural or legal person who enjoys its use either through hire or through any other means, for other than commercial purposes and in particular other than for the carriage of passengers or goods or for the supply of services for consideration or for the purposes of public authorities . . . 2) Member States may limit the scope of the exemptions provided for in paragraph 1(b) and (c) to international and intra-­Community transport. In addition, where a Member State has entered into a bilateral agreement with another Member State, it may also waive the exemptions provided for in paragraph 1(b) and (c). In such cases, Member States may apply a level of taxation below the minimum level set out in this Directive.” 14 Note that the United Kingdom has a very similar tax, without much loss for the City of London. This may be due to its very special and unique position in world financial markets.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Bibliography

Aigner, D. (2009) Die Sparzinsenrichtlinie, Vienna: Linde. Aigner, D. and Gläser, L. (2005) “Die Zahlstelle im Anwendungsbereich des EU-­ Quellensteuergesetzes”, Steuer und Wirtschaft International, 6: 261–267. Aivazian, V. A., Booth, L. and Cleary, S. (2006) “Dividend smoothing and debt ratings”, Journal of Financial and Quantitative Analysis, 41 (2): 439–453. Allen, F. and Michaely, R. (2003) “Payout policy”, in G. M. Constantinides, M. Harris and R. M. Stulz (eds.) Handbook of the Economics of Finance, 1st ed., vol. 1, Oxford: Elsevier. Allingham, M. G. and Sandmo, A. (1972) “Income tax evasion: a theoretical analysis”, Journal of Public Economics, 1: 323–338. Alm, J., McClelland, G. H. and Schulze, W. D. (1992) “Why do people pay taxes?”, Journal of Public Economics, 48 (1): 21–38. Altshuler, R. and Grubert, H. (2001) “Where will they go if we go territorial? Dividend exemption and the location decisions of U.S. multinational corporations”, National Tax Journal, 54: 787–809. Altshuler, R. and Grubert, H. (2002) “Repatriation taxes, repatriation strategies and multinational financial policy”, Journal of Public Economics, 87 (1): 73–107. Altshuler, R. and Grubert, H. (2005) “The three parties in the race to the bottom: host governments, home governments and multinational corporations”, Florida Tax Review, 7: 1–33. Altshuler, R., Grubert, H. and Newlon, S. T. (2001) “Has U.S. investment become more sensitive to tax rates?”, in J. R. Hines Jr (ed.) International Taxation and Multinational Activity, Chicago: University of Chicago Press. Amoroso, L. (1930) “La curva statica di offerta”, Giornali degli Economisti e Rivista di Statistica, 70: 1–26. Andreoni, J., Erard, B. and Feinstein, J. (1998) “Tax compliance”, Journal of Economic Literature, 36: 818–860. Andres, C., Betzer, A., Goergen, M. and Renneboog, L. (2009) “Dividend policy of German firms: a panel data analysis of partial adjustment models”, Journal of Empirical Finance, 16: 175–187. Arnold, B. J. (2000) “Controlled foreign corporation rules, harmful tax competition, and international taxation”, in 2000 World Tax Report, Report of the Proceedings of the First World Tax Conference: Taxes without Borders, Tampa Bay, Florida, 26 February–1 March 2000, 17: 1–26, Toronto: Canadian Tax Foundation. Arnold, B. J. and McIntyre, M. J. (1995) International Tax Primer, The Hague: Kluwer Law International.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

180   Bibliography Auer, K. (2007) “Ausschüttungsbemessung bei kapitalmarktorientierten Unternehmen im Wandel von Rechnungslegung und Kapitalerhaltung”, unpublished thesis, Vienna University of Economics and Business. Auerbach, A. and Hassett, K. (1993) “Taxation and foreign direct investment in the United States: a reconsideration of the evidence”, in A. Giovannini, R. G. Hubbard and J. Slemrod (eds.) Studies in International Taxation, Chicago: University of Chicago Press. Aujean, M. (2008) “The CCCTB project and the future of European taxation”, in M. Lang, P. Pistone, J. Schuch and C. Staringer (eds.) Common Consolidated Corporate Tax Base, Vienna: Linde. Avi-­Yonah, R. S. (1996) “The structure of international taxation: a proposal for simplification”, Texas Law Review, 74 (6): 1301–1359. Avi-­Yonah, R. S. (2006) International Tax as International Law: An Analysis of the International Tax Regime, New York: Cambridge University Press. Avi-­Yonah, R. S. (2007) “Double tax treaties: an introduction”. Online, available at: http://ssrn.com/abstract=1048441. Avi-­Yonah, R. S. and Clausing, K. A. (2007–2008) “Reforming corporate taxation in a global economy: a proposal to adopt formulary apportionment”, Hamilton Project Discussion Paper, Washington, DC: Brookings Institution. Azémar, C., Desbordes, R. and Mucchielli, J. L. (2007) “Do tax sparing agreements contribute to the attraction of FDI in developing countries?”, International Tax and Public Finance, 14: 543–562. Bacchetta, P. (1992) “Liberalization of capital movements and of the domestic financial system”, Economica, NS, 59 (236): 465–474. Bacchetta, P. and Espinosa, M. P. (1995) “Information sharing and tax competition among governments”, Journal of International Economics, 39: 103–121. Bacchetta, P. and Espinosa, M. P. (2000) “Exchange-­of-information clauses in international tax treaties”, International Tax and Public Finance, 7 (3): 275–293. Bach, S. (1993) Die Idee der Cash-­Flow-Steuer vor dem Hintergrund des gegenwärtigen Steuersystems, Berlin: Duncker & Humblot. Baker, P. (2001) Double Taxation Conventions and International Law, London: Sweet & Maxwell. Baker, P. and Mitroyanni, J. (2008) “The CCCTB rules and tax treaties”, in M. Lang, P. Pistone, J. Schuch and C. Staringer (eds.) Common Consolidated Corporate Tax Base, Vienna: Linde. Baldry, J. C. (1986) “Tax evasion is not a gamble: a report on two experiments”, Economics Letters, 22 (4): 333–335. Barba Navaretti, G. and Venables, A. J. (2004) Multinational Firms in the World Economy, Princeton, NJ: Princeton University Press. Barker, W. B. (2007) “An international tax system for emerging economies, tax sparing, and development: it is all about source!”, University of Pennsylvania Journal of International Law, 29 (2): 349–389. Barnard, J. (2003) “Former tax havens prepared to lift bank secrecy”, Bulletin for International Fiscal Documentation, 57 (1): 9–13. Bartelsman, E. J. and Beetsma, R. M. W. J. (2003) “Why pay more? Corporate tax avoidance through transfer pricing in OECD countries”, Journal of Public Economics, 87: 2225–2252. Barthel, F., Busse, M. and Neumayer, E. (2008) “The impact of double taxation treaties on foreign direct investment: evidence from large dyadic panel data”. Online, available at: http://ssrn.com/abstract=1234163.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Bibliography   181 Baumol, W. J., Panzar, J. C. and Willig, R. D. (1982) Contestable Markets and the Theory of Industry Structure, New York: Harcourt Brace Jovanovich. Becker, G. S. (1968) “Crime and punishment: an economic approach”, Journal of Political Economy, 76 (2): 169–217. Becker, G. S. (1997) “The economic way of looking at life”, in T. Persson (ed.) Nobel Lectures, Economics 1991–1995, Singapore: World Scientific Publishing. Behm, U. and Zimmermann, H. (1993) “The empirical relationship between dividends and earnings in Germany”, Zeitschrift für Wirtschafts- und Sozialwissenschaften, 113: 225–254. Bellak, C. and Leibrecht, M. (2008) “Corporate income taxation in Central and Eastern European countries and tax competition for foreign direct investment”, in J. H. Dunning and Ph. Gugler (eds.) Foreign Direct Investment, Location and Competitiveness, Progress in International Business Research, vol. 2, Amsterdam: Elsevier. Bellak, C., Leibrecht, M. and Wild, M. (2009) “Does lowering dividend tax rates increase dividends repatriated? Evidence of intra-­firm dividend repatriation policies by German multinational enterprises”, Deutsche Bundesbank Discussion Paper no. 19. Benartzi, S., Michaely, R. and Thaler, R. H. (1997) “Do changes in dividends signal the future or the past?”, Journal of Finance, 52 (3): 1007–1033. Bergsman, J. (1999) “Advice on taxation and tax incentives for foreign direct investment”. Online, available at: www.ifc.org/ifcext/fias.nsf/AttachmentsByTitle/Advice_ on_Taxation_and_Tax_Incentives_for_FDI.pdf/$FILE/Advice+on+Taxation+and+Tax +Incentives+for+FDI.pdf. Bernard, A., Jensen, J. B. and Schott, P. K. (2008) “Transfer pricing by U.S.-based multinational firms”, US Census Bureau Center for Economic Studies Paper no. CES-­WP08-29. Bezborodov, S. (2007) “Freedom of establishment in the EC economic partnership agreements: in search of its direct effect on direct taxation”, Intertax, 12: 658–712. Bhabra, G. S., Jeong, J. and Powell, J. G. (2002) “Dividend smoothing and the cross-­ sectional determinants of corporate payout policy”, Advances in Financial Economics, 7 (2): 163–183. Bhattacharya, S. (1979) “Imperfect information, dividend policy, and the ‘bird in the hand’ fallacy”, Bell Journal of Economics, 10: 259–270. Biehl, D. (1985) “A federalist budgetary policy strategy for the European Union”, Policy Studies, 6 (2): 66–76. Biehl, D. (1990) “Financing the ECC budget”, in R. Prud’homme (ed.) Public Finance with Several Levels of Government, The Hague: Foundation Journal Public Finance, The Hague. Biehl, D. (1992) “Reform of the EC financial constitution”, in H. Cowie (ed.) Towards fiscal federalism: Delors II Budgetary Proposals 1993–7, London: Federal Trust for Education and Research. Bird, R. M. and Wilkie, J. S. (2000) “Source vs. residence-­based taxation in the European Union: the wrong question?”, in S. Cnossen (ed.) Taxing Capital Income in the European Union: Issues and Options for Reform, New York: Oxford University Press. Blonigen, B. A. and Davies, R. B. (2000) “The effects of bilateral tax treaties on US FDI activity”, NBER Working Paper no. 7929, Washington, DC: National Bureau of Economic Research. Blonigen, B. A. and Davies, R. B. (2002) “Do bilateral tax treaties promote foreign direct investment?”, NBER Working Paper no. 8834, Washington, DC: National Bureau of Economic Research.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

182   Bibliography Blonigen, B. A. and Davies, R. B. (2004a) “Do bilateral tax treaties promote foreign direct investment?”, in E. K. Choi and J. Hartigan (eds.) Handbook of International Trade: Economic and Legal Analysis of Laws and Institutions, Oxford: Blackwell. Blonigen, B. A. and Davies, R. B. (2004b) “The effects of bilateral tax treaties on U.S. FDI activity”, International Tax and Public Finance, 11: 601–622. Bond, M. T. and Mougoué, M. (1991) “Corporate dividend policy and the partial adjustment model”, Journal of Economics and Business, 43: 165–178. Bond, S. and Devereux, M. (2002) “Cash flow taxes in an open economy”, CEPR Discussion Paper no. 3401. Boskin, M. J. and Gale, W. G. (1987) “New results on the effects of tax policy on the international location of investment”, in The Effects of Taxation on Capital Accumulation, A National Bureau of Economic Research Project Report. Chicago: University of Chicago Press. Bradford, D. (2000) “Blueprint for international tax reform”, Brooklyn Journal of International Law, 26 (4): 1449–1463. Bradford, D. (2004) “The X tax in the world economy”, CESifo Working Paper no. 1264. Brähler, G. (2009) Internationales Steuerrecht: Grundlagen für Studium und Steuerberaterprüfung, 5th ed., Wiesbaden: Gabler. Bridges, D. (2006) “The disciplines and the discipline of educational research”, Journal of Philosophy of Education, 40 (2): 259–272. Brooks, K. (2007) “Denying tax sparing provisions: another way for high-­income countries to dictate the tax policy of low-­income countries?” Online, available at: www. allacademic.com/meta/p177927_index.html. Brown, K. (2002) “Missing Africa: should U.S. international tax rules accommodate investment in developing countries?”, University of Pennsylvania Journal of International Economic Law, 23 (1): 45–83. Büttner, T., Overesch, M., Schreiber, U. and Wamser, G. (2006) “Taxation and capital structure choice: evidence from a panel of German multinationals”, CESifo Working Paper no. 1841. Carayol, N. and Uyen Nguyen Thi, T. (2004) Why Do Academic Scientists Engage in Interdisciplinary Research?, Strasbourg: Université Louis Pasteur, Strasbourg I, Department of Economics. Casti, J. L. (1994) Complexification, New York: HarperCollins. Cattoir, P. (2004) “Tax-­based EU own resources: an assessment”, European Commission, Taxation Working Papers no. 1. Chetty, R. and Saez, E. (2007) “An agency theory of dividend taxation”, NBER Working Paper no. 13538, Washington, DC: National Bureau of Economic Research. Chisik, R. and Davies, R. B. (2004a) “Asymmetric FDI and tax-­treaty bargaining: theory and evidence”, Journal of Public Economics, 88: 1119–1148. Chisik, R. and Davies, R. B. (2004b) “Gradualism in tax treaties with irreversible foreign direct investment”, International Economic Review, 45: 113–139. Clausing, K. A. (2003) “Tax-­motivated transfer pricing and US intra-­firm trade prices”, Journal of Public Economics, 87: 2207–2223. Clausing, K. A. (2006) “International tax avoidance and US international trade”, National Tax Journal, 59 (2): 269–287. Cnossen, S. (ed.) (1987) Tax Coordination in the European Community, Deventer, Netherlands: Kluwer Law and Taxation Publishers. Cnossen, S. (2001) “Tax policy in the European Union: a review of issues and options”, FinanzArchiv, 58 (4): 466–558.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Bibliography   183 Council Directive 2003/96/EC (2003) Restructuring the Community framework for the taxation of energy products and electricity, 27 October. Coupe, T., Orlova, I. and Skiba, A. (2008) “The effect of tax and investment treaties on bilateral FDI flows to transition countries”, mimeo. Cowell, F. A. (1990) Cheating the Government, Cambridge, MA: MIT Press. Cressy, R. and Olofsson, C. (1997) “European SME financing: an overview”, Small Business Economics, 9 (2): 87–96. Cross, R. B. and Shaw, G. K. (1981) “The evasion–avoidance choice: a suggested approach”, National Tax Journal, 34 (4): 489–491. Cummins, J. G. and Hubbard, G. R. (1995) “The tax sensitivity of foreign direct investment: evidence from firm-­level panel data”, in M. Feldstein, J. R. Hines Jr and R. G. Hubbard (eds.) The Effect of Taxation on Multinational Corporations, Chicago: University of Chicago Press. Da Silva, C., Goergen, M. and Renneboog, L. (2004) Dividend Policy and Corporate Governance, Oxford: Oxford University Press. Dagan, T. (2000) “The tax treaties myth”, Journal of International Law and Politics, 32: 939–996. Damodaran, A. (2005) Applied Corporate Finance: A User’s Manual, 2nd ed., New York: Wiley. Davies, R. B. (2004) “Tax treaties and foreign direct investment: potential vs. performance”, International Tax and Public Finance, 11: 775–802. Davies, R. B., Norbäck, P.-J. and Tekin-­Koru, A. (2009) “The effect of tax treaties on multinational firms: new evidence from microdata”, The World Economy, 32: 77–110. De Mooij, R. A. (2005) “Will corporate income taxation survive?”, De Economist, 153: 277–301. De Mooij, R. A. and Devereux, M. P. (2009) “Alternative systems of business tax in Europe: an applied analysis of ACE and CBIT reforms”, European Commission DG Taxation and Customs Taxation Paper no. 17. Debatin, H. (1965) “Die Steuerharmonisierung in der EWG in Form der Konzernbesteuerungsrichtlinie”, Deutsche Steuer-­Zeitung, Ausgabe A: 146–153. Delipalla, S. and O’Donnell, O. (2001) “Estimating tax incidence, market power and market conduct: the European cigarette industry”, International Journal of Industrial Organization, 19 (6): 885–908. Denis, D. J. and Osobov, I. (2008) “Why do firms pay dividends? International evidence on the determinants of dividend policy”, Journal of Financial Economics, 89 (1): 62–82. Desai, M. A. and Hines, J. R. Jr. (2003) “Evaluating international tax reform”, National Tax Journal, 56 (3): 487–502. Desai, M. A., Foley, C. F. and Hines, J. R. Jr (2001) “Repatriation taxes and dividend distortions”, National Tax Journal, 54 (4): 829–851. Desai, M. A., Foley, C. F. and Hines, J. R. Jr (2002) “Dividend policy inside the firm”, NBER Working Paper no. 8698, Cambridge, MA: National Bureau for Economic Research. Desai, M. A., Foley, C. F. and Hines, J. R. Jr (2006a) “Dividend policy inside the multinational firm”, Financial Management, 361: 5–26. Desai, M. A., Foley, F. C. and Hines, J. R. Jr (2006b) “Do tax havens divert economic activity?”, Economics Letters, 90: 219–224. Desai, M. A., Foley, F. C. and Hines, J. R. Jr (2006c) “The demand for tax haven operations”, Journal of Public Economics, 90: 513–531.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

184   Bibliography Devereux, M. P. (1990) “Capital export neutrality, capital import neutrality and capital ownership neutrality and all that”, IFS Working Paper, London: Institute for Fiscal Studies. Devereux, M. P. (2007) “The impact of taxation on the location of capita, firms and profit: a survey of empirical evidence”, Oxford University Centre for Business Taxation Working Papers no. 07/02. Devereux, M. P. and Sørensen, P. B. (2006) “The corporate income tax: international trends and options for fundamental reform”, Economic Papers 264, Brussels: European Commission. Devereux, M. P., Lockwood, B. and Redoano, M. (2004) “Do countries compete over corporate tax rates?”, mimeo, University of Warwick, UK. Dewenter, K. L. and Warther, V. A. (1998) “Dividends, asymmetric information and agency conflicts: evidence from a comparison of the dividend policies of Japanese and US firms”, Journal of Finance, 53: 879–904. Dhami, S. and al-­Nowaihi, A. (2005) “Why do people pay taxes: expected utility versus prospect theory”, mimeo, Department of Economics, University of Leicester. Dharmapala, D. (2008) “What problems and opportunities are created by tax havens?”, Oxford University Centre for Business Taxation Working Papers no. 08/20. Dischinger, M. and Riedel, N. (2008) “Corporate taxes and the location of intangible assets within multinational firms”, Munich Economics Discussion Papers no. 2008-15, Munich: University of Munich. Doralt, W. (2009) Umsatzsteuergesetz 1994, Vienna: Manz. Dourado, A. and de la Feria, R. (2008) “Thin capitalization rules in the context of the CCCTB”, Oxford University Centre for Business Taxation Working Paper no. 08/04. Duncan, J. (2000) “General Report”, in IFA Cahiers de droit fiscal international, vol. 85a, Tax Treatment of Hybrid Financial Instruments in Cross-­border Transactions, The Hague: IBFD. Easson, A. (1999) Taxation of Foreign Direct Investment: An Introduction, The Hague: Kluwer Law International. Easterly, W. (2001) The Elusive Quest for Growth: Economists’ Adventures and Misadventures in the Tropics, Cambridge, MA: MIT Press. Easterly, W. (2003) “Can Foreign Aid Buy Growth?”, Journal of Economic Perspectives, 17 (3): 23–48. Eberhartinger, E. (2000) Ertragsteuerliche Konsequenzen der Internationalisierung der Rechnungslegung, Vienna: LexisNexis ARD Orac. Eberhartinger, E. (2005) “Besteuerung und steuerliche Gestaltung von hybriden Finanzierungsinstrumenten”, in M. Bischof and E. Eberhartinger (eds.) Hybride Finanzierungsinstrumente, Vienna: Linde. Eberhartinger, E. and Pummerer, E. (2009) “Hybride Konzernfinanzierung”, in S. Urnik, G. Fritz-­Schmied and S. Kanduth-­Kristen (eds.) Steuerwissenschaften und betriebliches Rechnungswesen: Strukturen – Prinzipien – Neuerungen, Vienna: Linde. Eberhartinger, E. and Pummerer, E. (2010) “Grenzüberschreitende Konzernfinanzierung und die Nutzung ertragsteuerlicher Tarifdifferenzen”, zfb, forthcoming. Eberhartinger, E. and Six, M. (2009) “Taxation of cross-­border hybrid finance: a legal analysis”, Intertax, 37: 4–18. ECJ 6 March 2007, C-­292/04 Meilicke ECR [2007] I-­1885. ECJ 7 September 2002, C-­319/02 Manninen ECR [2004] I-­7486. Eden, L. and Kudrle, R. T. (2005) “Tax havens: renegade states in the international tax regime”, Law and Policy, 27: 100–127.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Bibliography   185 Egger, P., Larch, M., Pfaffermayr, M. and Winner, H. (2006) “The impact of endogenous tax treaties on foreign direct investment: theory and evidence”, Canadian Journal of Economics, 39: 901–931. Einem, C. von and Tränkle, C. (2004) “Neue steuerliche Rahmenbedingungen für den Mittelstand”, Venture Capital, 1: 38–41. Endfellner, C. and Dornhofer, J. (2005) “Die Einkommensbesteuerung von EU-­Beamten und sonstigen Bediensteten”, Steuer und Wirtschaft International (SWI), 1: 32–33. Engel, K. (2001) “Tax neutrality to the left, international competitiveness to the right, stuck in the middle with subpart F ”, Texas Law Review, 79: 1525–1607. Engelschalk, M. (2008) “Art 26”, in K. Vogel and M. Lehner (eds.) Doppelbesteuerungsabkommen Kommentar, Munich: C. H. Beck. Erard, B. and Feinstein, J. S. (1994) “Honesty and evasion in the tax compliance game”, Rand Journal of Economics, 25 (1): 1–19 . European Commission (1977) “Value added tax directive”, EC-­Directive 77/388/EWG modified by Directive 91/680/EWG. European Commission (1997) Agenda 2000 – For a Stronger and Wider Union. Part 3: The New Financial Framework (2000–2006), COM(1997) 2000 – Vol. 1. European Commission (2000) “Communication on taxation of aircraft fuel”, COM(2000) 110, 2 March. European Commission (2001) “Towards an internal market without tax obstacles: a strategy for providing companies with a consolidated corporate tax base for their EU-­wide activities”, COM(2001) 582 final, Brussels: European Commission. European Commission (2004) Report from 26 May 2004 on the Rates of Excise Duty Applied on Alcohol and Alcoholic Beverages, COM(2004) 223 final. European Commission (2007) “CCCTB: possible elements of a technical outline”, 20 November (CCCTB/WP057 annotated). European Commission (2008a) Taxation Trends in the European Union, Data for the EU Member States and Norway, 2008 edition, Luxembourg: Office for Official Publications of the European Communities. European Commission (2008b) Joint Report on Social Protection and Social Inclusion 2008, March. European Commission (2008c) Report from 16 July 2008 on the structure and rates of Excise Duty Applied on Cigarettes and Other Manufactured Tobacco Products, COM (2008) 460/2. European Commission (2009a) “Proposal of 28 January 2009 to reduce burdens on business and to help Member States tackle fraud”, IP/09/132. European Commission (2009b) “Taxation and Customs Union, Common Tax Base”. Online, available at: http://ec.europa.eu/taxation_customs/taxation/company_tax/ common_tax_base/ (accessed 8 May 2009). European Court of Justice (1993) “EuGH, C-­183/91”, Kommission/Griechenland, Tz: 17ff. European Parliament (1994a) Report on the System of Own Resources in the European Union, A3-0228/94, Rapporteur: H. Langes, 8 April. European Parliament (1994b) Report on a New System of Own Resources for the European Union, A3-0060/94, Rapporteur: H. Langes, 7 February. European Parliament (2002) “Sixth Community Environment Action Programme”, Decision no. 1600/2002 EC of the European Parliament and of the Council of 22 July. Eynatten, W. (2006) “Belgium”, in D. Aigner, L. Gläser and M. Tumpel (eds.) The Taxation of Interest Savings Income in the European Capital Market, Vienna: Linde. Feld, L. P. and Frey, B. S. (2002) “Trust breeds trust: how taxpayers are treated”, Economics of Governance, 3 (2): 87–89.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

186   Bibliography Fischer, H. (1968) “Harmonisierung der Steuergesetzgebung in der EWG”, Neue Jurist­ ische Wochenschrift: 236–244. Fleming, J. C. Jr, Peroni, R. J. and Shay, S. E. (2001) “Fairness in international taxation: the ability-­to-pay case for taxing worldwide income”, Florida Tax Review, 5 (4): 299–354. Foray, D. and Gibbons, M. (1996) “Discovery in the context of application”, Technological Forecasting and Social Change, 53: 263–277. Frankfurter, G. M. and Wood, B. G. Jr (2002) “Dividend policy theories and their empirical tests”, International Review of Financial Analysis, 11 (1): 111–138. Frey, B. S. (1992) Economics as a Science of Human Behaviour, Dordrect: Kluwer Academic. Frey, B. S. (1997) “A constitution for knaves crowds out civic virtues”, Economic Journal, 107: 1043–1053. Frey, R. L. and Torgler, B. (2002) “Entwicklung und Stand der Steuermoralforschung”, Wirtschaftswissenschaftliches Studium, 3: 130–135. Frisch, D. J. (1990) “The economics of international tax policy: some old and new approaches”, Tax Notes, 47: 581–591. Genschel, P., Kemmerling, A. and Seils, E. (2010) “Accelerating downhill: how the EU shapes corporate tax competion in the single market”, Journal of Common Market Studies, in print. Gibbons, M., Limoges, C., Novotny, H., Schwartzman, S., Scoot, P. and Trow, M. (1994) The New production of Knowledge, London: Sage. Gläser, L. (2005) “Der Begriff der ‘Umlauffähigen Schuldtitel’ im EU-­Quellensteuergesetz”, taxlex: 138–153. Gläser, L. (2005a) “EU-­Zinsenbesteuerung – Vermeidung der Doppelbesteuerung”, Steuer und Wirtschaft International, 15 (7): 325–330. Gläser, L. (2005b) Handbuch der EU-­Quellensteuer: 138–144. Gläser, L. and Halla, M. (2006) “The Savings Directive in the EU”, in D. Aigner, L. Gläser and M. Tumpel (eds.) The Taxation of Interest Savings Income in the European Capital Market, Vienna: Linde. Gläser, L. and Moritz, H. (2005) “EU-­Quellensteuerpflicht von Zertifikaten”, Steuer und Wirtschaft International, 80: 827–833. Goddard, J., McMillan, D. G. and Wilson, J. O. S. (2006) “Dividend smoothing vs dividend signalling: evidence from UK firms”, Managerial Finance, 32 (6): 493–504. Goergen, M., Renneborg, L. and da Silva, L. (2005) “When do German firms change their dividends?”, Journal of Corporate Finance, 11: 375–399. Goodspeed, T. J. (2006) “Taxation and FDI in developed and developing countries”, in J. Alm, J. Martinez-­Vaszquez and M. Rider (eds.) The Challenges of Tax Reform in a Global Economy, New York: Springer. Goodspeed, T. J., Martinez-­Vazquez, J. and Zhang, L. (2009) “Public policies and FDI location: differences between developing and developed countries”, Hunter College Economics Department Working Paper. Gordon, R. and Dietz, M. (2006) “Dividends and taxes”, NBER Working Paper no. 12292, Cambridge, MA: National Bureau of Economic Research. Graetz, M. J. (2003) Foundations of International Income Taxation, New York: Foundation Press. Griffith, R. and Klemm, A. (2005) “What has been the tax competition experience of the last 20 years?”, Institute for Fiscal Studies Working Paper no. 04/05, London: Institute for Fiscal Studies. Groeben, H. von der (1962) “Nationale Steuersysteme und gemeinsamer Markt”, Europäisches Steuerzeitung, 1: 6–14.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Bibliography   187 Gross, S. and Steiger, J. (2004) “Korb II-­Gesetz: Einkommen- und körperschaftsteuerliche Konsequenzen”, Bilanzbuchhalter und Controller, 2: 33–37. Grubert, H. (2006) “Comment on Desai and Hines, old rules and new realities: corporate tax policy in a global setting”, National Tax Journal, 58: 263–274. Grubert, H. and Mutti, J. (1991) “Taxes, tariffs and transfer pricing in multinational corporate decision making”, Review of Economics and Statistics, 73: 285–293. Grubert, H. and Newlon, T. S. (1997) Taxing Consumption in a Global Economy, Washington, DC: AEI Press. Grund, W. (1965) “Grundprobleme der Steuerharmonisierung und des internationalen Steuerrechts”, Steuer und Wirtschaft: 393–399. Gupta, S. and Tareq, S. (2008) “Mobilizing revenue”, Finance and Development, 45 (3) (September): 44–47. Online, available at: www.imf.org/external/pubs/ft/fandd/2008/09/ gupta.htm. Hartman, D. (1984) “Tax policy and foreign direct investment in the United States”, National Tax Journal, 37: 475–487. Haufler, A. and Runkel, M. (2008) “Firms’ financial choices and thin capitalization under corporate tax competition”, Oxford University Centre for Business Taxation Working Paper no. 08/15. Hayunga, D. K. and Stephens, C. P. (2008) “Dividend smoothing in equity REITs”, mimeo. Helminen, M. (2004) “Classification of cross-­border payments on hybrid instruments”, Bulletin for International Fiscal Documentation, 58: 56–61. Hey, J. (2008) “CCCTB – optionality”, in M. Lang, P. Pistone, J. Schuch and C. Staringer (eds.) Common Consolidated Corporate Tax Base, Vienna: Linde. Hiller, M. (2003) Cash-­Flow-Steuer und Umsatzsteuer: eine ökonomische Analyse unter Einbezug der US-­amerikanischen Reformüberlegungen, Wiesbaden: Deutscher Universitäts-Verlag. Hines, J. R. Jr (1996) “Altered states: taxes and the location of foreign direct investment in America”, American Economic Review, 86 (5): 1076–1094. Hines, J. R. Jr (1997) “Tax policy and the activities of multinational corporations”, in A. Auerbach (ed.) Fiscal Policy: Lessons from Economic Research, Cambridge, MA: MIT Press. Hines, J. R. Jr (1998) “ ‘Tax sparing’ and direct investment in developing countries”, NBER Working Paper no. 6728, Cambridge, MA: National Bureau of Economic Research. Online, available at: www.nber.org/papers/w6728. Hines, J. R. Jr (1999) “Lessons from behavioral responses to international taxation”, National Tax Journal, 52 (2): 305–322. Hines, J. R. Jr and Gordon, R. (2002) “International taxation”, in A. Auerbach and M. Feldstein (eds.) Handbook of Public Economics, vol. 4, Amsterdam: North-­Holland. . Hines, J. R. Jr and Rice, E. M. (1994) “Fiscal paradise: foreign tax havens and American business”, Quarterly Journal of Economics, 109: 149–182. Hochgatterer, C. and Leibrecht, M. (2009) “Tax competition as cause of falling corporate income taxes: a literature survey”, Vienna University of Economics and Business Discussion Papers SFB International Tax Coordination no. 32. Homburg, S. (2007) “Germany”s company tax act 2008”, FinanzArchiv, 63: 591–612. Huizinga, H. and Laeven, L. (2007) “International profit shifting within multinationals: a multi-­country perspective”, mimeo. Jarass, L. (2007) “Lobbyisten haben das Steinbrück-Konzept kastriert”, Frankfurter Rundschau, 23 April.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

188   Bibliography Jensen, M. C. and Meckling, W. H. (1976) “Theory of the firm: managerial behavior, agency costs and ownership structure”, Journal of Financial Economics, 3 (4): 305–360. Kaminski, B. and Strunk, G. (2006) Steuern in der internationalen Unternehmenspraxis, Hamburg: Gabler. Keen, M. and Ligthart, J. E. (2006a) “Incentives and information exchange in international taxation”, International Tax and Public Finance, 13 (2): 163–180. Keen, M. and Ligthart, J. E. (2006b) “Information sharing and international taxation: a primer”, International Tax and Public Finance, 13 (1): 81–110. Kessler, W. and Eicke, R. (2007) “Germany’s corporate tax reform: the road not taken”, Tax Notes International, 46: 1135–1137. Khan, T. (2006) “Company dividends and ownership structure: evidence from UK panel data”, The Economic Journal, 116: C172–C189. Kirchler, E., Maciejovsky, B. and Schneider, F. G. (2003) “Everyday representations of tax avoidance, tax evasion, and tax flight: do legal differences matter?”, Journal of Economic Psychology, 24 (2): 535–553. Kolm, S. C. (1973) “A note on optimum tax evasion”, Journal of Public Economics, 2 (3): 265–270. Krishnan, A. (2009) “What are academic disciplines? Some observations on the disciplinarity vs. interdisciplinary debate”, ESRC National Centre for Research Methods, NCRM Working Paper 03/09. Kubik, K. and Massoner, C. (2009) “Der aktuelle Stand der Common Consolidated Corporate Tax Base”, Finanz Journal, 13–17. Kuhn, R. D. (1963) “United States tax policy with respect to less developed countries”, George Washington Law Review, 32 (2): 261–268. Kumar, P. (1988) “Shareholder–manager conflict and the information content of dividends”, Review of Financial Studies, 1 (Summer): 111–136. Kummer, V. and Blome, H. (2008) “Die deutsche Unternehmensteuerreform 2008”, Steuer- und Wirtschaft International, 10: 470–474. Kuttin, M. (2003) “EU: Richtlinienentwurf über die Besteuerung von Zinserträgen”, Österreichische Steuer Zeitung: 127–134. La Porta, R., Lopez de Silanes, F., Shleifer, A. and Vishny, R. W. (1999) “Investor protection and corporate governance”. Online, available at: http://ssrn.com/abstract=183908 . Lang, M. (1993) Country chapter Austria, in IFA Cahiers de droit fiscal international, vol. 78a, Interpretation of double taxation conventions, Deventer, Netherlands: Kluwer. Lang, M. (1998) “Off-­track interpretation of double taxation treaties as exemplified by the taxation of lecturers”, Steuer und Wirtschaft International: 508–513. Lang, M. (1999) “The significance of national tax law in interpreting and applying double taxation treaties”, Steuer und Wirtschaft International: 61–69. Lang, M. (2003) Avoidance of Double Non-­taxation, Vienna: Linde. Lang, M. (2009) “Conflicts of qualification and double non-­taxation”, Bulletin for International Taxation, 63 (5): 204–207. Lang, M., Pistone, P., Schuch, J. and Staringer, C. (eds.) (2008) Common Consolidated Corporate Tax Base, Vienna: Linde. Layard, R., Nickell, S. and Jackman, R. (2005) Unemployment: Macroeconomic Performance and the Labour Market, Oxford: Oxford University Press. Lewis, K. (1999) “Trying to explain home bias in equities and consumption”, Journal of Economic Literature, 37 (2): 571–608.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Bibliography   189 Li, J. (2003) International Taxation in the Age of Electronic Commerce: A Comparative Study, Toronto: Canadian Tax Foundation. Lintner, J. (1956) “Distribution of incomes of corporations among dividends, retained eamings and taxes”, American Economic Review, 46: 97–113. McDaniel, P. (2003) “The U.S. tax treatment of foreign source income earned in developing countries”, George Washington International Law Review, 35 (2): 265–295. McDaniel, P. (2007) “Territorial vs. worldwide international tax systems: which is better for the U.S.?”, Florida Tax Review, 8 (3): 263–291. McKenzie, G. A., Osmond, D. W. H. and Gearson, G. P. (1997) “The composition of fiscal adjustment and growth: lessons from fiscal reforms in eight economies”, IMF Occasional Paper 149. Mackscheidt, K. (1973) “Die Entwicklung der Kölner Schule der Finanzpsychologie”, in C. Smekal and E. Theurl (eds.), Stand und Entwicklung der Finanzpsychologie, Baden-­ Baden: Nomos-­Verlagsgesellschaft. McLure, C. E. (2001) “Globalization, tax rules and national sovereignty”, Bulletin for International Fiscal Documentation, 55 (8): 328–341. McLure, C. E. and Weiner, J. M. (2000) “Deciding whether the European Union should adopt formula apportionment of company income”, in S. Cnossen (ed.) Taxing Capital Income in the European Union: Issues and Options for Reform, Oxford: Oxford University Press. Maisto, G. (2008) International and EC Tax Aspects of Groups of Companies, Amsterdam: IBFD. Maiterth, R. (2002a) “Die steueroptimale Finanzierung ausländischer Tochtergesellschaften”, Finanz Betrieb, 4: 566–576. Maiterth, R. (2002b) “Zur sachgerechten Behandlung von Beteiligungsaufwendungen im Steuerrecht”, Die Betriebswirtschaft, 62: 169–183. Marshall, A. (1890) Principles of Economics, London: Macmillan. Mason, J. (2006) “Six strategies for mixing methods and linking data in social science research”, ESRC National Centre for Research Methods, NCRM Working Paper 4/06, University of Manchester. Mazz, A. and Pistone, P. (eds.) (2010) Reflexiones en torno a un modelo latinoamericano de convenio de doble imposición, Montevideo: Fundación de Cultura Univer­ sitaria. Meade, J. E. (1978) “The structure and reform of direct taxation”, Economic Journal, 89 (354): 243–260. Miller, M. H. and Rock, K. (1985) “Dividend policy under asymmetric information”, Journal of Finance, 40: 1031–1051. Millimet, D. and Kumas, A. (2007) “Reassessing the effects of bilateral tax treaties on US FDI activity”, mimeo. Mirrlees, J. A. (1971) “An exploration in the theory of optimal income taxation”, Review of Economic Studies, 38: 175–208. Moran, J. (2001) Interdisciplinarity: The New Critical Idiom, London: Routledge. Morrow, R. (2005) “Turning a trickle into a flood”, Asiamoney, 16 (9): 12–14. Mühlehner, J. (2005) “Der Begriff der Zinszahlung im EU-­Quellensteuergesetz”, Österreichische Steuer Zeitung: 76–83. Musgrave, P. B. (2006) “Combining fiscal sovereignty and coordination. national taxation in a globalizing world”, in I. Kaul and P. Conceição (eds.) The New Public Finance: Responding to Global Challenges, Oxford: Oxford University Press.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

190   Bibliography National Academy of Sciences, National Academy of Engineering and Institute of Medicine (2004) Facilitating Interdisciplinary Research, Washington, DC: National Academies Press. Neumayer, E. (2007) “Do double taxation treaties increase foreign direct investment to developing countries?”, Journal of Development Studies, 43: 1501–1519. Newell, W. H. (2001) “A theory of interdisciplinary studies”, Issue in Integrative Studies, 19: 1–25. Nowotny, E. and Zagler, M. (2009) Der öffentliche Sektor, Berlin and Heidelberg: Springer. Oberson, X. (2003) “The OECD Model Agreement on Exchange of Information: a shift to the applicant state”, Bulletin for International Fiscal Documentation, 57 (1) 14–17. OECD (1998) Harmful Tax Competition: An Emerging Global Issue, Paris: OECD. OECD (2000) Improving Access to Bank Information for Tax Purposes, Paris: OECD. OECD (2001) Corporate Tax Incentives for Foreign Direct Investment, Paris: OECD. OECD (2005) Model Tax Convention on Income and on Capital, Paris: OECD. OECD (2007) Fundamental Reform of Corporate Income Tax, OECD Tax Policy Studies 16, Paris: OECD. OECD (2008) Model Tax Convention on Income and on Capital, Paris: OECD. Ondrich, J. and Wasylenko, M. (1993) Foreign Direct Investment in the United States: Issues, Magnitudes, and Location Choice of New Manufacturing Plants, Kalamazoo, MI: W. E. Upjohn Institute for Employment. Ortmann-­Babel, M. (2007) “Unternehmensteuerreform 2008 in Deutschland”, Finanzjournal, 46 (7): 266–273. Overesch, M. (2006) “Transfer pricing of intrafirm sales as a profit shifting channel: evidence from German firm data”, ZEW Discussion Paper no. 06-084. Overesch, M. and Wamser, G. (2006) “German inbound investment, corporate tax planning, and thin-­capitalization rules: a difference-­in-difference approach”, Ifo Working Paper 37, Munich: Ifo Institute for Economic Research. Owens, J. and Bennett, M. (2008) “OECD model tax convention: why it works”, OECD Observer, 269. Online, available at: www.oecdobserver.org/news/fullstory.php/ aid/2756/OECD_Model_Tax_Convention.html. Pak, S. and Zdanowicz, J. (1994) “A statistical analysis of the U.S. merchandise trade data base and its uses in transfer pricing compliance and enforcement”, Tax Management Transfer Pricing Report, 3 (1): 50–57. Papke, L. E. (2000) “One-­way treaty with the world: the US withholding tax and the Netherlands Antilles”, International Tax and Public Finance, 7: 295–313. Pattenden, K. and Twite, G. (2008) “Taxes and dividend policy under alternative tax regimes”, Journal of Corporate Finance, 14 (1): 1–16. Picciotto, S. (1992) International Business Taxation: A Study in the Internationalization of Business Regulation, New York: Quorum. Pindyck, R. S. and Rubinfeld, D. L. (1998) Econometric Models and Economic Forecasts, 4th ed., New York: McGraw-­Hill/Irwin. Pistone, P. (2002) “An EU model tax convention”, EC Tax Review, 3: 129–136. Pistone, P. (2006) “The impact of European law on the relations with third countries in the field of direct taxation”, Intertax, 34 (5): 234–244. Polivanova-­Rosenauer, T. (2005) “Welche Finanzprodukte umfasst die EU-­ Quellensteuer?”, Zeitschrift für Gesellschafts- und Steuerrecht: 256–265. Popper, K. R. (1963) Conjectures and Refutations: The Growth of Scientific Knowledge, New York: Routledge & Kegan Paul. Prigogine, J. (1980) From Being to Becoming, San Francisco: Freeman.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Bibliography   191 Prokisch, R. G. (2005) “Steuerfragen des internationalen Unternehmenserwerbs, Deutscher Nationalbericht zum Thema II IFA-­Kongress Buenos-­Aires 2005”, Cahiers de Droit Fiscal International, 90b: 275–298. Quiñones, N. (2010) “Establecimiento permanente e imposición exclusiva de la empresa en el Estado de la fuente”, in A. Mazz and P. Pistone (eds.) Reflexiones en torno a un modelo latinoamericano de convenio de doble imposición, Montevideo: Fundación de Cultura Universitaria. Ramsey, F. P. (1927) “A contribution to the theory of taxation”, Economic Journal, 37 (145): 47–61. Reinganum, J. F. and Wilde, L. L. (1986a) “Equilibrium verification and reporting policies in a model of taxpayer compliance”, International Economic Review, 27 (3): 739–760. Reinganum, J. F. and Wilde, L. L. (1986b) “Settlement, litigation, and the allocation of litigation costs”, RAND Journal of Economics, 17 (4): 557–566. Richman (Musgrave), P. (1963) Taxation of Foreign Investment: An Economic Analysis, Baltimore: Johns Hopkins Press. Rixen, T. (2008) The Political Economy of International Tax Governance, Basingstoke, UK: Palgrave Macmillan. Rixen, T. (2010) “Why bilateralism (and multilateralism)? The political economy of avoiding international double taxation”, European Journal of International Relations, in print. Rixen, T. (forthcoming) “From double tax avoidance to tax competition: explaining the institutional trajectory of international tax governance”, Review of International Political Economy. Roin, J. (2008) “Tax treaties and tax competition”, Columbia Law School Tax Policy Colloquium. Online, available at: www2.law.columbia.edu/taxcolloquium (last accessed 25 May 2009). Rowohlt, H. and Simon, A. (1955) “A behavioral model of rational choice”, Quarterly Journal of Economics, 69 (1): 99–118. Rozycki, J. J. (1997) “A tax motivation for smoothing dividends”, Quarterly Review of Economics and Finance, 37 (2): 563–578. Ruding Report (1992) Report of the Committee of Independent Experts on Company Taxation, Luxembourg: Commission of the European Communities. Ruppe, H. (2005) Umsatzsteuergesetz 1994 Kommentar, 3rd ed., Vienna: WUV. Sanz-­Menendez, L., Bordons, M. and Zulueta, M. A. (2001) “Interdisciplinarity as a multidimensional concept: measure in three different research areas”, Research Evaluation, 10 (1): 47–58. Schilcher, M. (2009) Subject-­to-tax Klauseln in der österreichischen Abkommenspraxis, Vienna: Linde-­Verlag. Schmidt, L., Sigloch, J. and Henselmann, K. (2005) Internationale Steuerlehre: Steuerplanung bei grenzüberschreitenden Transaktionen, Wiesbaden: Gabler. Schmidtchen, D. (1973) “Vom nichtmarginalen Charakter der Steuermoral”, in C. Smekal and E. Theurl (eds.) Stand und Entwicklung der Finanzpsychologie, Baden-­Baden: Nomos-­Verlagsgesellschaft. Schmölders, G. (1951/1952) “Finanzpsychologie”, FinanzArchiv, 13: 1–36. Schmölders, G. (1960) Das Irrationale in der öffentlichen Finanzwirtschaft, Hamburg: Rowohlt. Scientific Council at the German Federal Ministry of Finance (1997–1998) Reforming International Capital Income Taxation. Online, available at: www.taxjustice.net/cms/ upload/pdf/Reform.income_tax.pdf.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

192   Bibliography Shaviro, D. (2002) “Exploring the need for international harmonization: money on the table? Responding to international tax arbitrage”, Chicago Journal of International Law, 3: 317–330. Sinn, H.-W. (1997) “The selection principle and market failure in systems competition”, Journal of Public Economics, 66: 247–274. Six, M. (2007) Hybride Finanzierung im Internationalen Steuerrecht am Beispiel von Genussrechten, Vienna: Linde. Six, M. (2009) “Hybrid finance and double taxation treaties”, Bulletin for International Taxation, 63 (1): 22–25. Skinner, D. J. (2007) “Comment”, in A. J. Auerbach, J. R. Hines and J. Slemrod (eds.) Taxing Corporate Income in the 21st Century, Cambridge: Cambridge University Press. Skinner, D. J. (2008) “The evolving relation between earnings, dividends, and stock repurchases”, Journal of Financial Economics, 87 (3): 582–609. Slemrod, J. and Yitzhaki, S. (2002) “Tax avoidance, evasion, and administration”, in A. J. Auerbach and M. Feldstein (eds.) Handbook of Public Economics, vol. 3, Amsterdam: Elsevier. Sørensen, P. B. (1995) “Changing views of the corporate income tax”, National Tax Journal, 48 (2): 279–294. Sørensen, P. B. (2004) “Company tax reform in the European Union”, International Tax and Public Finance, 11: 91–115. Speidel, R. and Widinski, M. (2007) “Überblick über die Unternehmenssteuerreform 2008 in Deutschland”, Österreichische Steuerzeitung, 18: 423–424. Springael, B. (2004) “News analysis: Belgium moves forward on EU savings tax directive”, Tax Notes International, 315–317. Srinivasan, T. N. (1973) “Tax evasion: a model”, Journal of Public Economics, 2 (4): 339–346. Starmer, C. (2000) “Developments in non-­expected utility theory: the hunt for a descriptive theory of choice under risk”, Journal of Economic Literature, 28: 332–382. Surrey, S. (1965) “International tax conventions: how they operate and what they accomplish”, Journal of Taxation, 23: 366–376. Tabak, L. (2004) Presentation at Convocation on Facilitating Interdisciplinary Research, Washington, DC, 29 January. Tanzi, V. and Zee, H. H. (2001) “Can information exchange be effective in taxing cross-­ border income flows?”, in K. Andersson, P. Melz and C. Silfverberg (eds.) Modern Issues in the Law of International Taxation, The Hague: Kluwer Law International. Tenore, M. (2010) “Neutralidad e imposición de las rentas pasivas”, in A. Mazz and P. Pistone (eds.) Reflexiones en torno a un modelo latinoamericano de convenio de doble imposición, Montevideo: Fundación de Cultura Universitaria. Thompson Klein, J. (2004) “Interdisciplinarity and complexity: an evolving relationship”, e:co, 6 (1–2): 2–10. Tiebout, C. (1956) “A pure theory of local expenditure”, Journal of Political Economy, 64 (5): 416–424. Tilly, R., Welfens, P. and Heise, M. (2007) 50 Years of EU Economic Dynamics: Integration, Financial Markets and Innovations, Heidelberg: Springer. Torgler, B. (2002) “Speaking to theorists and searching for facts: tax morale and tax compliance in experiments”, Journal of Economic Surveys, 16: 657–683. Torgler, B. (2003a) “Does culture matter? Tax morale in an East–West-­German Comparison”, FinanzArchiv, 59 (4): 504–528.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Bibliography   193 Torgler, B. (2003b) “Tax morale, rule-­governed behaviour and trust”, Constitutional Political Economy, 14: 119–140. Torgler, B. (2003c) “To evade taxes or not: that is the question”, Journal of Socio-­ Economics, 32 (3): 283–302. Torgler, B. (2004) “Tax morale in Asian countries”, Journal of Asian Economics, 15: 237–266. Torgler, B. (2005) “Tax morale in Latin America”, Public Choice, 122: 133–157. Torgler, B. and Murphy, K. (2004) “Tax morale in Australia: what factors shape it and has it changed over time?”, Journal of Australian Taxation, 7: 298–335. Torgler, B. and Schneider, F. G. (2004) “Does culture influence tax morale? Evidence from different European countries”, mimeo, Center for Research in Economics, Management and the Arts, Basel. Torgler, B. and Schneider, F. G. (2005) “Attitudes towards paying taxes in Austria: an empirical analysis”, Empirica, 32 (2): 231–250. Traversa, E. (2008) “The CCCTB and possible elements of a technical outline: the ‘switch-­over’ clause”, in M. Lang, P. Pistone, J. Schuch and C. Staringer (eds.) Common Consolidated Corporate Tax Base, Vienna: Linde. Tretter, B. (1974) “Die Steuermentalität: Ein internationaler Vergleich”, in G. Schmölders (ed.) Finanzwissenschaftliche Forschungsarbeiten 45, Berlin: Duncker & Humboldt. Tumpel, M. and Gläser, L. (2005) “EU-­Zinsenbesteuerung ab 1. 7. 2005”, Steuer und WirtschaftsKartei, 19: 850–856. Tversky, A. and Kahneman, D. (1992) “Advances in prospect theory: cumulative representation of uncertainty”, Journal of Risk and Uncertainty, 5 (4): 297–323. UNCTAD (2009) FDI Statistics. Online, available at: www.unctad.org. Van Dijkum, C. (2001) “A methodology for conducting interdisciplinary social research”, European Journal of Operational Research, 128: 290–299. van Eije, H. and Megginson, W. (2008) “Dividends and share repurchases in the European Union”, Journal of Financial Economics, 89 (2): 347–374. van Themaat, P. V. (1966) “Die Steuerharmonisierung in der EWG”, Der Betrieb, 19: 834–840. Vieira, E. S. and Raposo, C. C. (2007) “Signaling with dividends? The signaling effects of dividend change announcements: new evidence from Europe”, mimeo, Aveiro, Portugal. Online, available at: http://ssrn.com/abstract=1004523. Vogel, K. (1995) “Problems of a most-­favoured-nation clause in intra-­EU treaty law”, EC Tax Review, 4: 264–265. Vondra, K. (2008) “Economic perspectives of tax competition and coordination in the European Union”, dissertation, Wirtschaftsuniversität Wien. Wassermeyer, F. (2004) “Commentary on article 23A MA”, in F. Wassermeyer, M. Lang and J. Schuch (eds.) Doppelbesteuerung: OECD Musterabkommen, DBA Österreich– Deutschland: Kommentar, Munich: C. H. Beck. Weichenrieder, A. J. (2007) “Steuern und die Empirie der Unternehmensfinanzierung”, mimeo, Frankfurt: Johann-­Goethe University. Weichenrieder, A. J. (2009) “Profit shifting in the EU: evidence from Germany”, International Tax and Public Finance, 16: 281–297. Weichenrieder, A. J. and Windischbauer, H. (2008) “Thin capitalization rules and company responses”, CesIfo Working Paper no. 2456, Category 1: Public Finance, Munich: Ifo Institute for Economic Research. Wheeler, D. and Mody, A. (1992) “International investment location decisions: the case of U.S. firms”, Journal of International Economics, 33: 57–76.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

194   Bibliography Wiedermann-­Ondrej, N. (2007) Hybride Finanzierungsinstrumente in der nationalen und internationalen Besteuerung der USA, Vienna: Peter Lang Europäischer Verlag der Wissenschaften. Wiedermann-­Ondrej, N. (2008) “Tax treatment of revenue-­based payments”, International Tax Journal, 34 (2): 39–54. Wiesmann, U., Biber-­Klemm, W., Grossenbacher-­Mansuy et al. (2008) “Enhancing transdisciplinary research: a synthesis in fifteen propositions”, in G. H. Hadorn, H. Hoffmann-­Riem, S. Biber-­Klemm et al. (eds.) Handbook of Transdisciplinary Research, Dordrecht: Springer. Wildasin, D. E. (2002) “Tax coordination: the importance of institutions”, Swedish Economic Policy Review, 9: 171–194. Wilson, J. D. and Wildasin, D. E. (2004) “Capital tax competition: bane or boon”, Journal of Public Economics, 88: 1065–1091. Wittendorff, J. and Banner-­Voigt, E. (2000) “Taxation of hybrid instruments”, IBFD DFI: 3–8. Yitzhaki, S. (1974) “A note on income tax evasion: a theoretical analysis”, Journal of Public Economics, 3 (2): 201–202. Young, K. H. (1988) “The effects of taxes and rates of return on foreign direct investment in the United States”, National Tax Journal, 41 (1): 109–121. Zagler, M., Luptacik, M., Lang, M. and Horvath, T. (2008) Österreichs Ökosteuern und Ökoabgaben, project report for Wirtschaftskammer Österreich. Zimmermann, K. (2002) “European labour mobility: challenges and potentials”, De Economist, 153 (4): 425–450. Zodrow, G. R. (1991) “On the ‘traditional’ and ‘new’ views of dividend taxation”, National Tax Journal, 44 (4): 497–509. Zodrow, G. R. and Mieszkowski, P. (1986) “Pigou, Tiebout, property taxation, and the underprovision of local public goods”, Journal of Urban Economics, 19: 356–370.

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

Index

ability to pay 13, 14, 16, 34 academic disciplines 6 accountability 160 accounting science 12 administrative rules 164 agency 43 agency costs 102, 103, 104 agricultural policy 167 Allingham–Sandmo–Yitzhaki model 52 allocation of taxing powers 29, 30, 32–4 alternative approach 9 annual accounts 111 anti-avoidance rules 71 Article 10 of the Savings Directive 48 Article 26 of the OECD-MTC: Background 44–5; Key regulations 45–7 attorney–client privilege 56 audit 146 aviation 170 bankruptcy 145 beneficial owner 48 benefit theory 15, 16, 17 capital exporting countries 15 capital gains: taxation 107 capital–export neutrality 13, 15, 16, 29 capital–import neutrality 13, 15, 16, 29 capital–ownership neutrality 15, 16 cash flow 110 cash flow tax 139 central bank 176 CFC legislation 27 CFC rules 70, 76, 77, 94, 96 check the box 77, 78, 94, 96 climate charge on aviation 171, 172 CO2 emissions 169 Collaboration 10 common consolidated corporate tax base

(CCCTB) 79, 97, 138, 146, 154, 155, 156, 157 communication tax 173, 174 community law 148 competitiveness 145 complexity 3 compliance costs 156, 161 consolidated financial statements 111 control effect 170 controlled foreign corporation: rules 109 controlling shareholders 112 coordination 158 corporate finance 115 corporate income tax 79, 80, 154, 155, 156, 157 corporate taxes 106 country of residence 146 credit method 11, 13, 15, 18, 20, 27–31; foreign-source losses 24, 28; see also relief of double taxation cross-border transactions 115 crossdisciplinarity 7 deadweight loss of market power 150; loss of taxation 148 degree of concentration 168 degree of openness 149 destination-based tax 153 developing countries 11, 13–36; categorization 22, 34; direct aid 18, 32; tax incentives 14, 34 disciplinary research 3 distortionary effect 149, 157 distribution of tax burden 145 dividend policies 12 dividend repatriation 11 dividends: policy(ies) 98; repatriated 101; smoothing 98, 104; taxation 106 domestic tax policy 156

196   Index

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

domestic tax system 160 double non-taxation 61, 67, 69 double tax agreements 11 double tax convention 115 double tax treaties (DTT) 61 double taxation 62, 63, 64, 65, 67, 68, 69, 73, 80 ECJ (European Court of Justice) 14, 21; A 26; Cadbury Schweppes 23, 24; D 25, 116; Marks & Spencer 25; Saint-Gobain 25; Schumacker 25 economic distortions 147 economic turmoil 175 effective tax rates 137 effectiveness 172 efficiency 14, 15 electricity 169 Energy Directive 169 energy taxation 169, 170, 171 EU (/European) law 14, 21–33; EPAs 25–6; free movement of capital 26–7, 32–3; most favored nation 25–6 EU Parent–Subsidiary Directive 107, 108 EU personal income tax 159 EU standards on exchange of information on capital income 47–9 EU-wide tax 148 exchange of information in tax matters (econ. perspective) 50–1 excise duties on tobacco and alcohol 166, 167, 168 exemption method 11, 13, 18, 20, 27–31, 106; foreign-source losses 24, 28; see also relief of double taxation expected utility framework 51 export 141 export prices 145 fiduciary capacity 43 financial accounting 143 financial centre 174 financial instruments 115 financial market 158, 175 financial reasons 168 financial transaction 174 fiscal effects 146 fishing expeditions 44 foreign direct investment (FDI) 14, 16, 17, 19, 21; effects of taxes on 17, 17, 19, 20 foreseeable relevance 45 fraud 167 fuel 169

gross tax flows 147 harmful tax competition 39 harmonization 157, 169 harmonized personal income tax 161 harmonized rules 155 harmonized tax base 155 horizontal justice 167 hub 172 hybrid entities 66, 71, 72, 76, 77, 94 hybrid financial instruments 115 hybrid instruments 12 imperfect competition 151, 154 import 141 import prices 145 import turnover tax 142 income effect 149 incomplete ownership 112 individual tax evasion (econ. perspective): economics of crime 51–3; policy implications to combat tax evasion 53–4 inflation tax 175 information exchange 11 information upon request 42 interdisciplinarity 5, 6 interdisciplinary approach 2 interdisciplinary knowledge 2 interdisciplinary research 4, 5 international taxation 2 intra-community delivery 138 intra-community purchase 140 intra-community supply 140, 141 kerosene 171 Latin America 29–30, 32; Model LATC 33–4, 36 legal uncertainty of tax law, influences on financial decisions 115 manipulation by the taxpayer 165 market concentration 149 market for loanable funds 158 market power 150 matching credit 28 megadisciplinarity 8 model agreement on exchange of information: background 39–40; key regulations 40–4 modulated VAT 163, 164, 165 money 175 monopoly 176 multidisciplinarity 7

Index   197 multinational enterprise 11, 12

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

national tax sovereignty 2 negative tax base 142 negotiation process 10 non-residents 25–7 OECD Global Forum Working Group on Effective Exchange of Information 40 OECD model tax convention 25, 64, 66, 67, 68, 163 OECD principles on exchange of information: aim of the OECD principles 54–5; interim results 57–8; limits to exchange of information 57; refusal of exchange of information upon request 55–7 OECD standards on exchange of information 39–47 oil price 170 openness 152 ordre public 46 paper profits 62, 69, 70, 72, 78, 79 passive income 14, 16, 33–4, 109 per capita tax 159 permanent establishments 144 personal income tax 159, 160, 161, 162, 163 personal taxation 24, 25 philosophical knowledge 4 pitfalls 10, 148 price elasticity of demand 150 probability tree 115 problem-solving approach 3 profit margin 143 profit shifting 98 progressive tax rate 160 ratios 111 recursive process 10 relief of double taxation 13, 25, 27–31, 32; credit method 13, 15; exemption 13; tax sparing research, interdisciplinary 1, 8 resident 106 return to equity 157 revenue sharing 31, 33 Savings Directive: aims of the Savings Directive 58–9; background 47–8; equivalent measures in third countries 49; interim results 59; key regulations 48–9

segmented markets 158 seignorage 175 shareholder structure 112 signaling 102, 104 smuggling 167 social security contribution 161 source-based tax 153 statistical effects 145 statutory corporate income tax rate 134 subsidiary 98 subsidization of export prices 139 substitution effect 149 superdisciplinarity 6 surcharge 159, 162 tax audit 123 tax authority 162 tax base 164 tax burden 141 tax competition 1, 61, 62, 63, 69, 70, 72, 74, 77, 78, 80, 95, 96, 97 tax coordination 12 tax credit 142 tax deferrals 19, 20 tax distortion 150 tax evasion 51 tax fraud 11 tax harmonization 1 tax haven 19, 37, 38, 69, 70, 77, 94, 96, 97 tax holidays 17, 19 Tax Information Exchange Agreement 58 tax jurisdiction 13, 32–4; boundaries 13, 32–4 tax planning 174 tax policy 155 tax sparing 19, 28, 31; partial 21 tax treaties 23, 29, 31, 157, 162 taxable income 135, 143 taxing power, attribution 134 taxing rights 151, 153, 154 tax-related information exchange system 50 territorial taxation 13, 14, 15, 16, 18, 19, 20, 25, 29, 33 thin capitalization 62, 66, 70, 71, 72, 75, 76, 77, 78, 92, 93, 94, 96 third countries 2, 22, 26–8 Tobin tax 174, 175 transdisciplinarity 7 transfer of seignorage revenue 175, 176 transfer pricing 70, 72, 75, 76, 77, 78, 89, 90, 91, 92, 134, 135, 136 transition costs 145 transportation 173

198   Index unduly prevent or delay 41 United States 14, 18, 19, 20, 21; Bush tax reform 14, 19; Obama tax reform 14, 19, 20; tax reforms 14;

Downloaded by [INFLIBNET Centre] at 06:19 30 August 2012

VADCIT, practical implementation 142 value-added-type destination-based cash flow capital income tax (VADCIT) 135, 142

VAT revenues 165 virtues 10, 148 withholding tax 64, 67, 68, 95, 108, 142, 160 worldwide taxation 15, 16, 17, 19, 21, 25