A Common Consolidated Corporate Tax Base for Europe — Eine einheitliche Körperschaftsteuerbemessungsgrundlage für Europa [1 ed.] 3540794832, 9783540794837 [PDF]

Preface This book contains the proceedings of the International Tax Conference on the c- th th mon consolidated corporat

143 3 2MB

English-German Pages 184 [196] Year 2008

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Front Matter....Pages I-XVI
Concept and Necessity of a Common Tax Base — an academic introduction....Pages 1-47
Perspektiven der Konzernbesteuerung....Pages 49-84
CCCTB: Der steuerpflichtige Unternehmensgewinn — gemeinsame Strukturelemente....Pages 85-93
An optional and competitive CCCTB — a comprehensive approach to reach the Lisbon objectives —....Pages 94-100
Panel Statements....Pages 101-111
Evaluating the Common Consolidated Corporate Tax Base....Pages 113-127
Europe Slowly Lurches to a Common Consolidated Corporate Tax Base: Issues at Stake....Pages 128-138
Panel Statements....Pages 139-149
Lessons of US Subnational Experience for EU CCCTB Iniative....Pages 150-154
Administrative Aspekte einer Gemeinsamen Konsolidierten Körperschaftsteuerlichen Bemessungsgrundlage in der EU....Pages 155-166
Administrative Aspects of the CCCTB — a Business Perpective....Pages 167-172
Panel Statements....Pages 173-183
Papiere empfehlen

A Common Consolidated Corporate Tax Base for Europe — Eine einheitliche Körperschaftsteuerbemessungsgrundlage für Europa [1 ed.]
 3540794832, 9783540794837 [PDF]

  • 0 0 0
  • Gefällt Ihnen dieses papier und der download? Sie können Ihre eigene PDF-Datei in wenigen Minuten kostenlos online veröffentlichen! Anmelden
Datei wird geladen, bitte warten...
Zitiervorschau

A Common Consolidated Corporate Tax Base for Europe Eine einheitliche Körperschaftsteuerbemessungsgrundlage für Europa

Wolfgang Schön · Ulrich Schreiber Christoph Spengel Editors

A Common Consolidated Corporate Tax Base for Europe Eine einheitliche Körperschaftsteuerbemessungsgrundlage für Europa

123

Professor Dr. Wolfgang Schön Max Planck Institute for Intellectual Property, Competition and Tax Law Marstallplatz 1 80539 Munich [email protected] Professor Dr. Ulrich Schreiber Professor Dr. Christoph Spengel University of Mannheim Chair of Business Administration & Taxation and Center for European Economic Research (ZEW) Schloss 68131 Mannheim [email protected] [email protected]

ISBN 978-3-540-79483-7

e-ISBN 978-3-540-79484-4

DOI 10.1007/978-3-540-79484-4 Library of Congress Control Number: 2008926728 c 2008 Springer-Verlag Berlin Heidelberg  This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Production: le-tex publishing services oHG, Leipzig Coverdesign: WMXDesign GmbH, Heidelberg Printed on acid-free paper 987654321 springer.com

Preface

V

Preface This book contains the proceedings of the International Tax Conference on the common consolidated corporate tax base (CCCTB) that was held in Berlin on 15th – 16th may 2007. The conference was jointly organised by the German Federal Ministry of Finance, the Centre for European Economic Research (ZEW), Mannheim, and the Max Planck Institute (MPI) for Intellectual Property, Competition and Tax Law, Munich. More than 250 participants from all over Europe and other regions, scholars, politicians, business people and tax administrators, discussed the European Commission’s proposal to establish a CCCTB. Three panels of tax experts evaluated the common tax base with respect to structural elements, consolidation, allocation, international aspects and administration. The conference made clear that the CCCTB has the potential to overcome some of the most intriguing problems of corporate income taxation within the Common Market. Common tax accounting rules substantially reduce compliance and administrative costs. Consolidation of a group’s profits and losses effects crossborder loss compensation which removes a major tax obstacle for European crossborder investment. At the same time, tax planning with respect to financing and transfer pricing is pushed back within the European Union. Moreover, as far as the CCCTB applies, member states are able to remove tax provisions that are targeted at cross border tax evasion and that might be challenged by the jurisdiction of the European Court of Justice. In the course of the conference the severe problems of implementing a CCCTB became evident as well. The impending loss of member states’ tax sovereignty is one of the most important obstacles on the way to a CCCTB. Member states may see their corporate income tax base endangered by the agreed apportionment rule that divides the common tax base. Moreover, both the administration and the protection of the common tax base are major issues of concern. It is a highly controversial issue if the common tax base should come along with a common corporate income tax rate or at least with an agreed minimum tax rate. Given both the technical problems of a CCCTB and the noticeable reluctance of member states to agree on key elements of the CCCTB, alternative concepts of consolidation being more in line with existing and internationally accepted tax policies might have to be taken into consideration. The conference made a substantial contribution to the ongoing evaluation of the ambitious concept of a CCCTB. The proceedings give an insight into the many conceptual problems and technical details of establishing a common tax base in the European Union. The editors acknowledge the generous financial and personal support from the European Commission, the German Federal Ministry of Finance, MPI and ZEW with respect to the organisation of the International Tax Conference and the publication of the proceedings. Furthermore, we would like to take the opportunity to thank all conference speakers for their valuable contributions published in this book. Special thanks go to Silke Bruns, Rainer Wächter and Martin Umbach (all German Federal Ministry of Finance) for organising the International Tax Conference, to

VI

Preface

Sybille Haase, Martina Hapke, Antja Schiffl-Lenth, Wanda Sene, Gabriele Stricker (all German Federal Ministry of Finance), to Annina Schlueter (Mannheim University) and to Arne Friese and Stefan Mayer (Max Planck Institute) for their help in preparing this book.

Mannheim and Munich December 2007

Wolfgang Schön Ulrich Schreiber Christoph Spengel

Contents

VII

Contents A. Concept and Necessity of a Common Tax Base – an academic introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christoph Spengel 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Company Taxation in the European Union . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Country Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Corporate Tax Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Corporate Tax Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2.1 Starting Point: Relationship between Financial Accounting and Tax Accounting . . . . . . . . 2.1.2.2 Depreciation on Plant and Equipment . . . . . . . . . . . . 2.1.2.3 Recognition of Provisions . . . . . . . . . . . . . . . . . . . . . 2.1.2.4 Inter-company Dividends . . . . . . . . . . . . . . . . . . . . . . 2.1.2.5 Capital Gains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2.6 Loss Compensation . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2.7 Foreign Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2.8 Group Taxation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.3 Corporation Tax Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.4 Local Taxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Effective Tax Burdens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Purpose of the Survey and Outline of the Model . . . . . . . . . . 2.2.2 Domestic Investments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3 Cross-border Investments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Economic Perspective of the Common Market: Efficiency and Neutrality . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3 Perspective of Businesses: Obstacles to Cross-border Activities within the EU . . . . . . . . 2.3.4 Perspective of Member States Tax Administrations: Enforcement of Taxes and Protection of Tax Bases . . . . . . . 2.4 Is Harmonisation of Nominal Tax Rates a Way Out? . . . . . . . . . . . 3. Common Tax Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Implementation issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Determination of Taxable Income . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Determination of Group Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1 Group Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1.1 Taxable Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1.2 Nexus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1.3 Personal Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 2 2 2 3 3 5 6 6 7 8 9 12 13 15 16 16 18 20 23 23 23 24 24 25 26 26 28 30 30 31 31 31 33 33

VIII

Contents

4.2.2

Consolidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2.1 Starting Point . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2.2 Intra-group Loss Relief . . . . . . . . . . . . . . . . . . . . . . . 4.2.2.3 Intra-group Transactions . . . . . . . . . . . . . . . . . . . . . . 4.2.2.4 Scope of Income Qualifying for Consolidation . . . . . 4.2.2.5 Treatment of EU Outbound and Inbound Investments . . . . . . . . . . . . . . . . . . . . . . 4.2.3 Entry and Exit Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Apportionment of Group Income . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Related Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.1 Legal Obligation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.2 Local Profit Taxes, Non-profit Taxes and Social Security Contributions . . . . . . . . . . . . . . . . . . . . . 5. In Favour of a Minimum Corporate Tax Rate . . . . . . . . . . . . . . . . . . . . . . 6. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Perspektiven der Konzernbesteuerung . . . . . . . . . . . . . . . . . . . . . . . . . Wolfgang Schön 1. Einleitung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Ansprüche an eine Konzernbesteuerung . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Rechtliche Ansprüche . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Vorgaben des Grundgesetzes . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2 Vorgaben des EG-Vertrages . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Politische und ökonomische Ansprüche . . . . . . . . . . . . . . . . . . . . . 2.2.1 Politische Zielsetzungen im Konzernsteuerrecht . . . . . . . . . . 2.2.2 Entscheidungsneutralität und Ressourcenallokation . . . . . . . 2.2.3 Konzernneutralität als Zielsetzung des Unternehmenssteuerrechts . . . . . . . . . . . . . . . . . . . . . . . . 3. Sollbruchstellen im Konzernsteuerrecht . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Die Mehrfachbesteuerung von Gewinnen . . . . . . . . . . . . . . . . . . . . 3.1.1 Klassisches Körperschaftsteuersystem und Integrationsverfahren . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2 Der Abzug von Beteiligungsaufwand . . . . . . . . . . . . . . . . . . 3.2 Die Verrechnung von Verlusten . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Die Kontrolle von Schuldverhältnissen . . . . . . . . . . . . . . . . . . . . . . 3.4 Die zentrale Rolle von Finanzierungen . . . . . . . . . . . . . . . . . . . . . . 3.5 Die Vermeidung von Gewinnrealisierungen . . . . . . . . . . . . . . . . . . 4. Zum Verhältnis Betriebsstätten / Tochtergesellschaften . . . . . . . . . . . . . . 5. Das Konzept einer einheitlichen Konzernsteuerbemessungsgrundlage . . 5.1 Grundzüge und Stand des Konzepts . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Optionslösung oder Zwangsvereinheitlichung? . . . . . . . . . . . . . . . . 5.3 Internationale und nationale Konzerne . . . . . . . . . . . . . . . . . . . . . . 5.4 Die Definition der Unternehmensgruppe . . . . . . . . . . . . . . . . . . . . . 5.4.1 Beteiligungsschwelle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.2 Wirtschaftliche Integration der Tochtergesellschaft . . . . . . .

35 35 35 37 38 39 40 41 43 43 44 45 46 49 49 50 50 50 52 53 53 53 54 55 55 55 57 58 59 61 62 63 64 64 66 67 68 69 70

Contents

5.5 5.6 5.7 5.8 5.9 5.10

IX

Die Definition des steuerpflichtigen Einkommens . . . . . . . . . . . . . . Die Technik der Konsolidierung . . . . . . . . . . . . . . . . . . . . . . . . . . . Die Behandlung von Drittstaaten . . . . . . . . . . . . . . . . . . . . . . . . . . . Die Anwendung von Zuordnungsfaktoren . . . . . . . . . . . . . . . . . . . . Institutionelle Fragen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Schlusswort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

71 74 75 78 83 84

C. Taxable Corporate Income – Common Structural Elements . . . . . . .

85

I.

1. 2. 3. 4.

5.

CCCTB: Der steuerpflichtige Unternehmensgewinn – gemeinsame Strukturelemente . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gunter Mayr Einleitung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Problemstellung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ausgangssituation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Besteuerungsgrundsätze . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Realisationsprinzip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Imparitätsprinzip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Zurechnung von Wirtschaftsgütern und Leasing . . . . . . . . . . . . . . . 4.4 Einzel- oder Gruppenabschreibung? . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Rückstellungen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 Betriebsausgaben . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7 Eigene Steuerbilanz? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ergebnis und Ausblick . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

85 85 85 87 87 87 88 89 89 90 91 91 92

II. An optional and competitive CCCTB – a comprehensive approach to reach the Lisbon objectives – . . . . . . 94 Krister Andersson 1. Aims of the CCCTB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 2. Underlying tax principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 3. Important features of the CCCTB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 III. Panel Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Martina Baumgärtel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Christian Comolet-Tirman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Peter Essers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Malcolm Gammie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

101 101 104 107 109

D. Consolidation, Allocation and International Aspects . . . . . . . . . . . . . . 113 I.

Evaluating the Common Consolidated Corporate Tax Base . . . . . . . . 113 Ulrich Schreiber 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 2. The Commission’s Proposal: Formula Apportionment and Source Based Taxation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

X

Contents

2.1 Consolidation and Apportionment . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Investment Decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Tax Revenue Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Third Country-Tax Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. An Alternative to the Commission’s Proposal: Separate Entity Accounting and Residence Based Taxation . . . . . . . . . . 3.1 Separate Entity Accounting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Investment Decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Tax Revenue Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Third Country-Tax Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. The Way to European Group Taxation . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Common Corporate Income Tax Policy? . . . . . . . . . . . . . . . . . . . . . 4.2 Single-Stage Solution or Step-by-Step Solutions? . . . . . . . . . . . . . . 5. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

114 114 116 118 120 120 121 121 123 123 123 124 127

II. Europe Slowly Lurches to a Common Consolidated Corporate Tax Base: Issues at Stake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jack Mintz 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. What’s it all about? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. What are the Issues? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 The Tax Unit and Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Consolidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 The Formula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Non-Business Income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Inbound and Outbound Investment . . . . . . . . . . . . . . . . . . . . . . . . . 4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

128 128 130 130 131 132 134 135 135

III. Panel Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Mary Bennett . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Bruno Gibert . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Michael Lang . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Ivar Nordland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

139 139 141 144 148

IV. Lessons of US Subnational Experience for EU CCCTB Iniative . . . . Walter Hellerstein 1. Historical lessons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Political lessons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Technical lessons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

150

128

150 151 153 154

Contents

XI

E. Administrative Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 I.

1. 2. 3. 4. 5.

6. 7. 7.1

8. 9.

10.

11.

12.

Administrative Aspekte einer Gemeinsamen Konsolidierten Körperschaftsteuerlichen Bemessungsgrundlage in der EU . . . . . . . . Ernst Czakert Einleitung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Derzeitige Situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Angestrebte Situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Verwaltungsstrukturen und Abläufe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bestehende Verfahren . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Amtshilferichtlinie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Beitreibungsrichtlinie . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Entscheidungsmechanismen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anforderungen an ein Verwaltungssystem für die GKKB . . . . . . . . . . . . Grundsätzliche Möglichkeiten für eine Verwaltungsstruktur zur GKKB . Nationale Ebene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Zentrale Ebene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 Kombination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Idealtypische Verwaltungsstruktur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Idealtypisches Verwaltungsverfahren . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1 Erfassung/Registrierung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 Steuererklärung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3 Veranlagungsverfahren . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4 Betriebsprüfung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.5 Neufestsetzung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sicherstellung der gleichmäßigen Auslegung . . . . . . . . . . . . . . . . . . . . . . 10.1 Ständiger Ausschuss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2 Verfahren und Entscheidungsfindung . . . . . . . . . . . . . . . . . . . . . . . 10.3 Mechanismus zur Anpassung der Regeln . . . . . . . . . . . . . . . . . . . . . Rechtsbehelfsverfahren . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Außergerichtliches Verfahren . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Gerichtliches Verfahren . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fazit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

155 155 156 156 156 157 157 158 158 159 159 159 160 160 160 161 161 162 162 162 163 163 163 164 164 164 164 165 166

II. Administrative Aspects of the CCCTB – a Business Perpective . . . . . 167 Theo Keijzer III. Panel Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. María Teresa Soler Roch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Miklós Kok . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Matthias Mors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Philip Baker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

173 173 175 179 181

List of Contributors

Krister Andersson is Head of the Tax Policy Department, Confederation of Swedish Enterprise. Since 2005, he has been Chairman of the Working Group on Fiscal Affairs of BusinessEurope (formerly the UNICE), Brussels, and the chairman of their Task Force on CCCTB. He is an associate professor (Docent) in Economics at Lund University. He is a board member of the Swedish National Tax Authority, the Swedish IFA branch, and the Association of Industry Treasurers. He is also a member of the BIAC Taxation and Fiscal Policy Committee and the International Chamber of Commerce Tax Commission, in Paris. Philip Baker is a Queen’s Counsel in practice in London, and also a visiting fellow at the Institute of Advanced Legal Studies, London University. He has been responsible for masters level courses in International Tax Law and in European Community Tax Law. Dr. Martina Baumgärtel is Head of the “Group Tax Policy & Products” division of Allianz SE. Focal points of her activity since starting at Allianz in 1988 are the tax policy of the Allianz group and tax aspects in the development of insurance and bank products. In addition, Dr. Baumgärtel headed the tax department at Dresdner Bank. She is a member of various business associations and committees. Mary C. Bennett joined the OECD’s Centre for Tax Policy & Administration in Paris in 2005 as Head of the Tax Treaty, Transfer Pricing & Financial Transactions Division. Previously, she was a partner in the Washington, DC office of Baker & McKenzie LLP, where she specialized in international tax from 1990 to 2005. Between 1985 and 1990, she was the Deputy International Tax Counsel at the U.S. Treasury, and prior to that, she practiced international tax law with private firms in Boston and London. Christian Comolet-Tirman is Deputy Director of the French Tax Policy Directorate, Head of the “European Affairs and International Relations” Department. He has chaired the sub-group on taxable income of the European Commission’s CCCTB working group. He is also Chair of the OECD Forum on Harmful Tax Practices and Member of the UN Committee of Experts on Cooperation in International Tax Matters. Ernst Czakert is Head of Division in the Directorate-General IV (Taxation) of the Federal Ministry of Finance in Berlin. In this capacity, he takes part in the European Commission’s CCCTB Working Group.

XIV

List of Contributors

Professor Dr. Peter Essers is Professor of Tax Law at Tilburg University, the Netherlands, Director of the European Tax College and Chairman of the Finance Committee of the Senate of the Netherlands Parliament. Malcolm Gammie QC is in practice at the Revenue Bar in London and is Research Director for the Tax Law Review Committee of the Institute for Fiscal Studies. Bruno Gibert is since 2001 Partner at CMS Bureau Francis Lefebvre, Paris. He was previously Director International Division with the French Tax Authorities. He chairs the European Union Joint Transfer Pricing Forum since its creation in 2002. Walter Hellerstein is the Francis Shackelford Professor of Taxation at the University of Georgia Law School. He is co-author (with his late father) of the twovolume treatise, State Taxation (Warren Gorham & Lamont, 3rd ed., updated semiannually), the casebook, State and Local Taxation (West Publishing Co., 8th ed. 2005), and more than 90 articles on state taxation. He has also practiced extensively in the state tax field and has been involved in numerous state tax cases before the United States Supreme Court. Theo Keijzer is Vice President Tax Policy at Shell International B.V. in the Netherlands. He obtained his Master Degree in Business Economics at the Erasmus University Rotterdam and since worked for Shell in various countries and tax positions for more than 30 years. He chairs the VNO-NCW International Tax Committee and is vice chair of the EU Joint Transfer Pricing Forum and vice chair of the International Chamber of Commerce Tax Committee; he is also a member of the Tax Committee of BusinessEurope as well as BIAC. Since 2004, Miklós Kok has been the Director of the International Department of the Hungarian Tax and Financial Control Administration (APEH). From 1997 to 2004, he worked, in various positions, for the Intra-European Organisation for Tax Administrations, a non-profit inter-governmental organisation, which provides a forum to assist members in the European Countries in improving tax administrations. In his current position, he is in charge of coordinating European and international relations of APEH. Prof. Dr. Michael Lang is the director of the Institute for Austrian and International Tax Law, Vienna University of Economics and Business Administration (WU), and the academic director of the WU’s LLM programme in International Tax Law. He is the president of the Austrian country group within the IFA and member of the Permanent Scientific Committee of global IFA. Furthermore, he is chairman of the Academic Committee of the European Association of Tax Law Professors (EATLP). DDr. Gunter Mayr qualified as a university lecturer at Innsbruck University and then transferred to the Ministry of Finance in Vienna. Since 2006, he has been head of the Directorate VI/6 (Income and Corporation Tax).

List of Contributors

XV

Jack M. Mintz is the Palmer Chair of Public Policy at the University of Calgary. He has consulted widely with governments and international organizations on tax policy and served as the chair of the Technical Committee on Business Taxation in Canada that led to recent tax reforms in Canada. Matthias Mors joined the European Commission’s Directorate-General for Taxation and Customs Union as Head of Unit responsible for the analysis and coordination of tax policies in 1999. The unit has been responsible for preparing the 2001 study on Company Taxation in the Internal Market as well as for the preparation of various targeted and comprehensive initiatives to overcome the remaining tax obstacles to the smooth functioning of the Internal Market. Since July 2005, he is Principal Advisor to the Director-General. He is also chairing the European Commission’s CCCTB Working Group. Ivar Nordland has been with the Danish Ministry of Taxation since 1985. Since 1992, he is responsible for international direct taxation, including EU, OECD and DTA matters. Wolfgang Schön is since 2002 a Director at the Max Planck Institute for Intellectual Property, Competition and Tax Law (Department of Accounting and Taxation) in Munich and Honorary Professor at Munich University. He is Chairman of the Scientific Council to the German Tax Law Association and a Member of the Permanent Scientific Committee of IFA as well as a Member of the Board of Trustees of the International Bureau of Fiscal Documentation. He was appointed Anton Philips Professor at Tilburg/NL University’s Centre for Company Law for the Academic Year 2004/2005 and was invited to join the Global Law Faculty at NYU in 2006. Prof. Dr. Ulrich Schreiber holds a chair of Business Administration and Taxation at the University of Mannheim. He is research professor at the Centre of European Economic Research (ZEW) and appointed member of the Advisory Board to the Federal Ministry of Finance. Ulrich Schreiber is specialised in the fields of international taxation and tax harmonisation in the European Union. María Teresa Soler Roch is Professor of Tax Law at the University of Alicante. She has been visiting Professor at the Universities of Florida (2001) and Paris II (2007-2008). Her research is focused in domestic, comparative and EU Tax Law with books such as: Family Taxation in Europe (editor) and Tax Law in Spain published by Kluwer. Her recent line of research has been on ECJ doctrine on State Aid provisions applying to selective tax measures. She is a member of the Board of Trustees of the International Bureau for Fiscal Documentation, the Permanent Scientific Committee of the International Fiscal Association and the European Association of Tax Law Professors, as well as the Council for the Taxpayer’s Defense (Spanish Ministry of Finance) and the Working Group set up for the study of the CCCTB (Institute for Fiscal Studies, Spanish Ministry of Finance).

XVI

List of Contributors

Christoph Spengel is a Professor of International Taxation at the University of Mannheim. His main research interests are in the areas of international taxation (tax planning, comparison of company tax burdens), company taxation in the European Union (tax harmonization, group taxation, profit allocation) and financial and tax accounting. He is a member of the Scientific Council of the Federal Ministry of Finance, Visiting Professor at the Center for European Economic Research (ZEW), Mannheim, International Research Fellow at Oxford University, Saïd Business School, Academic Director of the Executive Master of Accounting and Taxation at Mannheim Business School and has been appointed as a member of the expert group on company taxation by the European Commission. Christoph Spengel also regularly teaches in German and foreign MBA programmes and is a member of the MBA Faculties for International Taxation at the Universities of Freiburg, Hamburg and Vienna.

A. Concept and Necessity of a Common Tax Base – an academic introduction

1

A. Concept and Necessity of a Common Tax Base – an academic introduction Christoph Spengel

1. Introduction The European Single Market aims to promote the convergence of economic performance (Article 2 EC Treaty), to protect economies against distortions of competition (Article 3 (1) EC Treaty) and to ensure an efficient allocation of resources (Article 98 EC Treaty). The Union’s own political objectives follow these guidelines. The strategic goal announced at the Lisbon summit in 2000 is to “become the most competitive and dynamic knowledge-based economy in the world by 2010.”1 In the meantime, the Union reaffirmed its commitment to this objective several times.2 Since company taxation has not yet been harmonised within the EU, each member state may autonomously decide within the legal limits of the EC Treaty on the number of taxes, the level of tax rates, the rules for the determination of tax bases, and the type of corporation tax system on an individual basis. Regarding crossborder business activities, the distinction between unlimited and limited liability to tax is accepted and the principles for the taxation of foreign source income laid down in the OECD model convention on the avoidance of double taxation (i.e. residence and source principle) are recognised. It is needless to stress that a great diversity of company taxation currently exists within the EU. The differences among member states with regard to company taxation cause several distortions and obstacles to EUwide business activities. They impede cross-border investment in the Single Market, creating distortions with regard to the type, the location and the source of finance of an investment. Another effect of considerable tax differentials within the EU is that book profits tend to be shifted to member states with low tax rates. This results in conflicts between member states over the distribution of tax revenues. In order to reduce or even eliminate these distortions, the European Commission already in 2001 has suggested introducing a Common Consolidated Corporate Tax Base (CCCTB) for the EU-wide activities of multinationals.3 A proposal for a directive should be released till the end of 2008.4 Up until now, two different degrees of a common tax base seem to be reasonable:5 The minimum degree is a harmonised tax base which should be based on a single set of tax accounting principles and includes a common practice for offsetting losses. According to the European Commission 1 2

3

See point 5 of the Presidency Conclusions, Lisbon European Council, 23-24.3.2000. See European Commission, Communication from the Commission to the European Parliament, the Council and the European Economic and Social Committee, COM(2007) 223 final, 2.5.2007. See European Commission, Company Taxation in the Internal Market, 2001.

2

Christoph Spengel

and others, a harmonised tax base within the EU could be established by using the International Financial Reporting Standards (IFRS) as a starting point for defining the tax base.6 Following the original idea of a CCCTB, however, a consolidated tax base has to be established which should be allocated to the different member states according to a formula. A CCCTB raises a lot of questions which will be discussed on three panels during the following two days. The aim and the purpose of this introductory presentation are as follows: First, it provides an overview on the different elements of the company tax regimes in the 27 member states (i.e. tax rates, tax bases, taxation of cross-border investment, corporation tax systems, and local taxes), estimates of the consequences on the effective levels of company tax burdens for both domestic and cross-border business activities within the EU, and an evaluation of the main findings with respect to the most relevant taxation requirements in the EU (Section 2). The results of the first part of this presentation therefore provide arguments for the necessity of a common tax base. The second part of the presentation describes the concept of a CCCTB and highlights the mayor advantages of a CCCTB compared to prevailing approaches for the taxation of cross-border business activities (Section 3). The third part of this presentation (Section 4) then deals with the most relevant implementation issues of a CCCTB which will be discussed in much more detail on the three conference panels. Finally, some considerations with respect to tax competition under a CCCTB and their impact on tax rates are provided (Section 5). The final section concludes (Section 6).

2. Company Taxation in the European Union 2.1 Country Practice 2.1.1 Corporate Tax Rates The overview on the different elements of the company tax regimes in the 27 member states begins with a comparison of the aggregated nominal tax rates on corporate profits. The comparison covers corporation tax rates and surcharges as well as local profit taxes and their interrelations with the corporation tax (i.e. the deductibility as a business expense from the base of the corporation tax).

4

5

6

See European Commission, Communication from the Commission to the Council, the European Parliament and the European Economic and Social Community, Implementing the Lisbon Programme: Progress to date and next step towards a Common Consolidated Corporate Tax Base (CCCTB), COM(2006) 157 final, Brussels, 5.4.2006. See European Commission, Communication from the Commission to the European Parliament, the Council and the European Economic and Social Committee, COM(2007) 223 final, 2.5.2007. See European Commission, Company Taxation in the Internal Market, 2001, p. 399; Oestreicher/Spengel, Recht der Internationalen Wirtschaft, 2001, pp. 889-902; Spengel, European Taxation 2003, pp. 253-256; Schön, European Taxation, 2004, pp. 432-440.

3

A. Concept and Necessity of a Common Tax Base – an academic introduction

Table 1: Nominal tax rates on corporate profits (federal and local level, 2007, per cent) Austria

25

Latvia

15

Belgium

34

Lithuania

15

Bulgaria

10

Luxembourg

29.6

Cyprus

10

Malta

35

Czech Republic

24

Netherlands

25.5

Denmark

28

Poland

19

Estonia

0

Portugal

27.5

Finland

26

Romania

16

France

34.4

Slovakia

19

Germany

38.7

Slovenia

23

Greece

25

Spain

32.5

Hungary

17.5

Sweden

28

Ireland

12.5

United Kingdom

30

Italy

37.3

The comparison of corporate tax rates reveals a great diversity within member states. Currently, nominal tax rates vary between zero (Estonia) and 38.7 (Germany) per cent (Table 1).7 On average, the nominal tax rate amounts to 22.41 per cent. Taxrate competition is in full swing within the EU. Among the 10 member states on top of the country ranking we find 9 accession countries and Ireland. All of these countries levy tax rates below 20 per cent. Austria also has a comparatively low rate of 25 per cent. The Netherlands lowered its corporation tax rate to 25.5 per cent in 2007; Germany will reduce the nominal tax rate on corporate profits to below 30 per cent in the course of the Company Tax Reform Act for 2008. Practitioners often measure the company tax burden by the nominal tax rate. Although nominal tax rates have an important signal function and also may be relevant for the decision of where to locate international mobile activities8 they are not at all useful estimates for the tax burden of real (productive) investment as the effects of the tax bases are omitted. 2.1.2 Corporate Tax Bases 2.1.2.1 Starting Point: Relationship between Financial Accounting and Tax Accounting With respect to the determination of taxable income stemming from business activities currently a broad conformity exists with respect to the basis, the so-called starting point. In all member states, the determination of taxable income is based on financial profits or losses, reference or adjustments are either made to domestic 7 8

See European Tax Handbook, Amsterdam 2007, and own calculations. E.g. financing structures, administration, coordination and distribution centres, European Headquaters. See Devereux, EC Tax Review 1992, pp. 105-117 for empirical evidence.

4

Christoph Spengel

GAAP (Generally Accepted Accounting Principles) or to IFRS (International Financial Reporting Standards). It has to be noted that no member state makes reference to a distinct set of tax accounting rules (see Figure 1).9 Reference / Adjustments to IFRS

Reference / Adjustments to domestic GAAP

Cyprus, Estonia, France, Ireland, Malta, Slovakia, Slovenia, United Kingdom

Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, United Kingdom

Distinct tax accounting rules

Figure 1: Relationship between financial accounting and tax accounting

There will be an increasing relevance of IFRS in the future, since a Regulation from 2002 obliges listed EU parent companies to draw up their consolidated accounts in accordance with IFRS since 2005 (2007 for companies which draw up their accounts in accordance with US-GAAP).10 For this reason, the 4th and 7th EC-Directive are adjusted to IFRS.11 In this context, the Commission is recommending member states to provide for the application of IFRS both on consolidated accounts of non-listed parent companies and on individual financial accounts. From the above findings one can conclude that a broad common basis of objectified rules for tax accounting already exists among member states. In particular, this is valid, for example, for documentation, framework and system principles (such as going concern and item-by-item valuation) as well as for particular principles of accounting (such as the definition of assets and liabilities). Moreover, in all member states profit taxation is based on the realisation principle. As a general rule, there-

9

10

11

Unless otherwise indicated, information about the determination of the tax bases was taken from Endres/Oestreicher/Scheffler/Spengel, The Determination of Corporate Taxable Income in the EU Member States, Aalphen 2007. See Regulation (EU) 1606/2002 by the European Parliament and the Council regarding the application of International Accounting Standards, OJ L 243/1. See Proposal for a Directive of the European Parliament and of the Council amending Council Directives 78/660/EEC, 83/349/EEC and 91/674/EEC on the annual and consolidated accounts of certain types of companies and insurance undertakings, COM(2002) 259/2 final.

A. Concept and Necessity of a Common Tax Base – an academic introduction

5

fore, unrealised profits are not recognised neither in financial accounts nor in member states’ tax bases. By contrast, member states’ tax bases are still far from being uniform. This can be traced back to two reasons: First, even when the concept of taxable income is based on domestic GAAP or on IFRS, the impact of financial profits and losses on the tax bases varies to a considerable extent among member states. Different techniques provide for strong dependency in a smaller number of countries, whereas there is no formal dependency in most member states, but linkage between accounting profits and taxable income. Thus, there exists considerable variation in a number of provisions for the determination of taxable income (e.g. depreciation on assets and recognition of reserves). Second, a considerable number of provisions for determining the tax base exist for which financial accounting rules provide no solution. This is particularly true for loss relief, the taxation of dividend income and the taxation of capital gains. The following deals briefly with these two categories of variations. 2.1.2.2 Depreciation on Plant and Equipment One example for a great divergence in the taxable base is depreciation practice for assets. Figure 2 provides an overview on the regular depreciation on plant and equipment in the member states. Straight-line over the useful life

Straight-line with specified rates

Straight-line or declining balance with specified rates

Straight-line or accelerated with specified rates

Declining balance (pool) with specified rates

IFRS, Estonia, Netherlands

Austria, Bulgaria, Cyprus, Hungary, Ireland, Italy, Malta, Slovenia

Belgium, France, Germany, Greece, Lithuania, Luxembourg, Portugal, Romania, Spain, Sweden

Czech Republic, Poland, Slovakia

Denmark, Finland, Latvia, United Kingdom

Figure 2: Regular depreciation on plant and equipment

In the majority of the member states, plant and equipment in accordance with IFRS can be depreciated on a straight-line basis. For tax purposes, however, depreciation on plant and equipment is linked closely to investment incentives. This explains why more than one half of all member states apply – mandatory or optional – the declining-balance method. Moreover, there is a considerable variation between the depreciation rates. Currently, they range for machinery from 3 to 30 per cent in case of the straight-line method and from 9 to 48 per cent in case of the declining-balance method. Four member states also apply the declining-balance method on a pool basis. The pool method stands for simplification and objectivity since depreciation rates do not have to be set for each asset on an individual basis. On the other hand, there is no possibility of reflecting individual useful lives.

6

Christoph Spengel

2.1.2.3 Recognition of Provisions In contrast to regular depreciation, the rules for the recognition of provisions in financial accounting are more generous compared to tax accounting. IFRS define provisions as liabilities of uncertain timing or amount. A provision should be recognised when (1) an enterprise has a present obligation as a result of a past event, (2) it is more likely than not that an outflow of resources embodying economic benefits will be required to settle the obligation, and (3) a reliable estimate can be made of the amount of the obligation. Permitted with certain exceptions

Disallowed (general rule) with certain exceptions

Disallowed or not tax-effective

IFRS, Austria, Belgium, Estonia, France, Germany, Ireland, Luxembourg, Netherlands, Romania

Denmark, Czech Republic, Finland, Greece, Hungary, Italy, Lithuania, Portugal, Slovakia, Spain, Sweden, United Kingdom

Bulgaria, Cyprus, Latvia, Malta, Poland, Slovenia

Figure 3: Recognition of provisions

According to member states’ practice (Figure 3), as a general rule, only nine countries permit provisions for tax purposes. Most member states do not allow provisions apart – in some countries – from exceptional cases. There is also a considerable variation among member states. Whereas all of the older member states permit provisions for tax purposes if certain objective facts similar to IFRS are fulfilled, one half of the new member states do not allow provisions at all. Therefore, on average, we should expect a tax base that is considerably broader in the new compared to the old member states. One has to be aware, however, that this broader tax base is normally accompanied by lower nominal tax rates (see above, 2.1.1). 2.1.2.4 Inter-company Dividends The most relevant tax provisions for which financial accounting provides no solutions cover the taxation of dividend income, the taxation of capital gains, and the relief for losses. As a common rule, corporations are taxed separately from their shareholders. In this respect, tax law follows civil law, which treats corporations as separate legal entities. Accordingly, profits are taxed at the corporate level when earned and the taxation at the shareholder level is deferred in case profits are retained. Shareholders are only subject to taxation, if the profits are distributed. This system of corporate taxation may result in a double taxation of distributed profits by imposing both corporation tax on the company level and corporation or income tax on the shareholder level. A distinction therefore has to be drawn between dividends distributed to corporate taxpayers and dividends distributed to natural persons. In both cases it is assumed, that the distributing company and its shareholder are resident in the same

A. Concept and Necessity of a Common Tax Base – an academic introduction

7

country. The following deals with corporate shareholders. The taxation of cross-border profit transfers (see below, 2.1.2.7) and the taxation of natural persons as shareholders depending on the type of the corporation tax system are considered later (see below, 2.1.3).

Exemption 100% Austria, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, Greece, Hungary, Ireland, Latvia, Lithuania, Luxembourg, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Sweden, United Kingdom

Tax Credit 95%

Belgium, France, Germany, Italy

Malta, Spain

Not applicable for IFRS

Figure 4: Taxation of inter-company dividends (domestic investments)

There are two different methods of avoiding double taxation of dividends distributed to corporations: (1) exemption and (2) tax credit. The vast majority of the member states follows the exemption method. This method exempts inter-company dividends in the hand of corporate shareholders. Profits are only subject to corporation tax when earned by the distributing company. Consequently, retained and distributed profits are burdened with an equal charge. Belgium, France, Germany and Italy also apply the exemption method, but with the variation of charging 5 per cent of the dividends received to tax. In some member states a certain minimum shareholding and/or a minimum period of ownership is required for the exemption method to apply. Malta and Spain are the only countries that apply the tax credit method to avoid double taxation of dividends. Dividends received by the shareholder are grossed-up by the underlying corporation tax and taxed. At the same time, the corporation tax paid by the distributing company is credited against the corporation tax of the dividend receiving parent company. As a result, corporation tax on distributed profits is levied only once. In Spain, full imputation is only granted if the receiving company has had a direct or indirect participation of at least 5 per cent in the capital of the subsidiary for a continuous period of at least one year. Otherwise only one half of the corporation tax attributable to the gross dividend is credited. 2.1.2.5 Capital Gains With regard to capital gains taxation, distinction has to be made between capital gains upon the disposal of shares and capital gains realised from the sale of other business assets. In general, the taxation of capital gains upon the disposal of shares should follow the concept of dividend taxation (i.e. avoidance of double taxation).

8

Christoph Spengel

Therefore, most member states exempting dividends from corporation tax also exempt such capital gains. However, there are some exceptions; in particular no relief is granted in the Czech Republic, Ireland, Lithuania, Poland, Slovakia, and Slovenia. France applies a special (i.e. lower) capital gains tax rate. Exemption, indexation, roll-over relief

Cyprus

Indexation, Roll-over relief

Portugal

Partial exemption, indexation

Malta, Spain, United Kingdom

Roll-over relief

Belgium, Denmark, Germany, Luxembourg, Netherlands

Special tax rates

France, Greece

Deferment of taxation

Estonia, Italy

No relief

Not applicable for IFRS

Austria, Bulgaria, Czech Republic, Finland, Hungary, Ireland, Latvia, Lithuania, Poland, Romania, Slovakia, Slovenia, Sweden

Figure 5: Taxation of capital gains (other than shares)

For capital gains arising on the sale of business assets other than shares (Figure 5) it is apparent that about one half of the member states do not grant any relief (i.e. these capital gains are taxes in the same way as ordinary income). In the other member states, various different forms of relief exist. Relief can be granted with respect to the tax base (i.e. full or partial exemption, indexation allowance, and roll-over relief) or to the tax rate (i.e. reduced tax rate). The application of a specific preferential tax treatment often depends on the type of asset and on the fulfilment of certain conditions. Most member states concentrate on the tax base, despite full exemption, however, a reduction of the tax base only results in a deferment of the tax charge. 2.1.2.6 Loss Compensation Accounting for potential losses is a common principle of financial accounting and taxation practice. The lower of cost or market valuation is accepted by both IFRS and taxation practice in most member states. However, considerable variation exists in the fields of provisions and the recognition of bad debts. While these items are recognised under IFRS and affect income as expenses, they are limited and restricted for tax purposes in most member states. In particular, this is true with respect to the recognition of provisions (see above, 2.1.2.3). The limitations and restrictions of the recognition of losses in tax accounting, however, are compensated by offsetting losses for tax purposes.

9

A. Concept and Necessity of a Common Tax Base – an academic introduction

carry back

carry forward

unlimited

Austria Belgium Cyprus Denmark France Germany Hungary Ireland Luxembourg Malta Netherlands Sweden United Kingdom

limited

limited

15 years

10 years

6 years

5 years

3 years

1 year

Spain

Finland

Portugal

Bulgaria Czech Republic Greece Italy Latvia Lithuania Poland Romania Slovakia Slovenia

France Netherlands

Germany Ireland United Kingdom

Not applicable for IFRS

Figure 6: Loss compensation

As a general rule there are no restrictions on the offset of profit and losses within the same period, at least where the losses were incurred in the ordinary course of business. Essentially, the carry back and carry forward of losses is an important element of the tax base since they link the tax bases of different periods. In all member states, in the field of inter-period loss compensation, different rules and various limitations exist (see Figure 6). All member states allow the carry forward of losses. However, various limitations exist. In eleven countries losses can be carried forward only for a certain period varying between five and fifteen years. Among those member states which allow for an unlimited loss carry forward, some have introduced a kind of minimum taxation which limits the amount of the loss offset in any given subsequent period to a certain percentage of the profits of that period (e.g. Austria, Germany and Poland). In contrast to the loss carry forward, a loss carry back is only available in France, Germany, Ireland, the Netherlands and the United Kingdom. Moreover, losses can be carried back only for a period of one or three years. In Germany, the amount of losses that can be set off against profits of the preceding period is limited to € 511,500. 2.1.2.7 Foreign Income Personal and corporate income tax, as a general rule, is based on the residence principle. It is the worldwide income whether generated at home or abroad which is subject to tax in the country of residence. Residence is defined as the domicile in the event of individual persons and as the seat or the place of management in the event of corporations. If part of the worldwide income is earned abroad a (limited) tax

10

Christoph Spengel

credit is granted for the underlying foreign taxes in order to avoid or mitigate double taxation on foreign income. With respect to the taxation of foreign income, the residence principle is restrained in the event of foreign direct investments. Direct investments can be carried out either through foreign subsidiaries or through permanent establishments located abroad. For both legal forms of direct investment, the source country is entitled to tax business profits, if these profits belong to a subsidiary or, in the event of a permanent establishment, are attributable to a permanent establishment located therein. Therefore, first of all, the tax burden on direct investment is determined by the level of foreign taxes (= source principle). In depends on the legal form of direct investment and the home country’s rules for the taxation of foreign income whether the source principle is maintained or the residence principle will be applied (Figure 7). Dividends of Subsidiary

Permanent Establishment

Exemption

Tax credit

Exemption

Tax credit

Tax credit

Austria Belgium Bulgaria Cyprus Czech Republic Denmark Estonia Finland France Germany Hungary Italy Latvia Lithuania Luxembourg Netherlands Portugal Romania Sweden Slovakia Slovenia Spain

Greece Ireland Malta Poland United Kingdom

Austria Belgium Bulgaria Cyprus Denmark Estonia France Germany Greece Hungary Luxembourg Netherlands Poland Romania Spain

Czech Republic Finland Ireland Italy Lithauen Latvia Malta Portugal Sweden Slovenia Slovakia United Kingdom

EU Member States

Source Principle

Residence Principle

Source Principle

Residence Principle

Other Income

Residence Principle

Figure 7: Taxation of foreign income

– In the event of a foreign subsidiary, taxation according to the source principle is always maintained, if the subsidiary retains its profits. Due to the deferral principle no income is allocated to the shareholders. Moreover, even in the event of an ultimate repatriation of profits, the source principle is upheld, if the home country of the parent company exempts dividends from the domestic corporation tax base. For dividend income, the exemption method currently is prevailing in 22 member states. Only five member states tax foreign dividends and grant a (limited) tax credit for the underlying (foreign) corporation tax which results in a taxation according to the residence principle (i.e. it is at least the domestic tax that determines the tax burden on foreign profits).

A. Concept and Necessity of a Common Tax Base – an academic introduction

11

– In the event of a foreign permanent establishment, the source principle is upheld, if the home country of the parent company exempts foreign business profits from the domestic corporation tax base. The exemption method currently is prevailing in 15 member states; in the remaining 12 member states the credit method is applied which results in a taxation according to the residence principle. The coexistence of the residence principle and the source principle with regard to the taxation of foreign income has at least three consequences: (1) If a direct investment is carried out through a subsidiary and profits are retained or if the home country of the parent company exempts dividends or profits of foreign permanent establishments from the domestic corporation tax base, domestic investments and foreign direct investments are suffering different effective tax burdens. This might cause distortions with respect to the location of investments. (2) Under same conditions as described above (1), the exemption of profits from foreign direct investments offers considerable tax planning opportunities to multinational investors. Since foreign income other than business profits or profits of a subsidiary are always taxable in the home country of the parent company and a tax credit is granted for a foreign withholding tax (i.e. the residence principle applies, see Figure 7), a multinational investor can – in principle – decide whether to pay lower foreign or lower domestic taxes on cross-border investments. A relevant field for such profit shifting is the area of inter-company financing.12 In the case of equity financing, the tax burden at the level of the foreign subsidiary is the relevant burden on foreign direct investments. In the case of debt financing, interest paid by a foreign subsidiary is – as a general rule – deductible from the profits of the subsidiary and subject to corporation tax at the level of the parent company. Consequently, foreign profits shifted via debt financing bear the same tax burden as domestic investments of the parent company. As a general rule, therefore, equity financing is more tax-efficient than debt financing if foreign taxes are lower than domestic taxes (and vice versa). (3) The incentive of multinational investors to allocate profits preferably to low tax jurisdictions causes problems for the tax authorities above all in high tax countries to collect taxes from cross-border business transactions. The coexistence of the residence principle and the source principle is also decisive for different concepts prevailing in the member states with respect to relief for losses from foreign permanent establishments (Figure 8).13

12

13

See ZEW/Ernst & Young, Company Taxation in the New EU Member States, 2nd edition, Frankfurt/Mannheim 2004, pp. 23-26. Profit shifting by transfer prices results in similar effects. See European Commission, Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee, COM(2006) 824 final, 19.12.2006; European Commission, Commission Staff Working Document, SEC(2006) 1690, 19.12.2006.

12

Christoph Spengel

Losses of a Permanent Establishment

Credit method

Yes

Czech Republic, Finland, Ireland, Italy, Latvia, Lithuania, Malta, Portugal, Slovakia, Slovenia, Sweden, United Kingdom

Exemption method

Deduction and recapture

No deduction and no recapture

Yes

No

Austria, Belgium, Cyprus, Netherlands, Spain

Bulgaria, Denmark, Estonia, France, Germany, Greece, Hungary, Luxembourg, Poland, Romania

Figure 8: Treatment of losses of a foreign permanent establishment

As a general rule, relief for losses from foreign permanent establishments is granted automatically in the home country of the investor, if double taxation on income from foreign permanent establishments is avoided by the credit method. In this event, since the worldwide income is subject to tax, a foreign loss immediately reduces the taxable income in the home country. Double dips of foreign losses are prevented since a future loss relief in the source country results in additional taxes in the home country of the head office. The credit method currently is prevailing in 12 member states. If the exemption method applies (15 member states), as a general rule, losses from foreign permanent establishments have no impact on the domestic tax base. Austria, Belgium, Cyprus, the Netherlands and Spain, however, grant relief of foreign losses even if foreign profits are exempt from tax. Recapture provisions prevent deducting the same loss in both countries. The remaining ten member states do not grant any immediate loss relief. Foreign losses, therefore, do only result in a tax saving if they can be deducted from future foreign profits. Therefore, in loss situations, foreign direct investments through permanent establishments are discriminated against domestic investments. Among these ten member states, only Denmark and France apply a strict territorial principle excluding all profits and losses from foreign permanent establishments from domestic taxation. This might justify the denial of the relief for losses from foreign permanent establishments. 2.1.2.8 Group Taxation Accounting for losses of subsidiaries is an even more complex issue. Since a corporation is an own legal entity, as a consequence of the deferral principle corporate losses are not transferred to the shareholder. Therefore, a transfer of corporate losses to the shareholder can only be based on special regimes for group taxation.

A. Concept and Necessity of a Common Tax Base – an academic introduction

13

Group Taxation

No special regime

Special regime

Domestic Loss + Deferred compensation taxation of intra-group transfers

Belgium, Bulgaria, Czech Republic, Estonia, Greece, Hungary, Lithuania, Poland, Portugal, Romania, Slovakia

Cyprus, Finland, Germany, Latvia, Luxembourg, Slovenia

Ireland, Malta, Netherlands, Spain, Sweden, United Kingdom

Cross-border Loss + Deferred compensation taxation of intra-group transfers

Austria, Denmark, Italy

France

Figure 9: Group taxation

The survey of the member states practices reveals that 11 member states do not provide any special provision in this field (Figure 9). 12 member states apply special group taxation regimes. In all of these countries, qualifying group members are required to be aligned via ownership, which is a legal criteria. Parent companies must hold at least more than one half of the shares in the subsidiaries; in certain cases, direct and indirect participation might be added. Most of these member states require holding quotas of more than 90 per cent. Among the group taxation regimes, there are many differences between systems and consequences in detail. With respect to the technique of combining the individual results of group members one can distinguish between a pure loss compensation which can be based on a pooling mechanism, a loss transfer or on subvention payments, and a deferred taxation of intra-group transfers based on consolidation or simpler techniques. As the results in Figure 9 display, pure loss compensation is the prevailing instrument. A deferred taxation of intra-group transfers is only applied by seven member states. International group taxation within the EU is still in its infancy. Figure 9 shows that in most countries, the regime stops at the border. Non-resident subsidiaries can be part of a tax group only in Austria, Denmark, France and Italy. Among these four member states, only France provides a deferred taxation of intra-group transfers in addition to loss compensation. Therefore, international groups are discriminated against domestic groups in the vast majority of member states applying special group taxation regimes. 2.1.3 Corporation Tax Systems There are various systems governing the taxation of corporate profits distributed to personal income tax payers. Regarding the extent of integration of the corporation

14

Christoph Spengel

tax into the personal income tax of the shareholder, the classic system, double taxation reducing systems and double taxation avoiding systems can be distinguished (Figure 10).14

Corporate Level

Dividend Deduction < 100 %

Ireland

Double Taxation Avoiding Systems

Double Taxation Reducing Systems

Classical System

Split Rate System

Shareholder Level

Shareholder Relief

Corporate Level

Partial Dividend Imputation Deduction 100 %

Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Finland, France, Germany, Hungary, Italy, Lithuania, Luxembourg, Netherlands, Poland, Portugal, Spain, Slovenia, Sweden, United Kingdom

Shareholder Level

Split Full Tax Dividend Corporate Imputation Exemption Tax Rate System

(Estonia)

Malta, Romania

Estonia, Greece, Latvia, Slovakia

Figure 10: Corporation tax systems

The classic system is currently only applied in Ireland. This corporation tax system results in an unrelieved double taxation of dividends. Profits are first taxed by the corporation and then, on distribution, by the shareholder. In contrast, double taxation avoiding systems ensure that distributed profits are only taxed once. In Malta and Romania, double taxation is avoided by applying a full imputation system. Corporation tax paid is credited against the personal income tax of the shareholder as an “imputation credit”. The ultimate tax burden thus depends on the personal income tax of the shareholder only. Greece, Latvia and Slovakia eliminate double taxation through a system of dividend exemption for shareholders. Thus, the tax burden of distributed profits is only determined by the corporation tax. Estonia has her own, unique system of taxing profits only once – on the corporation at the time of distribution. This clearly places a burden on the distribution of the profits as opposed to profit retention. Shareholders enjoy exemption of dividends personal income tax, as is the case in Greece, Latvia and Slovakia. The vast majority of member states, however, grant only partial relief from double taxation of distributed profits. All of these countries provide a preferential treatment for dividend income at the shareholder level, either by a reduced tax base or a reduced rate on dividends (shareholder relief).

14

See Endres/Oestreicher/Scheffler/Spengel, The Determination of Corporate Taxable Income in the EU Member States, Aalphen 2007, pp. 20-21.

A. Concept and Necessity of a Common Tax Base – an academic introduction

15

In summary, the survey of the member states’ corporation tax systems reveals, that except for Ireland, the corporation tax is integrated into the personal income tax to different extents in all member states. Therefore, a harmonisation of the corporation tax system and the tax rate would have an immediate impact on the personal income tax. In the vast majority of the member states, this integration, however, does not cover the total amount of corporation tax underlying distributed profits. Therefore, part of the corporation tax always becomes a definite burden which causes distortions with respect to domestic investments. In particular, equity financing is discriminated against debt financing and corporate businesses are in the event of profit distributions discriminated against incorporated businesses (sole proprietors, partnerships).15 2.1.4 Local Taxes Corporations may be subject to additional taxes on business profits or on business assets other than corporation income tax. The most important additional taxes are real estate tax, property tax and local business taxes (Figure 11).16 AT BE BG CY CZ DE DK EE ES FI FR GR HU IE IT LT LU LV MT NL PL PT RO SE SK SL UK Real property tax



 

 





 



     

Net wealth tax Payroll tax Franchise tax on income Franchise tax on capital

     





















Figure 11: Local taxes in EU Member States (2007)

Real estate tax on industrial buildings is levied in all member states except Estonia, Greece, Malta and Slovenia. The tax base covers land and buildings. The taxable value is derived from either market prices, lower standard tax values or the area of land (square meters). Hence, the taxable value may be completely different in different countries even though the tax base covers the same items. Although the amount of real estate tax varies from country to country, real estate tax has no significant impact on the effective tax burden of companies, since the tax base is limited to only a few numbers of assets and tax rates are rather low. 15 16

See Jacobs, Intertax 1999, pp. 264-278, for a detailed analysis. See European Tax Handbook, Amsterdam 2007, and own investigations.

16

Christoph Spengel

A property or net wealth tax on business assets of corporations is only imposed by Luxembourg. Property tax, however, can be credited against corporation tax. As a result, profitable corporations in Luxembourg do not suffer from property tax. Additional local business taxes are only levied in a few member states. They can be levied on a profit basis (Hungary, Germany, Italy, Luxembourg); on a capital basis independent of the profits (France, Spain); or on a wage-oriented basis (Austria, France). The bases for these local business taxes vary from country to country but are defined uniformly within one country. As far as non-profit taxes are concerned, they do usually not include intangibles, financial assets or stocks. Moreover, in most cases it is obvious that liabilities connected with the financing of the assets are not deductible from the tax base. The tax rates are different and vary from country to country as well. Moreover, the tax rates differ within a country, as local authorities or municipalities are entitled to set the tax rates. In summary, the survey of the local taxes levied by the member states reveals that local taxes in general are of minor importance compared to the corporation tax. Above all, therefore, local taxes are not an issue when harmonising business taxation in the EU. In contrast, past experience proves that member states were willing to reform their local taxes on an individual basis (e.g. Austria abolished the trade tax on capital and on income; Germany abolished the trade tax on capital and the property tax; Luxembourg abolished the trade tax on capital; the Netherlands abolished the property tax). However, a harmonisation of the corporation tax as proposed by the European Commission would be linked indirectly with local taxes. First, most of the local taxes are deductible from the base of the corporation tax as business expenses. Second, local business taxes on profits in some member states take as a starting point for the determination of the tax base the base of the corporation tax which is modified afterwards to certain degrees (e.g. trade tax on income in Germany).

2.2 Effective Tax Burdens 2.2.1 Purpose of the Survey and Outline of the Model The main purpose of this chapter is to provide information about the effective average tax rates (EATR) for EU-based corporations enabling to work out the impact of the tax systems on the decisions multinational investors make about the location, investment strategies and financing options. Moreover, the most important elements of the tax systems that determine effective company tax burdens (i.e. the tax drivers) should be identified. The calculation of effective tax burdens is based on the commonly accepted approach of King and Fullerton,17 which has been extended by Devereux and Griffith.18 The most important studies by the OECD19 and the European Commission20 17 18

19 20

See King/Fullerton, The Taxation of Income from Capital, Chicago 1984. See Devereux/Griffith, The Taxation of Discrete Investment Choices – Revision 2, IFS Working Paper Series No, W98/16, London 1999. See OECD, Taxing Profits on a Global Economy, Paris 1991. See European Commission, Company Taxation in the Internal Market, 2001.

17

A. Concept and Necessity of a Common Tax Base – an academic introduction

apply this methodology. The model firm considered here (Figure 12) takes into account five different types of investment (intangibles, buildings, machinery, financial assets and inventories) and three different sources of finance (retained earnings, new equity capital and debt). Investments are carried out in a subsidiary that is located in one of the member states. The sources of finance of the subsidiary are provided by a parent company that is also located in one of the member states. There is a complete repatriation of the profits of the subsidiary to the parent. In the case of new equity financing, we assume a full distribution of profits as dividends in the same period. In the case of financing with retained earnings, we assume that profits will be distributed in subsequent periods. In the case of debt financing, we assume that the subsidiary pays interest to the parent at a fixed rate and distributes the remaining profits (i.e. real profits exceeding the assumed market interest rate) as a dividend. In each case, the parent needs to raise funds itself, and this is done by its own shareholders which are supposed to be natural persons. Shareholder taxation, however, is not taken into account here because personal taxation is in most cases not relevant for investment, financing and location decisions of multinationals.21 Intangibles

Buildings

Machinery

Financial Assets

Inventories

45 Investments

5 Assets

Subsidiary 3 Sources of Finance

Debt

Retained Earnings

Dividends

Retained Dividends Earnings

Dividends

Interest Dividends Equity

Dividends

Parent Company

3 Sources of Finance

Debt

Interest Dividends

External Funds

Equity

Shareholders (Not considered here)

Capital Market

Figure 12: Outline of the model

The consideration of five types of assets at the level of the subsidiary and three sources of finance at both the level of the subsidiary and the parent company results in 45 possible combinations of assets and financing for both the level of the subsidiary and the level of the parent. In order to keep the comparison of EATR for all 21

For technical details of the model, reference is made to Annex A of the European Commission’s report from 2001. See European Commission, Company Taxation in the Internal Market, 2001, pp. 519-533.

18

Christoph Spengel

countries manageable, mean (weighted average) EATR for each type of asset, mean (weighted average) EATR for each source of finance and an overall mean EATR for all combinations of assets and financing are calculated.22 The results presented below provide estimates of EATR for corporations taking into account the tax regimes as of 1 January, 2006.23 2.2.2 Domestic Investments The following presents effective average tax rates (EATR) at the level of subsidiaries resident in the 27 member states of the EU. Figure 13 illustrates overall average EATR for the five types of investment and the three sources of finance. The results reveal that, on average, company taxation in the member states is not neutral towards investment and financing decisions. (

)

30 25

EATR in %

20 15 10 5 0 Buildings Intangibles Machinery Financial Inventories Assets

Retained Earnings

New Equity

Debt

Figure 13: Effective average tax rates – investment and finance (EU27 overall averages, level of subsidiary)

With respect to the types of investment, capital allowances (i.e. depreciation) for buildings, intangibles and machinery exceed true economic depreciation on these assets. By contrast, non-depreciable assets (i.e. financial assets and inventories) receive no similar relief. The preferential depreciation for tax purposes, therefore, leads to modest EATR for depreciable assets. Since real estate tax imposes an extra tax load on industrial buildings, the EATR for intangibles and machinery is (on average) approximately the same and the lowest among the five assets considered here. 22

23

For simplification purposes, the sources of finance and the types of assets are weighted equally, i.e. 33.33 per cent for each source of finance and 20 per cent for each type of asset. Additional economic data incorporated into the model are an inflation rate set at 2 per cent, a pre tax interest rate of 7.1 per cent and economic depreciation of fixed assets on a declining basis at the rate of 3.1 per cent for buildings, 15.35 per cent for intangibles and 17.5 per cent for machinery. The pre-tax financial return of the investment is set at 20 per cent. The results were taken from an ongoing survey of the Centre for European Economic Research (ZEW) on behalf of the European Commission.

A. Concept and Necessity of a Common Tax Base – an academic introduction

19

From the perspective of a corporation, debt financing is always more tax efficient compared to equity-financed investments since nominal interest payments are deductible from taxable income. In contrast to debt financing, financing through retained earnings and new equity is disadvantageous because no deduction from the taxable income is allowed for the corresponding dividend payments. The EATR on retained earnings and new equity is the same for all member states except for Estonia. Here, taxation can be deferred by profit retention at the corporate level since retained earnings are exempt from taxation (see above, 2.1.3). Therefore, on average, retained earnings are bearing a lower EATR compared to new equity. The preferential treatment of debt financing at the level of the subsidiary does not mean, however, that debt financing is more tax efficient than equity financing. Since interest received by a subsidiary is always taxable at the level of the parent company and double taxation of dividends in the case of equity financing is avoided as a rule, a multinational investor always has to compare the aggregate tax burden on interest payments and repatriated profits. This is examined below (see 2.2.3). 45 EATR

nom. tax rate

40

EATR/stat. tax rate in %

35 30 25 20 15 10 5 0 CY BG LV IR RO LT HU SK PL EE CZ SI AU FI PT SE DK GR LU NL UK BE IT MA FR DE ES

Figure 14: Nominal tax rates and effective average tax rates (EU27 overall averages of investment and finance, level of subsidiary)

The analysis of average EATR for company investment and finance hides considerable dispersions among the EATR in each member state. There is a wide range of EATR within the member states (Figure 14). The overall spread is 25.9 percentage points; EATR is lowest in Cyprus (10.6 per cent) and highest in Spain (36.50 per cent). The average EATR in the member states is 23.13 per cent. The elements that determine the effective tax burdens cannot be explained by one single feature of the tax regime (i.e. the number of taxes, the tax bases and the tax rates). However, the statutory (nominal) tax rate on corporate profits plays an important role in determining the EATR. The ranking of the countries based on the nominal tax rates on corporate profits is a good indicator of the ranking with respect to the EATR. Here, there would be few changes in the country ranking (Figure 14). This is because the

20

Christoph Spengel

calculations of effective company tax burdens concentrate on profitable investments and only few member states levy substantial non-profit taxes. The spread between the EATR and the nominal tax rate on corporate profits is not more than 4 percentage points except in Estonia and Italy.24 The lower EATR compared to the nominal tax rate on corporate profits in all countries except Cyprus and Ireland is explained by the tax-reducing impact of the tax base (e.g. depreciation and stock valuation) and the deduction of interest payments in case of debt financing.25 2.2.3 Cross-border Investments This final chapter presents EATR on investment undertaken in one member state (the source country) by a subsidiary of a parent company resident in a different member state (the home country). In order to assess the impact of taxation on location and cross-border investment and financing decisions, the EATR on cross-border investment are compared with the EATR on domestic investment presented in the chapter above (see 2.2.2). As well as taking account of the domestic tax system, it therefore also incorporates taxes charged on the payment of dividends and interest between the subsidiary and the parent, and any further taxes levied by the country of the parent company. mean EU27 30

EATR in %

25 20 15 10 5 0 Domestic

Outbound

Inbound

Average Finance

Outbound

Inbound

Optimal Finance

Figure 15: Cross-border investment – mean of EU27

In Figure 15, the first bar presents the average EATR for domestic investment in the 27 member states, averaged over five types of investment and three sources of finance. The average EATR amounts to 23.13 per cent as analysed above (see 2.2.2). 24

25

In Estonia and Italy, the high spread between the two measures is caused by special features of the national tax systems, the taxation of distributed profits only in Estonia and the levy of a special local business tax (IRAP) in Italy. In both countries, the interaction of low nominal tax rates on corporate profits and a considerable real estate tax results in higher EATR compared to the nominal tax rate.

A. Concept and Necessity of a Common Tax Base – an academic introduction

21

The second bar presents the average EATR for outbound investment. This is an unweighted average taken across the 26 potential source countries for each home country. The third bar is equivalent, but presents the average EATR for inbound investment (i.e. treating each member state as a home country, rather than a source country). For both outbound and inbound investment presented in the second and the third bar, each source of finance of the subsidiary provided by the parent company is given an equal weight. The comparison between EATR for domestic investment and for average cross-border outbound and inbound investment indicates that, on average, cross-border investment are more heavily taxed than domestic investment and, thus, discriminated against domestic investment. Moreover, the results indicate that the EU tax regimes neither support the residence nor the source principle. If this were true, EATR on domestic investment should be identical either to the EATR on outbound investment (residence principle) or the EATR on inbound investment (source principle). The perspective so far is based on EATR averaged over the three possible sources of finance of the subsidiary (i.e. debt financing, retained earnings and new equity). Multinational investors, however, can choose the source of finance of the subsidiary; if one particular source of finance is tax disadvantaged, then it would not be used. Therefore, rather than take an average of the sources of finance, the results presented in the fourth and the fifth bar simply take the most tax efficient means of financing the subsidiary. That is, the form of finance which generates the lowest EATR was chosen.26 This so-called tax minimisation approach generates a rather different picture. On average, member states now have lower EATR on both outbound and inbound investment than for domestic investment. This suggests that if companies do indeed choose the most tax-favoured source of finance and thus taking advantage of the considerable variation of effective company tax burdens among member states, then – other thing being equal – the international tax system works such that foreign multinationals operating in a source country are likely to face a lower EATR than domestic companies. Therefore, one cannot conclude that cross-border investment is discriminated against domestic investment. Again, however, this is also inconsistent with both the residence and the source principle. The average results presented in Figure 15 hide considerable variations between different home and source countries with respect to advantages or discriminations of cross-border investment compared to domestic investment. Therefore, Figure 16 presents results for outbound and inbound investment from the perspective of a low tax country and a high tax country respectively. Among member states, Cyprus is considered as a low tax country and Spain is considered as a high tax country.27 For both outbound and inbound investment the EATR is illustrated for each source of 26

27

Recall (see above, 2.1.2.7), for example, that from the perspective of an investor located in a high tax country equity financing of a subsidiary located in a low tax country is more tax efficient than debt financing. The opposite (i.e. debt financing is more tax efficient than equity financing) turns out to be true, if the home country imposes a higher tax burden than the source country. See above Figure 14, 2.2.2.

22

Christoph Spengel

finance of the subsidiary provided by the parent company. Subsidiaries (outbound investment) and parent companies (inbound investment) are located in the remaining 26 member states. 40 35

EATR in %

30 25 20 15 10 5

Cyprus

t

ty

eb D

gs

ui

in rn Ea

ew N

d ne ai et R

et R

Inbound

Eq

t

ty

eb D

gs

ui

in Ea d ne ai

N

D

ew

rn

es om

D

ai et R

Outbound

Eq

tic

t eb

ty

gs in rn

Eq

Ea

ew N

d

ui

t

ty

eb D ne

Ea

ui

in N

d R

et

ai

ne

ew

rn

es om D

Eq

tic

gs

0

Outbound

Inbound

Spain

Figure 16: Cross-border investment – highest and lowest level

From the perspective of low tax countries (here: Cyprus), parent companies located therein have no incentives to invest abroad from a tax point of view. Irrespective of the source of finance of a foreign subsidiary, outbound investments are bearing a higher EATR compared with domestic investment. However, outbound investment should be financed preferentially with debt capital because then foreign profits shifted via interest payment are subject to the lower domestic tax burden. Equivalently, inbound investments to low tax countries are suffering a low EATR. In order to benefit from the lower local taxes, the parent company should finance subsidiaries rather with equity than with debt capital. The perspective of high tax countries (here: Spain) is the opposite. Parent companies located therein have incentives to invest abroad from a tax point of view. Irrespective of the source of finance of a foreign subsidiary, outbound investments are bearing a lower EATR compared with domestic investments. However, outbound investment should be financed preferentially with equity capital and profits should be retained at the level of the foreign subsidiary because then foreign profits are subject to the lower foreign tax burden only. Equivalently, inbound investments to high tax countries are suffering a lower EATR than local investments. In order to benefit from the lower taxes in the home country, the parent company should finance subsidiaries rather with debt than with equity capital. As a result, multinational investors have an incentive to shift their mobile tax bases from international investments to low tax countries increasing the tax revenues of these countries in favour of the tax revenues of high tax countries. High tax countries, therefore, are seeking for measures restricting international profit shifting to protect their tax revenues. One of the major reasons for this unbalanced distribution

A. Concept and Necessity of a Common Tax Base – an academic introduction

23

of taxable profits of multinational investors to different locations for investment can be seen in the coexistence of the residence and the source principle when taxing foreign income.

2.3 Evaluation 2.3.1 Criteria The criteria to evaluate the prevailing situation of company taxation in the EU member states cover (1) efficiency and neutrality as a major economic goal of the EC Treaty, (2) compliance costs and risks for double taxation of income from the perspective of businesses and (3) enforcement of taxes and protection of tax bases from the perspective of member states’ tax administrations. 2.3.2 Economic Perspective of the Common Market: Efficiency and Neutrality Efficiency is widely recognized as a fundamental economic principle of optimal taxation. This principle is also inherent to the EC Treaty. According to Articles 2 and 98 EC Treaty, the promotion of an efficient allocation of resources is a fundamental economic goal of the Common Market. In order to reach efficiency, taxes should be neutral to economic agents’ behaviour and not distort their decisions.28 There are different possible definitions of neutral and efficient taxation. Efficiency takes a macro-economic perspective and demands for a pareto-optimal allocation of resources which should not be changed by the presence of taxation.29 Neutrality focuses on the tax effects on economic decisions from a firm’s perspective. In this sense neutral taxation does not interfere with the decision-making of businesses. It has been shown elsewhere that a neutral system of company taxation (i.e. without taking personal taxes into account) promotes an efficient taxation at the macro level.30 The taxation of EU-wide businesses is not in line with the idea of a Single Market. With regard to the wide range of effective tax burdens, company taxation affects decisions of investors with respect to the location of an investment, the type of investment and its source of finance (see above, 2.2). Thus, an efficient allocation of resources cannot be achieved; the competitiveness of EU-based companies is restricted, and the Community as a whole suffers welfare losses.

28

29

30

See European Commission, Company Taxation in the Internal Market, Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee, COM(2001) 582 final, Brussels, 23.10.2001, p. 26. See Musgrave/Musgrave, Public Finance in Theory and Practice, 5th Ed., New York 1989, pp. 279-294. See Homburg, Allgemeine Steuerlehre, 5th Ed., München 2007, pp. 238-256.

24

Christoph Spengel

2.3.3 Perspective of Businesses: Obstacles to Cross-border Activities within the EU From the perspective of businesses, the coexistence of 27 separate tax systems causes several tax obstacles to cross-border activities within the EU. First, the need to comply with different rules concerning the tax rates, the tax bases, the corporation tax systems, the local taxes and the taxation of foreign income (see above, 2.1) entails a considerable compliance cost and represents itself a significant tax obstacle. According to a recent survey of the Commission, compliance costs amount to 1.9 per cent of the tax payments in the case of multinational enterprises while they amount to 30.9 per cent in the case of medium-sized enterprises.31 Moreover, since no single taxation of multinationals exists and each member state is a separate tax jurisdiction, this entails a number of further consequences. Since separate taxation in each member state and the coexistence of the residence and the source principle for taxing foreign income prevail, double taxation may occur as a result of conflicting taxing rights. In particular, (1) relief for losses incurred by associated companies located in other member states is not allowed in many cases, (2) the allocation of profits of multinationals to different jurisdictions on an arm’s length basis by transfer prices causes methodological problems and results in double taxation, and (3) cross-border reorganisations give rise to capital gains taxation and bear the risks of double taxation in many situations. 2.3.4 Perspective of Member States Tax Administrations: Enforcement of Taxes and Protection of Tax Bases Tax administrations of high tax member states in particular have difficulties to enforce taxes and to protect their tax bases against exhaustive profit shifting of multinationals. Many member states, therefore, have introduced measures to prevent the transfer of profits abroad and to protect national tax revenue from burgeoning tax planning. These measures include, for instance, restrictions on cross-border loss compensation. In addition, in the area of company finance, measures comprise thin capitalisation rules, CFC-legislation, and limitations for the allocation of expenses and the deductibility of headquarter costs. Moreover, reorganisations cause considerable exit taxes preventing companies from the relocation of business units and their head offices to other member states. Therefore, European businesses are not at all in a position to take full advantage of the economic benefits of the Single Market. Finally, there are intensified controls, corrections and documentation requirements relating to intragroup transfer prices, but only to the extent that the contracting unit is a foreign subsidiary or a permanent establishment. Substantial elements of these tax provisions contravene the ban on discrimination and may violate the fundamental freedoms laid down in the EC Treaty, in particular the freedom of establishment (Article 43 EC Treaty) and the free movement of capital (Article 56 EC Treaty). For instance, the European Court of Justice (ECJ) 31

See European Commission, European Tax Survey, Commission Staff Working Paper, SEC(2004) 1128/2, Brussels 2004.

A. Concept and Necessity of a Common Tax Base – an academic introduction

25

has overturned defensive tax measures aimed at protecting high-tax member states from profits being transferred to low-tax countries by means of external financing. The decisions of the ECJ cover thin-capitalisation rules (Lankhorst-Hohorst32), deduction of costs for refinancing (Bosal33 and Keller Holding34) and CFC-legislation (Cadbury Schweppes35). In two decisions, the ECJ ruled that a denial of crossborder loss relief is against EU law, if a member state refuses to allow deducting losses of a subsidiary only because it is resident in another member state (Marks & Spencer36) or if the deduction of individual losses is refused simply because the source of income is abroad (Rewe Zentralfinanz37). Also, the change in the ownership of a corporation may not give rise to an immediate taxation of capital gains, even if the member state where the corporation is resident loses its right to tax such capital gains (X and Y 38). ECJ case law is making it increasingly difficult for member states to place restrictions on international tax planning. The only choice left to member states is either to extend to domestic cases a restrictive regulation that is actually aimed at international cases, or to abolish restrictions for international cases if they do not also apply to domestic cases. Both options are bearing a conflict of objectives. In the first case, additional burdens are placed on domestic companies for which there is otherwise no justification. Moreover, this destroys the coherence of domestic company taxation at the same time. In the second case, more scope is given to international tax planning than is justified from a national perspective. This includes a considerable fiscal risk. Moreover, a continued existence of potential conflicts against the fundamental freedoms remains. One can therefore conclude that without further tax coordination member states presumably are not able to reform their tax systems so that they respect the fundamental freedoms of the EC Treaty for cross-border activities and do not destroy the coherence of domestic company taxation at the same time.

2.4 Is Harmonisation of Nominal Tax Rates a Way Out? The causes of the variations between the effective average company tax burdens (EATR) among the EU member states are complex. They lie in the many differences between the different elements of the national tax systems. It has, however, been shown that the statutory (nominal) tax rate on corporate profits plays a significant role in determining the EATR (see above, 2.2.2). A harmonisation of the corporation tax rate in itself would therefore go a long way towards reducing distortions in the decisions made by companies with respect to location, investments and financing. However, this would mean giving up a key element of national tax sovereignty. 32 33 34 35 36 37 38

See ECJ of 12.12.2002, C-324/00 (Lankhorst-Hohorst), ECR 2002, p. I-11779. See ECJ of 18.9.2003, C-168/01 (Bosal), ECR 2003, p. I-9409. See ECJ of 23.2.2006, C-471/04 (Keller Holding), ECR 2006, p. I-2107. See ECJ of 12.9.2006, C-196/04 (Cadbury Schweppes), ECR 2006, p. I-7995. See ECJ of 13.12.2005, C-446/03 (Marks & Spencer), ECR 2005, p. I-10837. See ECJ of 29.3.2007, C-347/04 (Rewe Zentralfinanz), ECR 2007, p I-2647. See ECJ of 21.11.2002, C-436/00 (X and Y), ECR 2002, p. I-10829.

26

Christoph Spengel

Therefore, and also with regard to the subsidiarity principle (Article 5 EC Treaty), the harmonisation of corporation tax rates is not on the list of objectives – at least not for the present.39 Despite their importance, harmonisation of the tax rates alone would be insufficient in terms of the above-mentioned problems because the different rules on determining taxable profits would continue to exert an influence on the effective tax burdens. Isolating national tax bases also causes major obstacles to cross-border business activities in the Single Market thanks to the coexistence of the residence principle and the source principle. After all, this coexistence is also a decisive factor in the increasing disputes over the distribution of tax revenues between member states and in the related conflicts with EU law caused by the domestic measures to combat profit transfers to low-tax countries.

3. Common Tax Base 3.1 Principles The current initiatives of the European Commission are based on the 2001 report “Company Taxation in the Internal Market”.40 The Commission believes that only a comprehensive solution can help to eliminate tax obstacles systematically. The long-term objective in removing obstacles to cross-border business activities is the creation of a Common Consolidated Corporate Tax Base (CCCTB) for the EU-wide activities of multinationals. A proposal for a directive should be released till the end of 2008.41 According to the concept of the CCCTB three distinct steps are necessary to arrive at the tax base for each jurisdiction: (1) Each group member calculates its taxable profits separately but according to the same set of rules; (2) The individual tax bases are aggregated to the consolidated tax base; (3) The consolidated tax base is allocated to the different member states by applying specific factors (formula apportionment). Finally, each member state preserves its right to tax the allocated portion of the consolidated tax base applying its own tax rate. According to the Commission’s idea, therefore, a group of companies forming an economic entity is not treated as a single taxpayer. The collection of taxes on a group member’s income will continue to be assessed on an individual basis; in particular each group member preserves its own and distinct legal capacity. Since the income attributed to a jurisdiction is ultimately 39

40 41

See European Commission, Communication from the Commission to the Council and the European Parliament: The Contribution of Taxation and Customs Policies to the Lisbon Strategy, COM(2005) 532 final, 25.10.2005, p. 6. See European Commission, Company Taxation in the Internal Market, 2001. See European Commission, Communication from the Commission to the Council, the European Parliament and the European Economic and Social Community, Implementing the Lisbon Programme: Progress to date and next step towards a Common Consolidated Corporate Tax Base (CCCTB), COM(2006) 157 final, Brussels, 5.4.2006.

A. Concept and Necessity of a Common Tax Base – an academic introduction

27

taxed according to this country’s rules, the CCCTB is conceptually aligned to the source principle.42 The allocation of group profits to the individual group companies represents a paradigm shift. Consolidating the separately determined profits of the group members makes it impossible to maintain the prevailing system of direct division of profits using transfer prices based on the arm’s length principle for individual transactions (separate entity accounting). Instead, consolidating the individual results requires an indirect division of the profits of the consolidated overall result using a formula, i.e. by breaking down the group's result among the individual group companies (formula apportionment). Formula apportionment has a long tradition in North America concerning group taxation at the level of the States (USA) and the level of the Provinces (Canada).43 The rationale of formula apportionment begins with the limits of separate entity accounting. Concerning economically integrated groups of companies, the transactional approach seems theoretically questionable. By setting up an integrated group of companies, coordination of transactions via markets is abandoned in favour of coordination using intra-organisational hierarchies. The aim is to generate economies of integration, for example, by means of lower transaction costs, improvement of information flow or managerial efficiency.44 As a result, the profits of an integrated group of companies are greater than the aggregate profits earned by its separate entities. Since the excess profits accrue at group level, it seems difficult to determine the source of these profits as they cannot be attributed to specific transactions either. Therefore, the comparison of controlled transactions with uncontrolled transactions – as the arm’s length principle implies – seems conceptually questionable and systematically inapplicable.45 Another problem arising in the context of the arm’s length principle is that of double taxation. One tax jurisdiction may adjust a given transfer price because it is deemed not to be at arm’s length. If the other jurisdiction does not make a corresponding adjustment, there is a risk of double taxation.46 Against this background, the current international tax system is inadequate with reference to the principles of efficiency and neutrality as well as simplicity and enforceability. The arm’s length principle ignores the differences between controlled 42

43

44 45

46

See Devereux, International Tax and Public Finance 2004, p. 83; Wissenschaftlicher Beirat beim Bundesministerium der Finanzen, Einheitliche Bemessungsgrundlage der Körperschaftsteuer in der Europäischen Union, Gutachten, Berlin, 2007, p. 28. See Martens Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, European Commissions’ Taxation Papers, Working Paper No. 8, Brussels, 2005, pp. 10-15. See Berry/Bradford/Hines, Tax Notes 88, 1992, p. 737. See Mclure, Defining a Unitary Business: An Economist’s View, in: McLure (Ed.), The State Corporation Income Tax: Issues in Worldwide Unitary Combination, Stanford, pp. 94, 105; Oestreicher, Konzern-Gewinnabgrenzung, München 2000, p. 20; Jacobs, Internationale Unternehmensbesteuerung, 6th Ed., München 2007, pp. 588-592. See Newlon, Transfer Pricing and Income Shifting in Integrating Economies, in: Cnossen (Ed.), Taxing Capital Income in the European Union: Issues and Options for Reform, Oxford, 2000, pp. 220-221.

28

Christoph Spengel

and uncontrolled transactions and entails scope for abusive transfer pricing. It is incapable of fairly allocating profits to the countries involved according to their source and thus, is inconsistent with the principle of inter-nation equity.47 At the same time, double taxation, arising if transfer prices are adjusted unilaterally, violates the principle of equity between taxpayers.48 If the CCCTB is introduced, the arm’s length principle as a means for the allocation of taxable income between jurisdictions would be replaced by formula apportionment. Formula apportionment does not seek to allocate income to its source perfectly. The rationale behind formula apportionment rather is to provide a pragmatic solution for profit allocation among jurisdictions in order to better cope with the issues of simplicity and enforceability.49 However, formula apportionment is not arbitrary. Depending on the choice of apportionment factors, this approach intends to allocate the consolidated tax base to the profit generating activities. Factors which are deemed to represent profit generating activities are typically property, payroll and sales. In this context it has to be kept in mind that in an economically integrated entity consisting out of separate legal entities it is not possible to determine the source of profits on an economic rationale.

3.2 Advantages Although there is strong support for the idea of a common tax base, it is, however, not clear whether such a harmonised tax base really corresponds to the original idea of a CCCTB. Three different interpretations of a harmonised tax base exist, each of which involves a different degree of mutual cooperation and each of which eliminates tax obstacles to cross-border EU-wide activities to a different extent (Table 2).50

47 48 49 50

See Jacobs,/Spengel/Schäfer, Intertax 2004, pp. 272-273. See Li, Canadian Tax Journal 2002, p. 840. See contribution of Walter Hellerstein in this volume. In the meantime, the Commission also identifies a number of alternative options to achieve a CCCTB. The Commission distinguishes between (1) no-change, (2) common base without consolidation and (3) common consolidated base. See European Commission, Communication from the Commission to the European Parliament, the Council and the European Economic and Social Committee, COM(2007) 223 final, 2.5.2007, p. 5. This is similar to the scenarios presented here.

A. Concept and Necessity of a Common Tax Base – an academic introduction

29

Table 2: Approaches to an EU-wide harmonised tax base Approaches to EU company taxation with different degree of international cooperation Reduction / elimination of tax obstacles to crossborder EU-wide activities

– Harmonised tax base throughout the EU with one set of tax accounting principles

– Harmonised tax base throughout the EU with one set of tax accounting principles – Cross-border loss relief

– Harmonised tax base throughout the EU with one set of tax accounting principles – Consolidation (including elimination of inter-company profits) – Allocation of the consolidated tax base

Compliance costs

Achieved

Achieved

Achieved

Cross-border loss relief

Not achieved Except to the extent that member states already provide cross-border loss relief

Achieved

Achieved

Transfer prices

Not achieved Transfer prices are still required for the division of the tax base

Not achieved Transfer prices are still required for the division of the tax base

Achieved Transfer prices are substituted by formula apportionment

Reorganisations

Achieved But only if the tax treatment of reorganisations is harmonised

Achieved But only if the tax treatment of reorganisations is harmonised

Achieved But only if the tax treatment of reorganisations is harmonised

Double taxation as a result of conflicting taxing rights

Not achieved

Not achieved

Achieved

The minimum degree is a harmonised corporate tax base (CTB without consolidation) which should be based on a single set of tax accounting principles. Clearly, such a model would reduce compliance costs. However, all other tax obstacles on cross-border activities identified above would remain (see above, 2.3). A harmonised corporate tax base is a prerequisite for cross-border loss relief. Otherwise, there would have to be separate accounting rules for the determination of foreign losses with all the attendant difficulties associated with the recapture of loss relief if the foreign subsidiary claims its own loss relief later. Some mechanism is necessary to prevent that loss relief is claimed more than once, as was emphasised in an earlier proposal for foreign loss relief by the European Commission.

30

Christoph Spengel

In order to fully eliminate tax obstacles to cross-border EU-wide activities a common consolidated corporate tax base has to be established. The main advantages of the CCCTB are, first, the lower tax-compliance costs for investors. Further advantages lie in the consolidated determination of taxable profits: consolidating the individual results of the group members leads to cross-border loss compensation at the level of the taxable entity. Furthermore, supplies and services can generally be invoiced at the respective tax book value because profits realized from transactions with third parties are distributed among the group companies by the division of profits. This leads to an elimination of inter-company profits, which takes the single economic unit aspect of the group into account. Moreover, the group’s tax burden cannot be influenced by changing intra-group financing, so that measures to limit shareholder debt financing (thin capitalisation rules) and the CFC rules would become obsolete. Also, costs for refinancing are deducted from the consolidated group income and they are therefore apportioned to all member states in which group members are resident. Finally, hidden reserves upon cross-border reorganisations or the transfer of assets do not have to be taxed immediately; they are divided among the group members according to the allocation formula upon realisation, irrespective of where the profits are generated. The indirect division of profits (Formula Apportionment) thus abolishes the incentives to transfer book profits from one group member to another. To this extent, the CCCTB takes away most of the companies’ tax-planning leeway but, on the other hand, abolishes obstacles to cross-border business activities, reduces current member states’ conflicts with EU-law and increases the member states’ tax autonomy.

4. Implementation issues A CCCTB raises several implementation issues. They cover the determination of taxable income at the level of each group member, the determination of group income (consolidation), the apportionment of group income, and some related issues.51

4.1 Determination of Taxable Income With regard to the determination of taxable income at the level of each group member different solutions seem to be possible. First, based on their factual importance, there could be a strict dependency of International Financial Reporting Standards (IFRS) for the determination of taxable income. However, there is no support to accept full dependency of IFRS. The evolution of IFRS is not yet clear, they include some accounting options which are not in line with fundamental and common principles of tax accounting (in particular fair-value accounting) and IFRS – like domes-

51

The implementation issues are addressed in more detail by Spengel/Wendt, A Common Consolidated Tax Base for Multinational Companies in the European Union. Some Issues and Options, Oxford University Center for Business Taxation, WP 07/17.

A. Concept and Necessity of a Common Tax Base – an academic introduction

31

tic GAAP – do not provide solutions for a considerable number of provisions for determining the tax base (see above, 2.1.2). Second, tax accounting therefore could be based on a separate and autonomous set of rules distinct from financial accounting. This second solution, however, does not receive strong support either, since it ignores the existing linkage of tax accounting and financial accounting in all member states and, therefore, would rather introduce a 28th system for tax accounting than reaching a mutual recognition of common standards for a harmonised tax base. Based on the existing linkage of financial and tax accounting there is a strong support to accept IFRS as a starting point for the common tax base. The analysis above shows that IFRS could provide elements of a common tax base in certain areas (see 2.1.2). Following more comprehensive studies, these areas could cover the recognition of assets and liabilities, the determination of cost values, amortisation, impairment and treatment of onerous contracts.52 Although based on IFRS, a common tax base first would require common standards of loss relief, the taxation of dividend income, and the taxation of capital gains since these issues are not addressed in financial accounting. Second, since member states address these issues in different manners, the introduction of common and uniform standards in the field of accounting for reserves and in the field of depreciation are necessary. Since the latter are likely to eliminate tax incentives from the tax base such incentives could be substituted by tax credits or grants.

4.2 Determination of Group Income 4.2.1 Group Definition Concerning the determination of group income, it is first necessary to define the taxable group. This includes criteria for the identification of the taxable unit, the nexus of group members to different countries, and the personal scope of a CCCTB. 4.2.1.1 Taxable Unit For the definition of the taxable unit it is necessary to find criteria according to which affiliated companies qualify for consolidation and formula apportionment. The fundamental question is whether to delineate the group in terms of legal criteria, economic criteria or a combination of both. In defining the group, the underlying rationale for the CCCTB has to be kept in mind. Accordingly, economic interrelationships among affiliated companies that make it impossible to determine the source of profits on the basis of separate accounting using arm’s length pricing should lay the foundation for the definition of the consolidated group.53 Legal ownership is widely recognised as a key element in defining the taxable unit both in 52

53

See Oestreicher/Spengel, European Taxation 2007, pp. 437-451; Gammie/Giannini/Klemm/ Oestreicher/Parascandolo/Spengel, Achieving a Common Consolidated Corporate Tax Base in the EU, Centre for European Policy Studies (CEPS), Brussels 2005; see also the contribution of Wolfgang Schön in this volume. See Hellerstein/McLure, International Tax and Public Finance 2004, p. 204.

32

Christoph Spengel

theory 54 and in member states’ practice (see above, 2.1.2.8). The underlying theoretic rationale is that legal ownership entails the ability to govern and control an affiliate’s business activities in order to obtain economic benefits. Legal ownership can be defined either with reference to voting stock or equity. As control requires voting power, ownership of voting rights seems more appropriate than ownership of equity.55 The ownership threshold should not be too restrictive, on the one hand, it should, on the other hand, however, reduce the scope for tax planning. A reasonable compromise could be to set the ownership threshold sufficiently low as to ensure that a further reduction would be associated with a substantial loss of control. Accordingly, an ownership threshold of 50% or 75% seems appropriate for defining the taxable unit subject to taxation under the CCCTB.56 A definition of the taxable unit based solely on legal ownership has the advantage of administrative simplicity, both for taxpayers and tax administrations. It provides a clear cut solution and certainty regarding the entities to be included.57 By contrast, this definition does not necessarily represent the economic substance of the relationship among group entities.58 A subsidiary might be under control but economically independent. If this is the case and the CCCTB is applied, the consolidated tax base of these entities would be sourced using apportionment factors that would not have contributed to the production of the income in question.59 Thus, although legal ownership is a prerequisite for economic integration, an extension of the definition of the taxable unit based on economic criteria seems conceptual more consistent with the concept of CCCTB.60 However, economic criteria are less feasible, as they are rather subjective and do not provide legal certainty. Experience from the US state corporate income taxation has shown that problems exist in attempting to define a unitary business.61 There is no common definition of a uni-

54

55

56

57

58

59 60

61

See Agúndez-García, The Delination and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, Taxation Papers, Brussels 2006, pp. 11-15. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on group taxation (SG – 05 & 06 February 2007), Working Document, Brussels 2007b, p. 4. See contribution of Wolfgang Schön in this volume; Schreiber, Steuer und Wirtschaft 2004, p. 232. See European Commission, Summary Report of the Workshop on the Allocation Mechanism, Brussels 2004b, p. 5. See Martens Weiner, European Taxation 2001, p. 383; Schreiber, Steuer und Wirtschaft 2004, p. 223. See Hellerstein/McLure, International Tax and Public Finance 2004, p. 205. See Agúndez-García, The Delination and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, Taxation Papers, Brussels 2006, p. 13; Jacobs/Spengel/Schäfer, Intertax 2004, pp. 277-278; Martens Weiner, Company Tax Reform in the European Union. Guidance from the United States and Canada on Implementing Formulary Apportionment in the EU, New York 2006, p. 71. See Hellerstein/McLure, International Tax and Public Finance 2004, pp. 203-206.

A. Concept and Necessity of a Common Tax Base – an academic introduction

33

tary business, but tests are applied which are mainly based on economic criteria.62 Against this background, reaching an agreement on economic criteria and applying them uniformly in the EU may become difficult and burdensome for taxpayers as well as tax authorities.63 Therefore, the definition of the taxable unit should rely on legal criteria or bright-line tests indicating whether there are economic interdependencies between affiliated entities rather than on economic criteria. Reference can be made to those bright-line tests developed by McLure and Hellerstein.64 Accordingly, the definition of a consolidated group could be based on a control test and tests determining whether there are economic interdependencies and a flow of value. Whether economic interdependencies or a flow of values between affiliated companies exist should be determined based on objective and easily administrable criteria. For example, the percentage of turnover realised on intra-group transfers of goods or provisions of services might serve as an indicator.65 4.2.1.2 Nexus A second necessary decision refers to the nexus of group members to different countries. In this respect, the concept of residence based on OECD principles is well established and should be maintained. Therefore, the nexus of a corporation could be either its seat or the effective place of management. 4.2.1.3 Personal Scope The personal scope of the CCCTB should include subsidiaries (i.e. corporations) and permanent establishments as group members. The difficulties caused by separate accounting if applied to an economically integrated group of companies are not only relevant for distinct legal entities but also for permanent establishments. In contrast, differentiating between subsidiaries and permanent establishments would entail scope for tax planning by not including relevant activities within the CCCTB and thus violate against the principle of neutrality. Moreover, an exclusion of permanent establishments could violate the freedom of establishment laid down in Article 42 of the EC Treaty.

62

63

64

65

See Oestreicher, Konzern-Gewinnabgrenzung: Gewinnabgrenzung – Gewinnermittlung – Gewinnaufteilung, München 2000, pp. 136-139. See Agúndez-García, The Delination and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, Taxation Papers, Brussels 2006, p. 14. See McLure, Defining a Unitary Business: An Economist’s View, McLure Jr., C. E. (Ed.), The State Corporation Income Tax: Issues in Worldwide Unitary Combination, Stanford 1984; Hellerstein, Allocation and Apportionment of Dividends and the delineation of the Unitary Business, in: Tax Notes, Special Report 1982. For a summary of test used to define a unitary business see Martens Weiner, Using the Experience in the U.S. States to Evaluate Issues in Implementing Formula Apportionment at the International Level, Office of Tax Analysis Paper 83, Washington, DC 1999, pp. 29-34. See Hellerstein, Tax Notes 1993; Hellerstein/McLure, International Tax and Public Finance 2004: p. 205.

34

Christoph Spengel

An important issue regarding the personal scope of the CCCTB concerns the treatment of transparent entities. This issue was already addressed by the Working Group of the European Commission.66 The general view so far is that transparent entities should not qualify as parent companies of a consolidated group. Although the principle of neutrality would call for an equal tax treatment of transparent entities and corporations, there are good reasons to exclude transparent entities from the application of the CCCTB. As outlined above, corporations are distinct legal entities which are taxed separately from their shareholders. In most countries the corporate income tax is not fully integrated into the personal income tax of the shareholders (see above, 2.1.3). By contrast, partnerships are not regarded as distinct legal entities. Not the partnership itself but, according to the transparency principle, each partner is subject to tax with his share in the total profits. If transparent entities would qualify as a parent of a tax group for CCCTB purposes, the share of the CCCTB apportioned to the transparent entity would be allocated directly to the partners and would be subject to personal income tax. This would contravene the distinction between corporate income tax and personal income tax drawn by corporate tax systems. Therefore, transparent entities should be excluded from the personal scope of the CCCTB in order to prevent national corporate tax systems from being evaded in the context of the CCCTB. However, member states could be granted an option whether to extend the personal scope of the CCCTB to transparent entities. An inclusion of transparent entities would also affect revenues from personal income tax. The fiscal effects of consolidation and formula apportionment could not be limited to corporate income tax. These effects are not intended by the European Commission, as the proposals for a consolidated tax base focus on corporate income tax rather than on personal income tax. According to the Commission, the personal income tax should remain in the sole responsibility of the member states.67 Although transparent entities should not qualify for the application of the CCCTB, they are still relevant. If a corporation included in the personal scope of the CCCTB is a partner of a transparent entity, its share of the income of the transparent entity should be included in the CCCTB. Accordingly, the relevant percentage of the factors of such transparent entities should be considered with respect to formula apportionment.68

66

67

68

See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on group taxation (SG – 05 & 06 February 2007), Working Document, Brussels 2007b, p. 5. See European Commission, Tax policy in the European Union – Priorities for the years ahead, Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee, COM(2001) 260 final, Brussels 2001, p. 17. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on group taxation (SG – 05 & 06 February 2007), Working Document, Brussels 2007, p. 5.

A. Concept and Necessity of a Common Tax Base – an academic introduction

35

4.2.2 Consolidation A key element essential to the concept of the CCCTB is that of consolidation. Consolidation raises questions with regard to the starting point of the consolidated tax base, intra-group loss relief, taxation of intra-group transactions, scope of income qualifying for consolidation, and treatment of EU outbound and inbound investment. The following addresses this questions consecutively. 4.2.2.1 Starting Point Since the EU-Regulation on consolidated accounts from 200269 obliges companies listed on an EU stock market to establish their consolidated accounts based on IFRS from 2005 onwards, for reasons of simplicity and practicability, consolidated IFRSaccounts could be used as a starting point for the determination of the CCCTB. This possibility was also considered by the European Commission in an earlier Communication.70 There are, however, certain arguments against this strategy. Both the definition and the territorial scope of the consolidated group are not satisfactory for tax purposes. Moreover, the method of consolidation is unlikely to be compatible with tax principles and current member states’ tax practice. Adjusting IFRS-group statements would be complex and is unlikely to reduce compliance costs, as intended by this approach. Therefore, consolidated IFRS-accounts cannot serve as a starting point for the CCCTB. By contrast, a tax-specific method of consolidation using the individual accounts as a starting point is preferable. This view follows member states’ practice with regard to consolidated taxation of domestic groups and has now been shared by the European Commission.71 Therefore, the starting point of the CCCTB should be the individual incomes of all members of the taxable unit computed according to a single set of tax accounting rules as outlined above (see 4.1). 4.2.2.2 Intra-group Loss Relief The offset of losses and profits between members of the taxable unit is achieved via consolidation. Adding together the individual tax bases of each group entity, losses incurred by one group entity are automatically offset against profits of other group entities. If the consolidated tax base is negative, the loss should be carried back to be offset against profits of prior years or carried forward to be offset against future profits. This is necessary to ensure the taxation of net income. Two possible mechanisms

69

70

71

See Regulation (EU) 1606/2002 by the European Parliament and the Council regarding the application of International Accounting Standards, OJ L 243/1. See European Commission, An Internal Market without company tax obstacles – achievements, ongoing initiatives and remaining challenges, Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee, Brussels 2003. See European Commission, The Application of International Accounting Standards (IAS) in 2005 and the Implications for the Introduction of a Consolidated Tax Base for Companies’ EUwide activities, Consultation Document, Brussels 2003, p. 21.

36

Christoph Spengel

for a loss carry over were discussed by the Commission’s working group on losses.72 The first approach would be to allocate the loss to each group member according to the apportionment mechanism. The allocated share of the group’s loss may then be carried back or forward at the level of each individual group member and offset against shares of a positive consolidated tax base in prior or future years. The second approach would be to carry back or forward such losses at the group level. The first approach seems to be more consistent with the concept of consolidation and formula apportionment. Although the approach of consolidation is based on the notion that a group of companies forms an economic unit, under the concept of the CCCTB each of the group members remains liable to tax for its apportioned share of the consolidated tax base on an individual basis. The first approach leads to a symmetric treatment of profits and losses. Not only a positive but also a negative consolidated tax base is apportioned to each group member. By contrast, the second approach results in an asymmetric treatment of profits and losses. Whereas a positive group income is subject to formula apportionment in the period it is earned, a negative consolidated tax base is carried forward at the group level and reduces future profits. As a consequence, losses are apportioned according to the allocation of apportionment factors in the period in which they are offset at the group level. If a negative CCCTB is – according to the preferred first approach – to be apportioned to the involved group entities and offset against shares of a positive consolidated tax base in prior or future years, common rules for loss carry back or loss carry forward in the EU are necessary in order to reduce incentives for tax planning. At present, the vast majority of member states do not provide for a loss carry back (see above, 2.1.2.6). It seems therefore more realistic, to allow only for a loss carry forward. In order to prevent that losses cannot be offset, there should be no time restrictions to the loss carry forward. Moreover, special rules are required in order to prevent companies from utilizing losses more than once. Losses may not only be consolidated at the group level but also be carried over in previous or subsequent years to be offset against profits in the individual account of the respective group member. There are two alternative approaches to avoid this so called “double dip”. Either, any carry forward or carry back of losses at the company level is added back to the consolidated tax base at the group level or the separate taxable income before compensation for loss carry forwards of each group member enters the consolidation. Both approaches would ensure that once losses are offset via consolidation they cannot be used at the company level. Another issue refers to pre-consolidation losses. Since under the concept of the CCCTB each group entity remains liable to tax with its share in the consolidated tax base on an individual basis, such losses should remain with the company that has incurred them and offset on a stand-alone-basis.73 72

73

See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – Progress to date and future plans for the CCCTB, Working Document, Brussels 2006, p. 12. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on group taxation (SG – 05 & 06 February 2007), Working Document, Brussels 2007, p. 6.

A. Concept and Necessity of a Common Tax Base – an academic introduction

37

4.2.2.3 Intra-group Transactions Another key element of consolidation is the deferred taxation of profits and losses stemming from intra-group transactions. The recapture of these profits and losses depends on the type of the respective asset. Profits and losses regarding intra-group transfers of non-depreciable assets should generally be deferred until the asset leaves the group, either by a sale or other disposition to a non-member. Intra-group income regarding depreciable assets is either restored in the course of depreciation or if the asset leaves the group. There are several different techniques with respect to the adjustments necessary to neutralise intra-group transactions. A first approach would be to ignore intra-group transactions for tax purposes. The seller’s basis in the asset (i.e. the tax book value) would simply carry over to the buyer in the intra-group sale. The company receiving the assets is obliged to compute any new depreciation and any profit or loss in respect of the asset transferred according to the seller’s basis in the asset ultimately before the transfer. As a result, profits or losses realised on an intra-group transaction are not documented for tax purposes. A second approach is to record intra-group transfers at their internal transfer price, both in the seller’s as well as the purchaser’s account. Any profit or loss realised upon the transaction is recorded as a balance sheet item and thus neutralised for tax purposes in the seller’s account. The acquiring company computes new depreciation or a capital gain on a subsequent transfer on the basis of the transfer price. If the intra-group profit or loss has to be restored because of a subsequent transfer or in the course of depreciation, the balance sheet item is dissolved. According to a third approach, each group member computes its individual taxable income taking into account all profits and losses realised upon an intragroup transfer. Thus, group member’s accounts correspond to those of unrelated companies. Instead of adjusting the individual accounts, intra-group profits and losses may be neutralised in a separate combined report. Taxable income of each group member is calculated as between unrelated companies. These individual results are combined and adjusted, eliminating intra-group profits or losses and any excess depreciation. This approach prevails in France and the United States for domestic group taxation. Although all three approaches lead to the same consolidated tax base in each year, only according to the second and third approach, intra-group transactions are recorded at their transfer price in the individual accounts. This seems beneficial in cases where intra-group income has to be restored, for example if a company involved in an intra-group transfer ceases to be a group member. The third approach has another advantage. As each group member’s separate account is determined as though the company was independent, the separate accounts may remain the basis for the claim of minority shareholders. As a result, there are good reasons for adjusting intra-group transactions in a separate combined report and determining the individual accounts using the method of separate accounting (third approach).

38

Christoph Spengel

4.2.2.4 Scope of Income Qualifying for Consolidation A guiding principle for the allocation of profits is to assign income to the source country. The source of profits is defined as the location where the profits are created.74 If the source of certain categories of income can be precisely determined, there is a strong argument to exclude this income from consolidation and apportionment and instead to assign this income to the tax jurisdiction having its source within its borders. The application of formula apportionment should be restricted to income arising from transactions and activities in the course of the economically integrated business, as the arm’s length principle fails to allocate the benefits from economic integration. A distinction between different categories of income is also made in the United States and in Canada. In both jurisdictions, the application of formula apportionment is restricted to the so called business income.75 Business income is generally referred to as any income arising from transactions and activities in the regular course of the business. Non-business income typically includes income such as royalties, dividends, interest receipts and rents. Whereas business income is consolidated and apportioned, non-business income is directly assigned to a tax jurisdiction. Despite that a distinction between different types of income seems justified from a theoretical point of view, there are some disadvantages to this approach.76 Separation between categories of income can get difficult and might be subject to manipulations, as deciding whether activities and resulting income belong to the economic unit is not an easy task. Moreover, a consequent application of this approach would require a separation of expenses aligned to the different categories of income as well. Taxing interest and licenses according to the current member states’ practice would enable companies to maintain their strategies of shifting income thus assigning expenses to high tax countries and corresponding income to low tax countries. A final problem linked to the distinction approach is that it removes one of the advantages of consolidation, namely the fact that loss offset disregarding the categories of income. Instead, losses incurred from the non-apportionable categories of income, excluded from the consolidated base, would not benefit from general intra-group cross-border loss compensation and therefore an additional system of loss transfers might be required. Taking all these arguments into account, it might be more apt not to differentiate between different categories of income.

74 75

76

See Jacobs, Spengel and Schäfer, Intertax 2004, p. 270. See Martens Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, Taxation Papers, Brussels 2005, p. 66. See Argúndez-Garzía, The Delination and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, Taxation Papers, Brussels 2006, pp. 17-20.

A. Concept and Necessity of a Common Tax Base – an academic introduction

39

4.2.2.5 Treatment of EU Outbound and Inbound Investments The application of the CCCTB is limited to the boundaries of the EU.77 For business activities between member states and third countries, separate accounting using arm’s length pricing remains.78 With respect to third countries, outbound and inbound investments can be distinguished. Typical patterns for outbound investments are an EU parent company doing direct investment in a third country through a permanent establishment or a subsidiary. Between parent and subsidiary different income flows may arise. The parent company may receive dividends, interest income, or royalties. Typical settings for inbound investments are a non-EU company doing direct investment in the EU either maintaining a permanent establishment or a subsidiary. Again, the non-EU company may receive dividend payments, interest income or royalties from its EU subsidiary. A starting point for the tax treatment of inbound and outbound investment with respect to third countries is the current tax practice as codified in the national tax laws and existing tax treaties. By contrast, any renegotiation or amendment to existing tax treaties is regarded as last resort solution.79 At present, tax rules concerning inbound and outbound investments differ to a substantial degree (see above, 2.1.2.7). Differences exist regarding various aspects. On the one hand, member states treat the same category of income differently. With regard to foreign permanent establishments some member states apply the exemption method either with or without loss relief while other member states apply the credit method to avoid double taxation. Similarly, dividends from a foreign subsidiary are either tax exempt or taxed and a foreign tax credit is granted. On the other hand, member states apply different approaches depending on the category of income. Interest on loans or royalties on intellectual property are subject to residence based taxation while source based taxation is rather applied to profits of a foreign permanent establishment or subsidiary. With respect to inbound investments, withholding taxes are levied at source at different rates. Furthermore, member states apply different anti-avoidance rules in order to prevent the erosion of the tax base. Due to these differences, it seems appropriate to exclude foreign income earned in third countries from the CCCTB and instead, to assign the taxing right solely to the jurisdiction of residence of the individual group member receiving the income. Otherwise, member states would have to accept each other’s tax rules regarding cross-border investment. For example, if one member state grants cross-border relief for losses incurred by a foreign permanent establishment, foreign losses would be consolidated thereby reducing the apportionable consolidated tax base and thus, the 77

78

79

See European Commission, Company Taxation in the Internal Market, Brussels 2001, pp. 482486, 498. Separate accounting under the arm’s length principle would also have to prevail with regard to EU affiliates resident in member states that do not apply the CCCTB and EU affiliates excluded from the consolidated group. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on International aspects (SG4), Working Document, Brussels 2006, p. 3.

40

Christoph Spengel

share in the consolidated tax attributable to other involved countries. Similar conflicts may arise, if the exemption method is meant to relief double taxation in one member state while the credit method is applied in another member state. Although, maintaining current tax practice seems attractive from a pragmatic point of view, there are good reasons to strive for common rules with respect to third country issues, at least in the long run. If different national rules are to be followed with respect to cross-border investments between EU group entities and non-EU group entities, this may facilitate strategic tax planning. EU companies might reorganise their businesses within the EU in such a way that cross-border investments in third countries are subject to the most beneficial rules provided by a member state. For instance, direct investment in a low-tax country via a permanent establishment or a subsidiary would be undertaken from a company resident in a member state providing the exemption method as a means to avoid double taxation. Similar tax planning strategies might occur with respect to non-EU companies investing in the EU if the aim is to reduce or avoid withholding taxes on income from inbound investments. Therefore, a CCCTB maintaining differing current tax practice with respect to third country issues is not neutral. As companies can reorganise their businesses in order to realise tax advantages, anti-avoidance rules applied by some member states are likely to be evaded. Thincapitalisation rules, for instance, could be evaded, if a non-EU company first grants a loan to a subsidiary resident in a country without thin-capitalization rules, and afterwards this loan is directed to the relevant company via intra-group transactions. This is facilitated by consolidation, as intra-group transactions are netted out and thus, are not subject to tax. As a consequence, it becomes difficult for member states to enforce their tax rules, especially if they are less favourable than comparable rules of other member states applying the CCCTB. Moreover, with different national tax rules regarding third country issues, multinational companies doing business not only within the internal market but also in third countries would still have to cope with 27 different tax systems. To conclude, at least in the long run, common rules for taxing inbound and outbound investments with respect to third countries are recommended.80 4.2.3 Entry and Exit Rules Entry and exit rules refer to the taxation of hidden reserves (i.e. difference between market value and tax book value of an asset) in the event companies are joining or leaving the taxable unit. Hidden reserves built before a company joins the group do not belong to the group. Instead, they belong to the individual company and, therefore, the respective tax jurisdiction preserves its taxing right. These taxing rights may be at risk since hidden reserves realised in a subsequent year are apportioned to all group members. The preservation of the taxing right of the country of origin of the hidden reserves under the CCCTB is similar to the one in the event of reorganisations (without having a permanent establishment in the relevant jurisdiction after80

See Hellerstein/McLure, International Tax and Public Finance 2004, p. 207; Mintz, International Tax and Public Finance 2004, p. 228.

A. Concept and Necessity of a Common Tax Base – an academic introduction

41

wards) or the cross-border transfer of assets to a permanent establishment. Therefore, the difference between the fair value of assets and liabilities and their value for tax purposes should be treated like an intra-group transaction. Group members are not to be taxed at the time of the transaction. If the intra-group income is restored in subsequent years, an amount equal to the hidden reserves assessed in the year the CCCTB is applied the first time should be directly allocated to the respective tax jurisdiction. If a company leaves the group or the group terminates, losses apportioned to each group member during the group phase should remain with each company. This is because each group member preserves its own legal capacity irrespective whether taxable income is apportioned or not. Moreover, if one of the group companies engaged in a former intra-group transaction leaves the group, this should be deemed as a restoration event. If the internal purchaser of the asset afterwards leaves the group, not only the purchaser but also the asset leaves the group. Therefore, this event should be treated in the same way as a sale of the asset to an unrelated party. In addition, this approach would prevent tax planning strategies aimed to avoid taxation of capital gains by transferring assets into a subsidiary and selling the shares of the subsidiary with full participation exemption.81 If the seller leaves the group, a previously deferred intra-group profit or loss should also become subject to tax. As the seller and the buyer do not belong to the same group, the requirement for deferred taxation is not met anymore. Another reason for this provision again refers to tax planning. Companies might avoid taxation of capital gains in a certain country by joining the group, transferring assets into other group members and afterwards leaving the group. In order to ensure that taxing rights of the country of residence are not at risk, the leaving company has to be subject to tax on the gain which is assessed as the difference between the seller’s basis in the asset and the arm’s length price at the time of the original transfer. To conclude, it is necessary to introduce a mechanism with regard to companies and assets entering and leaving the group. Such entry and exist rules serve to secure member states taxing rights for hidden reserves built up before (entry) and during (exit) consolidation. Therefore, separate entity accounting has to be maintained in addition to formula apportionment.

4.3 Apportionment of Group Income The idea of formula apportionment is to allocate the consolidated tax base to the member states according to the source principle (see above, 3.1). From this perspective, the main objective of profit allocation is to find apportionment factors which are linked to the economic activities of multinationals in different jurisdictions and

81

See European Commission, Implementing the Community Programme for improved growth and employment and the enhanced competitiveness of EU business: Further Progress during 2006 and next steps towards a proposal on the Common Consolidated Corporate Tax Base (CCCTB), Communication from the Commission to the European Parliament, the Council and the European Economic and Social Committee, COM(2007) 223 final 2007, p. 9.

42

Christoph Spengel

which are, at the same time, regarded as a fair share in the tax base by the member states. The ongoing discussion shows that a distinction can be made between macrofactors and micro-factors.82 Macro-factors can be based on overall economic ratios (e.g. GDP) or on sector ratios. The advantage of macro-factors is that they cannot be influenced by individual companies; opportunities for tax planning are eliminated in this respect. In individual cases, however, this could lead to a situation in which companies pay low taxes in one member state, even though they are very economically active there, or conversely pay high taxes in another member state where they are hardly economically active at all. It is therefore questionable whether the member states will accept distribution rules that are divorced from the activities of a resident enterprise and thus also completely detached from existing conventions of source taxation. Therefore, micro-factors are probably preferable. Following the example of formula apportionment operated in the United States and in Canada, micro-factors could be derived from company-specific production factors such as proportional property, payroll, and sales.83 The theoretical justification for the use of production factors stems from the source country’s entitlement to tax income. Company assets represented by the property factor and total wages represented by the payroll factor, therefore, seem generally suitable as criteria. Taking into account the fact that most of multinational companies’ assets consist of intangible assets such as patents, licenses, software and know-how, the inclusion of total wages is also justified because it serves as substitute for measuring intangible capital. The idea behind the reference to sales is that profits are only generated by the interaction of supply and demand (market-oriented source principle), so that in the case of cross-border transactions both the production location and the state where there is demand for the product have a claim on the company's profits. The use of all three of these company-specific ratios should therefore be considered. Value-added has been proposed as an alternative micro-factor.84 Compared to company-specific production factors, value-added has the advantage that companies activities are recorded more broadly. Moreover, value-added can be derived comparatively simply from the value added tax base, which is harmonized throughout the EU. However, the distribution of group profits according to the value-added of the group members causes problems if goods and services are delivered within the taxable unit. This is because, in this case, the determination of value-added presupposes that the services exchanged by the group companies are valued. This in turn requires transfer prices, which is precisely what the CCCTB is supposed to avoid.

82

83

84

See European Commission, Company Taxation in the Internal Market, Brussels 2001, pp. 407415, Argúndez-García, The Delination and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, Taxation Papers, Brussels 2006, pp. 39-87. See Hellerstein/McLure, International Tax and Public Finance 2004, pp. 199-220; Martens Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, Taxation Papers, Brussels 2005. See Lodin/ Gammie, Home State Taxation, Amsterdam 2001.

A. Concept and Necessity of a Common Tax Base – an academic introduction

43

To conclude, for the apportionment of group income the allocation formula should be based on a broad set of company-specific production factors such as proportional property, payroll, and sales which must be defined uniformly across member states. A uniform definition of these factors is necessary to exclude situations of both double taxation and lower taxation. Theory provides no clear answer for the weight of the company-specific production factors. A political agreement is therefore required on the weighting of these factors. Macro-factors and micro-factors, if derived from value-added, do not seem to be appropriate.

4.4 Related Issues 4.4.1 Legal Obligation Related issues first concern the legal obligation of a CCCTB. In order to administrate such a system, to restrict tax arbitrage85 and to avoid discrimination of purely domestic activities, a CCCTB should be compulsory (i.e. without option) and applied to both domestic and international companies. However, the European Commission intends to implement the CCCTB as an optional system. Its decision is based on the notion that the CCCTB provides companies with rules allowing them to benefit from the single market.86 Furthermore, the CCCTB is a very far reaching reform. Currently, consolidation including a deferred taxation of gains arising on intra-group transfers is only available in seven Member states (see above, 2.1.2.8). Consolidation and formula apportionment could be very complex and associated with high administrative costs. Therefore, companies should have the option of remaining with the current tax system in order to avoid compliance costs associated with the CCCTB. If the CCCTB is applied as an optional scheme, the option should be obligatory for a certain period of time in order to prevent taxpayers from having the choice between the two systems each year.87 Two contrasting approaches exist regarding the choice of members in the election. One requires an “all-in or all-out” treatment. If the group decides to apply for the regime, all members that meet the shareholding and other requirements must join. Another treatment allows “cherry-picking” of some members among the potential members that meet the shareholding requirements. If the application of the CCCTB is optional, all eligible entities should be consolidated. Cherry-picking as a means of tax planning should not be allowed.

85

86

87

See Mintz, International Tax and Public Finance 2004, pp. 226; Sørensen, International Tax and Public Finance 2004, p. 102. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on group taxation (SG – 05 & 06 February 2007), Working Document, Brussels 2007, p. 2. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on group taxation (SG – 05 & 06 February 2007), Working Document, Brussels 2007, p. 3.

44

Christoph Spengel

The question of whether the application of the CCCTB should be optional or mandatory is closely related to the definition of the taxable unit (see above, 4.2.1.1).88 If the criteria used to define the consolidated group can be adjusted without costs, companies would have a factual option regarding the entities to be included in the CCCTB. In this case, a compulsory scheme may not be effective. Similarly, if the application of the CCCTB is optional, the importance of a definition of the consolidated group resistant to tax planning would decrease. 4.4.2 Local Profit Taxes, Non-profit Taxes and Social Security Contributions In most member states, companies are subject to taxes other than the corporate income tax. These taxes are either imposed on income (profit taxes) or business factors (non-profit taxes), such as payroll or property (see above, 2.1.4). A CCCTB raises several issues regarding these additional taxes. First, the consolidation and apportionment approach may affect the determination of the tax base of local taxes. Local profit taxes would have to be levied on the share in the common tax base ascribed to the respective member state. This, however, will probably not cause conceptual problems, since both, the concept of the CCCTB and the determination of the tax base of local profit taxes follows the source principle. Another issue to be considered refers to the deductibility of local taxes.89 At present, most member states allow companies to deduct local taxes from the corporate income tax base. The CCCTB working group on the definition of income has elaborated a definition of deductible expenses in general. Deductible expenses are defined as decreases in economic benefits in the form of outflows which are made for business purposes and are incurred during the taxable period.90 Exceptions to this definition are to be shown in a special list. Regarding taxes, there seems to be a consensus to exclude the corporate income tax and VAT.91 Local taxes represent a mandatory cost incurred in the course of business. Thus, with reference to the general definition of deductible expenses, local taxes should be deductible. There are two alternative approaches to deducting local taxes. They can either be deductible against the CCCTB before apportionment or they can be deductible against the share of the CCCTB after apportionment.92 If local taxes are deducted from the CCCTB 88 89

90

91

92

See the contribution of Wolfgang Schön in this volume. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the fourth meeting of the subgroup on taxable income (SG3), Working Document, Brussels 2006, pp. 4-7. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on taxable income (SG3), Working Document, Brussels 2006, p. 5. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the third meeting of the subgroup on taxable income (SG3), Working Document, Brussels 2006, p. 6. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the fourth meeting of the subgroup on taxable income (SG3), Working Document, Brussels 2006, p. 6.

A. Concept and Necessity of a Common Tax Base – an academic introduction

45

before apportionment, the share of all participating member states will decrease. Therefore, mutual recognition of the different national practices as regards the division of competences and the methods of raising revenues seems to be a prerequisite. However, member states may not accept to provide relief for taxes levied by another member state. Moreover, there is also an incentive inherent to this approach for funding public needs by imposing local taxes because these taxes are effectively paid not only by the company resident in the respective jurisdiction but also by other participating countries. In order to avoid mutual agreement of different national tax systems, local taxes should be deductible against the share of the CCCTB after apportionment. This approach would be also consistent with the source principle. When regarding local taxes levied on property and payroll, these taxes are closely related to apportionment factors. Similar issues regarding the deducibility of expenses arise in the context of social security contributions.93 Generally, social security contributions should be tax deductible. However, it seems questionable whether member states would accept each other’s practice and level of social security system. Member states may not accept to provide relief for higher social standards provided by other countries. Therefore, similar to local taxes, social security contributions should be deductible from the consolidated tax base after apportionment. However, one has to bear in mind, that mutual recognition is already implied regarding different national levels of expenses such as wages including costs for occupational pension schemes or rental fees, since these expenses may be influenced by national regulations and habits. Thus, it would make conceptual sense not to distinguish between different kinds of expenses and instead to deduct social security contributions as well as local taxes from the consolidated tax base before apportionment. To conclude local profit and non-profit taxes as well as social security contributions should be tax deductible against the consolidated tax base. Whether these expenses reduce the CCCTB before or after apportionment mainly depends on the willingness of member states to accept each others national practices concerning the level and methods of raising revenues. As regards the determination of the base of local taxes, the introduction of a CCCTB would not lead to serious problems.

5. In Favour of a Minimum Corporate Tax Rate Formula apportionment would introduce a new concept of company taxation in the EU. It operates as a direct tax on the allocation factors (i.e. on property, payroll, and sales). According to the mechanism of formula apportionment, the effective tax rate applied to the consolidated tax base is a weighted average of the tax rates in the different jurisdictions in which group entities operate, with the weights given by the

93

See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG) – An overview of the main issues that emerged at the fourth meeting of the subgroup on taxable income (SG3), Working Document, Brussels 2006, p. 4.

46

Christoph Spengel

presence of the apportionment factors of the firm in each jurisdiction relative to its total factors.94 Formula apportionment is not immune against strategic tax planning. If the apportionment factors are mobile across jurisdictions and the company controls them there is leeway for profit shifting. There might be an incentive for companies to relocate economic activities that comprise the factors to low tax jurisdictions in order to reduce the effective tax rate.95 Moreover, they may invest in companies which are resident in low tax jurisdictions even if these companies are not profitable because due to the investment the effective tax rate applied to the consolidated tax base and thus the overall tax burden would decline.96 By reallocating business functions entering the allocation formula multinationals are able to shift a greater portion of the tax base to other member states compared to the prevailing division of the tax base applying transfer prices. Such a reallocation is promoted by a CCCTB since the transfer of functions is no taxable event due to the elimination of inter-company profits. Given the considerable EU-wide range of nominal tax rates, tax competition within the EU presumably will increase (tax rate differentials in the EU are by far more relevant than in Canada and the USA). Increasing tax competition might provoke member states to further tax cuts on corporate profits. Moreover, it worsens the proper functioning of the Common Market if decisions where to locate investments are mainly tax driven. This inevitably raises the question whether a CCCTB should be combined with a minimum tax rate on corporate profits. A minimum corporate tax rate has two objectives: it protects an efficient allocation of resources and, thus, the economic goals of the EC Treaty. Moreover, it protects the autonomy of member states with respect to the personal income tax. Therefore, a harmonised tax base combined with a minimum corporate tax rate serves as a compromise between economic efficiency in the EU and member states’ tax autonomy.97

6. Conclusion (1) The analysis of the present situation of taxing cross-border activities within the EU reveals several serious problems. There is no proper functioning of the Common Market in economic terms, there exist obstacles to cross-border activities of busi94 95

96

97

See McLure, PQF 1981, pp. 395-413; Mintz, FA 1999, pp. 406-407. See Martens Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, European Commissions’ Taxation Papers, Working Paper No. 8, Brussels 2005, pp. 38-50; Argúndez-Garzía, The Delination and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, Taxation Papers, Brussels 2006, pp. 59-69. See Gérard/Martens Weiner, Cross-border loss offset and formulary apportionment: How do they affect multijurisdictional firm investment spending and interjurisdictional tax competition?, CESifo Working Paper No. 1004, Munich, 2003. See Sørensen, To harmonise or not to harmonise?, CESifo Forum 2002, p. 35; Ruding, ECTR 2005, pp. 2-3; Wissenschaftlicher Beirat beim Bundesministerium der Finanzen, Einheitliche Bemessungsgrundlage der Körperschaftsteuer in der Europäischen Union, Gutachten, Berlin 2007, pp. 72-75.

A. Concept and Necessity of a Common Tax Base – an academic introduction

47

nesses and many of the member states’ provisions with respect to the taxation of cross-border activities are in conflict with EU law. (2) A CCCTB might help to abolish current tax obstacles to cross-border activities within the EU, reduce conflicts with EU law and increase member states’ tax autonomy. (3) For the implementation of a CCCTB, it is necessary to introduce harmonised rules for the determination of individual taxable income and the determination of group income including common standards for the definition of the group, the process for consolidation as well as for entry and exit rules. Moreover, common standards for the allocation of group income are necessary. Finally, agreements are required with respect to the legal obligation of a CCCTB and the treatment of local profit taxes, non-profit taxes and social security contributions. (4) A CCCTB can only reach its objectives in total if it is accompanied by a minimum corporate tax rate. Therefore, the idea of a CCCTB is very far reaching. Member states have to balance the need for harmonisation against the existing problems of taxing EU multinationals. European experience from VAT with a uniform tax system, a harmonised tax base and a fixed band of tax rates indicates that the allocation of the tax base to the member states is the most difficult task.

B. Perspektiven der Konzernbesteuerung

49

B. Perspektiven der Konzernbesteuerung Wolfgang Schön*

1. Einleitung Für das Unternehmen, das Bild einer künftigen Konzernbesteuerung zu entwerfen, empfiehlt es sich, in drei Schritten vorgehen. Er müssen zunächst die grundlegenden Ansprüche definiert werden, denen ein solches Regelungswerk zu genügen hat. Diese Ansprüche sind vielfältiger Natur: Aus der Sicht des Juristen stehen in vorderster Linie die Vorgaben höherrangigen Rechts – insbesondere des Grundgesetzes und des EG-Vertrages. In ihrer Ausformulierung durch das Bundesverfassungsgericht und den Europäischen Gerichtshof bilden die dort niedergelegten Grundrechte und Grundfreiheiten inzwischen einen engen Rahmen für die Gesetzgebung auf dem Gebiet des deutschen Unternehmenssteuerrechts. Innerhalb dieses Rahmens ist es Aufgabe der politischen Instanzen, die inhaltliche Zielsetzung eines Konzernsteuerrechts zu definieren. Hier kommen ganz unterschiedliche Tendenzen in den Blick: Man kann eine Vorzugsbehandlung für Konzerne anstreben und Deutschland im Wettbewerb der Rechtsordnungen als „Holdingstandort“ Attraktivität verleihen. Man kann einen kritischen Standpunkt einnehmen und Konzernstrukturen als grundsätzlich manipulationsanfällig und aufkommensschädlich einordnen. Man kann schließlich eine steuerpolitische Mittellage präferieren und darauf abzielen, der jeweiligen Wahl einer gesellschaftsrechtlichen Konzernstruktur weder besonders günstige noch besonders nachteilige Rechtsfolgen zuzuordnen. Auf der Grundlage dieser Ansprüche ist in einem weiteren Schritt der Zustand des geltenden Rechts zu bewerten und Reformbedarf zu formulieren. Diese Bewertung darf sich nicht auf solche Steuerregeln beschränken, die gezielt und ausschließlich für Konzerne gelten wie etwa die Vorschriften des Körperschaftsteuerrechts über die Organschaft. Man muss vielmehr weitergehend alle diejenigen Vorschriften des Ertragsteuerrechts in den Blick nehmen, die auf Unternehmensgruppen Anwendung finden, im geltenden Recht etwa die Steuerfreiheit von Gewinnausschüttungen zwischen Kapitalgesellschaften, die Sonderregeln über den Beteiligungsaufwand, die Kontrolle verdeckter Gewinnausschüttungen oder die Gewinnrealisierung bei Umstrukturierungen. Dabei wäre man von vornherein zum Scheitern verurteilt, wenn man die Validität des geltenden Konzernsteuerrechts ausschließlich am inländischen Konzern verproben wollte. Nicht nur die wirtschaftliche Realität internationaler Konzerne (auch im Mittelstand), sondern zugleich die Forderungen des europäischen Gemeinschaftsrechts nach diskriminierungsfreier *

Es handelt sich um einen Vortrag, den der Verf. im Januar 2007 auf dem Symposium „Zukunft des Konzernrechts“ in Königstein/Taunus gehalten hat (Erstveröffentlichung in ZHR 171 (2007), S. 409 ff.).

50

Wolfgang Schön

Besteuerung grenzüberschreitender Unternehmensstrukturen verlangen von dem Gesetzgeber, das Konzernsteuerrecht von vornherein europäisch, wenn nicht sogar „global“ anzulegen. Je nachdem, wie das Urteil über den Stand der Konzernbesteuerung ausfällt, wird man zu inkrementalen oder grundstürzenden Reformvorschlägen gelangen. Zu den revolutionär zu nennenden Konzepten gehört die Idee einer einheitlichen europäischen Steuerbemessungsgrundlage für Konzerne, welche die Europäische Kommission im Jahre 2001 in einer großangelegten Studie1 skizziert und in den vergangenen fünf Jahren in Arbeitsgruppen2 und Zwischenberichten 3 ausdifferenziert hat. Dieses Konzept hat erhebliche Aufmerksamkeit in der europäischen und US-amerikanischen Fachdiskussion gefunden; es stellt vertraute Grundannahmen sowohl des nationalen Steuerrechts als auch des internationalen Doppelbesteuerungsrechts in Frage und verdient daher erhebliche Aufmerksamkeit. Ihm ist der Hauptteil meiner Ausführungen gewidmet.

2. Ansprüche an eine Konzernbesteuerung 2.1 Rechtliche Ansprüche 2.1.1 Vorgaben des Grundgesetzes Es gehört zu den Selbstverständlichkeiten steuerpolitischer Arbeit, sich zunächst der verfassungsrechtlichen Rahmenbedingungen zu vergewissern, welche dem Gesetzgeber nicht nur durch die Kompetenznormen der Finanzverfassung, sondern vor allem durch die Grundrechte, namentlich den allgemeinen Gleichheitssatz des Art.3 Abs.1 GG, gesetzt werden. Im Zentrum steht dabei die Frage, ob und in welchem Umfang das Konzernsteuerrecht für bestimmte Unternehmensstrukturen eine Vorzugsbehandlung anordnen darf, inwiefern also für Kapitalgesellschaften allgemein oder für bestimmte Gesellschaftstypen oder bestimmte Beteiligungsverhältnisse Sonderregeln eingeführt werden dürfen. Eine erste Vorentscheidung hat das Bundesverfassungsgericht vor wenigen Monaten dahin getroffen, dass ein strenges Gebot der Gleichbehandlung von Personenunternehmen und Kapitalgesellschaften nicht existiert4. Die Begründung des 1

2

3

4

Mitteilung der Kommission an den Rat, das Europäische Parlament und den Wirtschafts- und Sozialausschuss v. 23.10.2001, Ein Binnenmarkt ohne steuerliche Hindernisse: Strategie zur Schaffung einer konsolidierten Körperschaftsteuer-Bemessungsgrundlage für die grenzüberschreitende Unternehmenstätigkeit in der EU, KOM (2001) 582 endg.; Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681. Vollständige Dokumentation der Sitzungsprotokolle auf der website der Kommission unter http://ec.europa.eu/taxation_customs/taxation/company_tax/common_tax_base/index_de.htm. Zuletzt: Arbeitsgruppe „Gemeinsame Konsolidierte Körperschaftsteuer-Bemessungsgrundlage“ (AG GKKB), Bisherige Fortschritte und künftiges Programm für die GKKB, 20.11.2006 (CCCT\WP\046\doc\de). BVerfG v. 21.6.2006, 2 BvL 2/99, DStR 2006, S.988 ff., Rz. 117 ff.; aus der Diskussion im juristischen Fachschrifttum siehe u. a. ausführlich zur Thematik der Rechtsformneutralität Hey, 24 DStJG (2001), S. 155 ff.; Jachmann, 23 DStJG (2000), S. 9 ff., 11 ff.; Kirchhof, StuW 2006, S. 3 ff., 11 f.; Sieker, 26 DStJG (2003), S. 145 ff., 149 ff.

B. Perspektiven der Konzernbesteuerung

51

Gerichts, dass die Steuergesetzgebung „verfassungsrechtlich bedenkenfrei die zivilrechtliche Grundentscheidung auf(nimmt), nach der bei Personengesellschaften das Gesellschaftsvermögen den Gesellschaftern zugerechnet wird (vgl. § 718 BGB i. V. m. § 105 Abs.3, § 161 Abs.2 HGB), während das Vermögen der Kapitalgesellschaften gegenüber dem Vermögen ihrer Gesellschafter grundsätzlich selbständig ist“5, wird in diesem Raum vermutlich nur wenige überzeugen. Dennoch ist mit dieser Entscheidung grundsätzlich der Weg für die im europäischen Ausland schon weitgehend vollzogene Ablösung der Besteuerung von Kapitalgesellschaften vom allgemeinen Einkommensteuerrecht freigemacht6. Es erscheint zulässig, ein Sonderrecht für Konzerne auf den Bereich der Kapitalgesellschaften zu beschränken. Verfassungsrechtlich angreifbar erscheinen allerdings Überlegungen, für die Konzernbesteuerung innerhalb des Kreises steuerpflichtiger Kapitalgesellschaftsformen weitergehende Differenzierungen einzuführen. Problematisch sind etwa die in den letzten Jahren diskutierten Vorschläge, für die Europäische Aktiengesellschaft als grenzüberschreitend angelegter Rechtsform ein präferentielles Sonderrecht zu etablieren7 oder den Zugang zu der bereits erwähnten europäischen Bemessungsgrundlage nur grenzüberschreitenden Konzernverbindungen zu gewähren. Eine solche Vorzugsbehandlung ist nicht durch Unterschiede in der steuerlichen Leistungsfähigkeit der jeweiligen Gesellschaften legitimiert und erscheint daher als mit dem in Art.3 Abs.1 GG niedergelegten Grundsatz steuerlicher Belastungsgleichheit nicht vereinbar 8. Sie würde wohl auch die in Art.9 Abs.1 GG und Art.14 Abs.1 GG angelegte freie Wahl zwischen verschiedenen Rechtsformen unzulässig beeinträchtigen. Diskutiert wird auch darüber, ob neuartige Konzernsteuerregeln gleichsam „auf Probe“ nur bestimmten Gesellschaften – etwa kleinen und mittleren Unternehmen – zur Verfügung gestellt werden sollen, um auf der Grundlage dieser Erfahrungen über eine flächendeckende Einführung zu entscheiden9.

5 6

7

8

9

BVerfG v. 21.6.2006, 2 BvL 2/99, DStR 2006, S. 988 ff., Rz. 118. Skeptisch zur Rechtsformneutralität aus betriebswirtschaftlicher Sicht Schneider, Steuerlast und Steuerwirkung, 2002, S. 215 ff.; Wagner, StuW 2006, S. 101 ff. Mitteilung der Kommission an den Rat, das Europäische Parlament und den Wirtschafts- und Sozialausschuss v. 23.10.2001, Ein Binnenmarkt ohne steuerliche Hindernisse: Strategie zur Schaffung einer konsolidierten Körperschaftsteuer-Bemessungsgrundlage für die grenzüberschreitende Unternehmenstätigkeit in der EU, KOM (2001) 582 endg., Abschn.5; Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss: Ein Binnenmarkt ohne unternehmenssteuerliche Hindernisse, Ergebnisse, Initiativen, Herausforderungen, 24.11.2003, KOM (2003) 726 endg., Tz. 4.5; kritisch González Sánchez/Franch Fluxa, European Taxation 2006, S. 197 ff., 201 ff.; Spengel in: Herzig (Hrsg.), Besteuerung der Europäischen Aktiengesellschaft, 2004, S. 101 ff., 115 f. Auch wenn das europäische Recht eine solche Differenzierung vorsehen würde, wäre der nationale Gesetzgeber durch Art. 3 Abs.1 GG gehalten, im Rahmen der verbleibenden eigenen Rechtssetzungskompetenz die bestimmten Rechtsformen gewährten Vorteile auf vom Europarecht nichtbegünstigte Gesellschaften auszudehnen. Non-Paper der Kommission v. 7.7.2004 für die informelle Tagung des Rates „Wirtschaft und Finanzen“ am 10. und 11.9.2004: Sitzlandbesteuerung bei kleinen und mittleren Unternehmen.

52

Wolfgang Schön

2.1.2 Vorgaben des EG-Vertrages Deutlich enger als die Anforderungen des deutschen Verfassungsrechts sind die Klammern, die das europäische Gemeinschaftsrecht der nationalen Gesetzgebung auf dem Gebiet der Konzernbesteuerung anlegt. Hier spielen die Grundfreiheiten des EG-Vertrages die primäre Rolle10. Die freie Gründung grenzüberschreitender Konzernverbindungen ist im Grundsatz durch die Niederlassungsfreiheit des Art. 43 Abs.1 EG garantiert, auf die sich nach Art.48 Abs.1 EG auch Kapitalgesellschaften berufen können11. Hinzu treten die Freiheit des Kapitalverkehrs – vor allem für Portfolio-Beteiligungen – und die Freiheiten der Dienstleistung und des Warenverkehrs für grenzüberschreitende Transaktionen innerhalb und außerhalb des Konzerns. Der Stand der Rechtsprechung des Europäischen Gerichtshofs lässt sich dahin zusammenfassen, dass grenzüberschreitende Konzernverbindungen weder vom Sitzstaat der Muttergesellschaft noch vom Staat der Tochtergesellschaft im Vergleich zu Inlandssachverhalten benachteiligt werden dürfen. Anknüpfungspunkt für die Kontrolle am Maßstab der Grundfreiheiten ist dabei stets der isolierte Blick auf das jeweilige nationale Steuerrecht. Kein Verstoß gegen die Grundfreiheiten lässt sich demgegenüber ausmachen, wenn Konzernverbindungen durch bloße Rechtsunterschiede zwischen den Mitgliedstaaten Nachteile erleiden – etwa unterschiedliche Gewinnermittlungsregeln gelten oder unterschiedliche Entlastungssysteme für körperschaftsteuerliche Vorbelastungen eingreifen. Schließlich bildet – nach jüngst bestätigter Rechtsprechung des Gerichtshofs12 – der bloße Tatbestand der Doppelbesteuerung von Unternehmensgewinnen in mehreren beteiligten Staaten keinen Verstoß gegen die Grundfreiheiten, sondern muss unilateral oder im Wege von Doppelbesteuerungsabkommen oder europäischen Richtlinien beseitigt werden. Zu diesen Richtlinien gehören die Mutter-Tochter-Richtlinie aus dem Jahre 199013 und die Richtlinie über Zinsen und Lizenzgebühren im Konzern aus dem Jahre 200314, die innerhalb der europäischen Union den Verzicht auf jegliche Quellensteuer auf grenzüberschreitende konzerninterne Dividenden-, Zins- und Lizenzgebührenzahlungen festschreiben. Nicht übersehen darf ein Gesetzgeber auf dem Gebiet der Konzernbesteuerung schließlich das Verbot steuerlicher Beihilfen15. Dieses ist in den vergangenen Jahren mehrfach aktualisiert worden, wenn nationale Steuergesetzgeber gezielt bestimmten Konzernfunktionen Vorteile angeboten hatten. Beispiele sind die belgischen

10

11 12 13

14

15

Überblick bei Schön, IStR 2004, S. 289 ff.; ausführlich M. Lang, Die Rechtsprechung des EuGH zu den direkten Steuern, Wiener Vorlesungen, 2007, S. 19 ff.; zur Bedeutung der Grundfreiheiten in Konzernverhältnissen siehe Witt, Die Konzernbesteuerung, 2006, S. 115 ff., 124 ff. Schön, EWS 2000, S. 281 ff., 283. EuGH v. 14.11.2006, Rs. C-513/04 (Kerckhaert-Morres), IStR 2006, S. 66 ff., Rz. 20 ff. Richtlinie 90/435/EWG über das gemeinsame Steuersystem der Mutter- und Tochtergesellschaften verschiedener Mitgliedstaaten. Richtlinie 2003/49/EG über eine gemeinsame Steuerregelung für Zahlungen von Zinsen und Lizenzgebühren zwischen verbundenen Unternehmen verschiedener Mitgliedstaaten. Überblick bei Schön, State Aid in the Area of Taxation, in: Hancher/Ottervanger/Slot, E.C. State Aids, 3. Aufl., 2006, S. 241 ff.

B. Perspektiven der Konzernbesteuerung

53

Regeln über Koordinierungszentren internationaler Konzerne16. Soweit ein Gesetzgeber jedoch allgemein günstige Steuerregeln schafft (etwa die Freistellung von Auslandsdividenden nach deutschem Steuerrecht), bewegt er sich auch dann außerhalb des Verbots selektiver Beihilfen, wenn diese sich auf die Konzernbildung günstig auswirken, und kann daher von der Europäischen Kommission nicht angegriffen werden17.

2.2 Politische und ökonomische Ansprüche 2.2.1 Politische Zielsetzungen im Konzernsteuerrecht Weniger deutlich als die rechtlichen Rahmenbedingungen erscheinen die politischen Zielsetzungen, die mit einem Konzernsteuerrecht verfolgt werden können. Viele Länder des europäischen Auslandes18 – aus deutscher Sicht namentlich Österreich19 – sehen es inzwischen als Teil nationaler Standortpolitik an, günstige Regeln für Konzerne und insbesondere für die Ansiedlung von Konzernleitungen auf ihrem Territorium zu schaffen. Dazu gehören z. B. großzügige Regeln für die Freistellung von Auslandseinkünften einerseits bei erweiterter grenzüberschreitender Verlustverrechnung andererseits. In Deutschland pflegt die Politik von jeher ein zwiespältiges Verhältnis zum Konzernsteuerrecht. Auf der einen Seite wird die Notwendigkeit erkannt, den Steuerstandort Deutschland sowohl für Konzernleitungen als auch für Entwicklungs-, Produktions- und Vertriebsstätten mittlerer und großer Unternehmen attraktiv zu gestalten. Auf der anderen Seite herrscht der Generalverdacht, dass ein Konzernverhältnis wegen der Möglichkeit, Gewinne in unterschiedlichen Konzerngesellschaften zu allozieren und Transaktionen alternativ auf gesellschaftsvertraglicher oder schuldvertraglicher Grundlage mit jeweils unterschiedlichen Rechtsfolgen durchzuführen, a priori auf missbräuchliche Gestaltungen angelegt sei. Hinzu kommt das unausrottbare Vorurteil, dass günstige Regelungen im Rahmen des Körperschaftsteuerrechts vorzugsweise Großunternehmen zugute kommen, während der in der deutschen Politik hoch geschätzte Mittelstand nahezu ausschließlich als Personengesellschaft organisiert sei. 2.2.2 Entscheidungsneutralität und Ressourcenallokation Versucht man, eine stimmige steuerpolitische Linie für die Ausgestaltung des Konzernsteuerrechts zu formulieren, so sollte man einen Schritt zurücktreten und sich die wirtschaftspolitischen Zielsetzungen des Unternehmenssteuerrechts in seiner Gesamtheit vor Augen führen. Wenn die Förderung des wirtschaftlichen Wachstums das Ziel einer Steuerpolitik bildet, dann muss das Steuerrecht darauf angelegt sein, 16

17

18

19

EuGH v. 22.6.2006, verb. Rs.C-182/03 (Kommission ./. Belgien) und Rs.C-217/03 (Forum 187 ASBL ./. Kommisison), IStR 2006, S. 568 ff. Schön, State Aid in the Area of Taxation, in: Hancher/Ottervanger/Slot, E.C. State Aids, 3. Aufl., 2006, S. 241 ff., Rz. 10-29 f. Rechtsvergleichender Überblick bei Dörr in Schön (Hrsg.), Steuerliche Maßgeblichkeit in Deutschland und Europa, 2005, S. 727 ff. Kessler/Daller, IStR 2006, S. 289 ff.

54

Wolfgang Schön

die optimale Nutzung von Ressourcen zu fördern. Für den Europäischen Binnenmarkt ist diese Zielsetzung in Art. 2, 98 EG deutlich formuliert20. Steuerliche Normen tragen zur effizienten Ressourcenallokation dann bei, wenn sie die wirtschaftlichen Entscheidungen der Akteure so wenig wie möglich beeinflussen, sich gleichsam „neutral“ zu ihnen verhalten21. Bezogen auf Konzernverbindungen heißt dies in concreto, dass Entscheidungen über den rechtlichen Aufbau eines Unternehmens, über die organisatorische Zuordnung von Leitungs- und operativen Funktionen, über die räumliche Anordnung von Produktionsfaktoren und die Lenkung von Finanzströmen im Grundsatz von steuerlichen Konsequenzen unabhängig sein sollten22. Diese Forderung nach steuerlicher Neutralität wirkt für zwei Fallgruppen: Sie verlangt zum einen, dass wirtschaftlich gleichwertige Gestaltungen keine wesentliche steuerliche Differenzierung erfahren sollen; sie verlangt zum anderen, dass der Übergang von einer Rechtsform, Organisationsstruktur oder räumlichen Allokation zu einer anderen – unternehmerisch erwünschten – Gestaltung nicht durch steuerliche Hindernisse, etwa eine steuerwirksame Aufdeckung stiller Reserven, verhindert werden soll. 2.2.3 Konzernneutralität als Zielsetzung des Unternehmenssteuerrechts Aus der Sicht des Gesellschaftsrechts wirft dies die Frage auf, ob unterschiedliche gesellschaftsrechtliche Strukturen – etwa die Aufspaltung eines Gesamtunternehmens in mehrere verselbständigte Tochtergesellschaften – im Grundsatz unterschiedliche steuerliche Belastungseffekte rechtfertigen oder ob eine strenge „Konzernneutralität“ in dem Sinne angestrebt werden soll, dass die zivilrechtlichen Differenzierungen bei der steuerlichen Betrachtung weitgehend in den Hintergrund treten. Diese Fragestellung kulminiert in der Problematik, ob eine Einheitsbesteuerung im Konzern nur bei einer weitgehenden rechtlichen Integration abhängiger Gesellschaften auf der Grundlage von Beherrschungs- und Ergebnisabführungsverträgen angeordnet werden oder ob auch im faktischen Konzern eine Einheitsbesteuerung angestrebt werden soll. Dies hängt davon ab, ob die wesentlichen Merkmale der gesellschaftsrechtlichen Verselbständigung von Unternehmensteilen – vor allem die Einsetzung von eigenständigen Unternehmensorganen und die Trennung der Vermögens- und Haftungssphären – zugleich für die Besteuerung von Unternehmensverbindungen als wesentlich angesehen werden können. Für eine Besteuerungsordnung, die in typi20

21

22

Schön, DStJG 23 (2000), S. 191 ff.; zur wirtschaftspolitischen Konzeption der Unternehmensbesteuerung im Binnenmarkt siehe die Mitteilung der Kommission an den Rat, das Europäische Parlament und den Wirtschafts- und Sozialausschuss v. 23.10.2001, Ein Binnenmarkt ohne steuerliche Hindernisse: Strategie zur Schaffung einer konsolidierten KörperschaftsteuerBemessungsgrundlage für die grenzüberschreitende Unternehmenstätigkeit in der EU, KOM (2001) 582 endg., Abschn. 2. Zum wirtschaftswissenschaftlichen Konzept der „Neutralität“ und seiner Bedeutung für das Unternehmenssteuerrecht siehe Schreiber, Besteuerung der Unternehmen, 2004, S. 512 ff. Herzig in: Herzig (Hrsg.), Organschaft, 2003, S. 3 ff., 24 ff.; kritisch zur Annahme einer „Neutralität“, welche Risiko- und Haftungsstrukturen unberücksichtigt lässt, Schneider, Steuerlast und Steuerwirkung, 2002, S. 2, 15 ff.

B. Perspektiven der Konzernbesteuerung

55

sierender Weise die Realität des Wirtschaftslebens abbilden muss, hängt die Antwort davon ab, ob die gesellschaftsrechtlichen Grenzziehungen die Lebenswirklichkeit der Unternehmen prägen. Mein Eindruck ist, dass für eine solche Differenzierung typischerweise dann kein Raum ist, wenn sich die verselbständigten Unternehmensteile unter gemeinsamer Kontrolle befinden23. Auch im faktischen Konzern ist die Ausrichtung der Tochtergesellschaften auf das Konzerninteresse eine wirtschaftliche Selbstverständlichkeit, die nur in Ausnahmefällen bei schwerwiegender Verletzung von Tochterinteressen aufgegeben wird. Diese seltenen Konflikte zwischen Tochter- und Mutterinteressen bildet den aus der Sicht des Gesellschaftsrechts den wichtigen, aber für die tägliche Unternehmenspraxis doch eher den pathologischen Fall und dürfen daher die steuerliche Würdigung nicht prägen. Gleiches gilt für die Frage der Verlusttragung. Die Voraussetzung persönlicher Haftung für die Schulden des Tochterunternehmens als Maßstab für das Eingreifen des Organschaftsrechts ist in der deutschen Rechtstradition zwar stets betont worden; man muss allerdings berücksichtigen, dass Gewinne und Verluste eines Tochterunternehmens auch ohne förmliche Ergebnisabführung auf die wirtschaftliche Situation des Mutterunternehmens durchschlagen. Es ist lediglich eine Frage bilanzrechtlicher Konventionen, ob die jährlichen Tochterergebnisse der Muttergesellschaft unmittelbar zugerechnet werden oder erst zeitlich verzögert im Rahmen der Bewertung des Tochteranteils oder bei Veräußerungsgeschäften Bedeutung erlangen. Vor dem Hintergrund, dass die Kapitalmärkte heute mit Selbstverständlichkeit die Bewertung börsennotierter Aktiengesellschaften von dem konsolidierten Konzernergebnis abhängig machen, sollte die juristische Separierung auch im Rahmen der Unternehmensbesteuerung keine taugliche Grundlage abgeben, um die steuerliche Belastung eines Konzerns im Vergleich zu einem Einheitsunternehmen zu modifizieren. Wir müssen daher die Analyse der gegenwärtigen Situation und die Bewertung von Reformvorschlägen davon abhängig machen, in welchem Umfang das Ziel einer „Konzernneutralität“ in der Unternehmensbesteuerung verwirklicht werden kann.

3. Sollbruchstellen im Konzernsteuerrecht 3.1 Die Mehrfachbesteuerung von Gewinnen 3.1.1 Klassisches Körperschaftsteuersystem und Integrationsverfahren Ausgangspunkt einer Analyse des geltenden Konzernsteuerrechts ist die Frage, ob und wie in Konzernverhältnissen die Mehrfachbesteuerung des Gewinns einer Tochtergesellschaft bei dessen Ausschüttung an die Muttergesellschaft verhindert werden soll24. Das „klassische“ Körperschaftsteuersystem ist darauf ausgerichtet, 23

24

Krebühl in: Herzig (Hrsg.), Organschaft, 2003, S. 595 ff., 596 f.; Witt, Die Konzernbesteuerung, 2006, S. 19 f., S. 139 ff. Ausführliche Analyse der grenzüberschreitenden Besteuerung von Dividenden bei: Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681, Rz. 326 ff.

56

Wolfgang Schön

jeder beteiligten Gesellschaft – sowohl der Mutter- als auch der Tochtergesellschaft – ihre eigene subjektive Steuerpflicht und ihre eigene objektive Leistungsfähigkeit vorzuhalten, so dass sowohl die Tochtergesellschaft auf den erwirtschafteten Gewinn als auch die Muttergesellschaft auf die bezogene Dividende jeweils ihre Körperschaftsteuer entrichtet25. Ein solches klassisches Körperschaftsteuersystem würde durchaus den Ansprüchen des EG-Vertrages und des Grundgesetzes genügen. Es ist vom Grundgesetz über Jahrzehnte stets akzeptiert worden und kann zugleich – das erscheint heute fast noch wichtiger – in grenzüberschreitenden Fällen diskriminierungsfrei zum Einsatz gebracht werden. Sowohl im Fall deutscher Tochtergesellschaften mit ausländischer Muttergesellschaft als auch im umgekehrten Fall deutscher Muttergesellschaften mit ausländischen Töchtern würde schlicht das bei der jeweils in Deutschland ansässigen Kapitalgesellschaft vereinnahmte Einkommen – der operative Gewinn der Tochter oder die bezogene Dividende der Mutter – steuerlich erfasst. Die im klassischen Körperschaftsteuersystem angelegte Mehrfachbelastung läuft jedoch der politischen Zielsetzung eines „konzernneutralen“ Steuerrechts konträr entgegen und behindert insbesondere die Gründung ausländischer Tochtergesellschaften. Vor diesem Hintergrund ordnet die Mutter-Tochter-RL seit dem Jahre 1990 an, dass bei grenzüberschreitenden wesentlichen Beteiligungen der Sitzstaat der Muttergesellschaft die bezogenen Dividenden entweder zur Gänze freistellen oder jedenfalls die von der Tochtergesellschaft gezahlte Körperschaftsteuer anrechnen muss26. Damit ist das „klassische System“ de lege lata für den grenzüberschreitenden Fall erledigt. Es erscheint unter dieser Prämisse auch nicht sinnvoll, ein klassisches Körperschaftsteuerrecht im nationalen Rahmen zu praktizieren. Angestrebt werden muss daher eine Einmalbelastung des bei der Tochtergesellschaft erwirtschafteten Gewinns innerhalb des Konzerns. Dies führt in die Notwendigkeit, die bei der Tochtergesellschaft und der Muttergesellschaft jeweils anfallenden Körperschaftsteuern in ein wechselseitiges Abhängigkeitsverhältnis zu bringen. Dies wirft im Inlandsfall keine Probleme auf, weil diese Steuern von demselben Fiskus erhoben werden. Es führt aber zu Friktionen im grenzüberschreitenden Fall, weil dieser nach einer Koordinierung zwischen inländischem Fiskalanspruch und ausländischem Steuerzugriff verlangt. Dabei würde jeder Versuch, für das jeweilige Entlastungssystem zwischen der inländischen Unternehmensgruppe und dem grenzüberschreitenden Konzern zu differenzieren, an der strengen Rechtsprechung des Europäischen Gerichtshofs scheitern, welche eine Gleichbehandlung grenzüberschreitender Konzerne mit inländischen Beteiligungsverhältnissen verlangt und dabei so weit geht, dass der inländische Fiskus eine Anrechnung oder vergleichbare Entlastung auch für solche Steuern vornehmen muss, die er nicht selbst vereinnahmt hat27. 25

26 27

Überblick zur Entwicklung der Körperschaftsteuersysteme bei Hey in: Herrmann/Heuer/Raupach, Einkommensteuergesetz und Körperschaftsteuergesetz, Einf. KSt (Stand 1999), Anm. 9 ff. Art. 4 Mutter-Tochter-RL. EuGH v. 7.9.2004, Rs.C-319/02 (Manninen), EuGHE 2004, S. I-7477 ff.; EuGH v. 6.3.2007, Rs.C-292/04 (Meilicke).

B. Perspektiven der Konzernbesteuerung

57

An dieser Vorgabe sind in jüngerer Zeit die vor einigen Jahrzehnten in Europa etablieren Anrechnungsverfahren zugrunde gegangen, da die nationalen Fisci verständlicherweise nicht bereit sind, eine volle Anrechnung ausländischer Körperschaftsteuern zu gewähren; an dieser Vorgabe scheitert auch das elegante „Dividendenabzugsverfahren“, welches bei der Tochtergesellschaft den vollen Betriebsausgabenabzug gezahlter Dividenden vorsieht, weil es im grenzüberschreitenden Fall dann nur noch zu einer Besteuerung bei der ausländischem Muttergesellschaft kommt und der operative Inlandsgewinn letztlich unversteuert bleibt. Akzeptiert werden kann allerdings der in Deutschland seit 2001 gewählte Weg einer gleichmäßigen Freistellung inländischer und ausländischer Dividenden auf der Ebene der Muttergesellschaft28. Diese Freistellung führt letztlich dazu, dass die Gewinne ausländischer Tochtergesellschaften in Deutschland solange keinem steuerlichen Zugriff unterliegen, wie sie nicht an die Anteilseigner der Muttergesellschaft ausgekehrt und dort im Rahmen der regulären Einkommensteuer (nach Maßgabe des „Halbeinkünfte-Verfahrens“) erfasst werden29. 3.1.2 Der Abzug von Beteiligungsaufwand Die Freistellung von in- und ausländischen intercompany-Dividenden hat allerdings auch einen Haken, nämlich das Problem des Abzugs von Beteiligungsaufwand. Wie ist mit den Kosten umzugehen, welche das Mutterunternehmen für den Erwerb und die Verwaltung der inländischen oder ausländischen Tochterbeteiligung aufwendet – vor allem Finanzierungszinsen und Verwaltungskosten. Die deutsche Steuergesetzgebung hat zunächst auf die Steuerfreiheit der Dividendeneinnahmen hingewiesen und korrespondierend den Abzug der durch die Beteiligung veranlassten Kosten ausgeschlossen (§ 3c Abs.1 EStG). Dabei wurde allerdings verkannt, dass die Vorbelastung der ausgeschütteten Gewinne mit Körperschaftsteuer auf der Ebene der Tochtergesellschaft sich letztlich als eine materielle Besteuerung der Dividende zu Lasten des Mutterunternehmens auswirkt, so dass systemgerecht auch der bei der Muttergesellschaft anfallende Aufwand für steuerlich abzugsfähig erklärt werden muss30. Dies ist inzwischen (§ 8b Abs.5 KStG) mit einer Einschränkung (5 %-Fiktion) vom Gesetzgeber akzeptiert worden. Das eigentliche Problem besteht darin, dass nach der Rechtsprechung des Europäischen Gerichtshofs diese Abzugsfähigkeit dann diskriminierungsfrei auch für den Aufwand für ausländische Beteiligungen gewährt werden muss31. Wenn eine inländische Muttergesellschaft die Gründung, den Erwerb oder den Ausbau einer ausländischen Tochtergesellschaft durch Kredite finanziert, werden daher die resultierenden Gewinne der Tochtergesellschaft nur im Ausland versteuert, die resultierenden Kosten jedoch nur in Deutschland abgezogen32. Dies hat zum dem polemisch überspitzten, aber in der 28 29

30 31

32

§ 8b Abs. 1 KStG. Dazu monographisch Otto, Die Besteuerung von gewinnausschüttenden Körperschaften und Anteilseignern nach dem Halbeinkünfteverfahren, 2006. Schön, FR 2001, S. 381 ff. EuGH v. 18.9.2003, Rs.C-168/01 (Bosal), IStR 2003, S. 666 ff.; EuGH v. 23.2.2006, Rs. C-471/ 04 (Keller Holding), IStR 2006, S. 235 ff. Spengel/Braunagel, StuW 2006, S. 34 ff., 37.

58

Wolfgang Schön

Sache bedenkenswerten Vorwurf eines „Super-Steuersparmodells für Konzerne“ 33 geführt, mit dem der deutsche Steuerzahler die Verlagerung von Arbeitsplätzen ins Ausland mitfinanziert. Schaut man näher hin, so liegt das eigentliche Problem in der Sphärentrennung zwischen der Mutter- und der Tochtergesellschaft: Die Kosten sind wirtschaftlich veranlasst durch die Geschäftstätigkeit der Tochter und müssten dort zum Abzug gebracht werden, sie können aber juristisch der Tochtergesellschaft nicht zugeordnet werden, so dass der Sitzstaat der Tochtergesellschaft diese nicht als Betriebsausgaben akzeptiert34.

3.2 Die Verrechnung von Verlusten Aus der Sicht des deutschen Steuerrechts liegt die wesentliche Wirkung der körperschaftsteuerlichen Organschaft nach der Beseitigung der körperschaftsteuerlichen Doppelbelastung des Gewinns der Tochtergesellschaft in der wechselseitigen Verrechnung von Verlusten zwischen den Gesellschaften eines Organkreises 35. In dieser Funktion ist das Rechtsinstitut der Organschaft in den letzten Jahren vor allem aus der Sicht der grenzüberschreitenden Verlustverrechnung in die Schusslinie geraten. Das deutsche Organschaftsrecht verlangt im Grundsatz die inländische Ansässigkeit sowohl des herrschenden als auch des abhängigen Unternehmens und beschränkt die Ergebniszurechnung damit auf inländische Sachverhalte36. Hinzu kommt das Erfordernis eines Ergebnisabführungsvertrages, welches deshalb mittelbar beschränkend wirkt, weil das deutsche Konzernvertragsrecht auf ausländische Tochtergesellschaften nach kollisionsrechtlichen Grundsätzen keine Anwendung findet und das ausländische Gesellschaftsrecht in der Regel keine vollgültigen Unternehmensverträge i. S. der §§ 291 ff. AktG kennt37. Die gegen die aktuelle Rechtslage vorgetragenen Gründe sind steuerpolitischer und steuerjuristischer Natur. Steuerpolitisch wird darauf hingewiesen, dass die Möglichkeit grenzüberschreitender Verlustverrechnung für die Ansiedlung von Konzernleitungen von herausgehobener Bedeutung sei38 und daher in anderen europäischen Jurisdiktionen – etwa in Österreich – zunehmend derartige Verrechnungsoptionen bestehen 39. Steuerjuristisch wird hervorgehoben, dass der Ausschluss ausländischer Tochterunternehmen von der Verlustverrechnung eine Behinderung der grenzüberschreitenden Niederlassung i. S. von Art.43, 48 EG darstelle40. Die mit 33 34 35 36 37 38

39 40

Jarass/Obermair, Geheimnisse der Unternehmenssteuern, 2004, S. 93 ff. Schön, FR 2001, S. 381 ff; 389 f. Orth in: Herzig (Hrsg.), Organschaft, 2003, S. 167 ff. Thömmes in: Herzig (Hrsg.), Organschaft, 2003, S. 525 ff. Herzig/Wagner, Der Konzern 2006, S. 176 ff., 184 ff. Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681, Rz. 347 ff. Kessler/Daller, IStR 2006, S. 289 ff. Fischer, DStR 2006, S. 2287 f.; Spengel/Braunagel, StuW 2006, S. 34 ff, 38 ff.; Witt, Die Konzernbesteuerung, 2006, S. 124 ff.; ausführlich zur Auswirkung der Grundfreiheiten auf die grenzüberschreitende Verlustverrechnung siehe die Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Steuerliche Behandlung von Verlusten bei grenzübergreifenden Sachverhalten, 19.12.2006, KOM (2006) 824 endg.

B. Perspektiven der Konzernbesteuerung

59

Spannung erwartete Entscheidung des Europäischen Gerichtshofs in der Rechtssache „Marks & Spencer“41 hat zu dieser Frage eher Verwirrendes als Klärendes beigetragen. Einerseits betont der Gerichtshof, dass die territoriale Abgrenzung der Steuerhoheit zwischen verschiedenen Mitgliedstaaten im Grundsatz Differenzierungen nach dem Entstehungsort von Verlusten erlaube. Andererseits meint der Gerichtshof, dass Verluste, die im Staat der Tochtergesellschaft nicht abgezogen werden können, jedenfalls im Staat der Muttergesellschaft in Ansatz gebracht werden müssten. Mit der Aussage, ein einmal entstandener Verlust müsse jedenfalls in einem der beteiligten Staaten verrechnet werden können, verstrickt sich der Gerichtshof allerdings in einem grundlegenden Problem des europäischen Steuerrechts, nämlich die fehlende Harmonisierung der Gewinnermittlung in den Mitgliedstaaten. Nach welchem Recht bestimmt sich überhaupt, ob ein verrechenbarer Verlust entstanden ist? Und weshalb soll der Staat der Muttergesellschaft verstärkt zur Verlustverrechnung verpflichtet werden, wenn und weil der Staat der Tochtergesellschaft gegenüber seinen Steuerpflichtigen Abzugs- und Ausgleichsmöglichkeiten einschränkt? Schließlich sind auch die Techniken der Verlustverrechnung in den Mitgliedstaaten der Gemeinschaft sehr unterschiedlich, sie reichen von der Vollkonsolidierung aller Geschäftsvorfälle über die Zurechnung von isoliert ermittelten Jahresergebnissen bis hin zu der bloßen Zuweisung von Verlustgutschriften oder einem Betriebsausgabenabzug für Verlustausgleichszahlungen des Mutterunternehmens. Indem der Gerichtshof die Rechtsfolge im Staat der Obergesellschaft an die Rechtslage im Staat der Untergesellschaft knüpft, führt er die Beteiligten in Aporien, die ohne eine weitergehende Harmonisierung der steuersystematischen und bilanzrechtlichen Grundlagen der Unternehmensbesteuerung nicht bewältigt werden können.

3.3 Die Kontrolle von Schuldverhältnissen Das geltende deutsche Unternehmenssteuerrecht legt der Ermittlung und Besteuerung von Gewinnen die Annahme zugrunde, dass für jede Konzerngesellschaft gesondert die betrieblichen Einnahmen und Ausgaben erfasst und dabei auch schuldrechtliche Vertragsbeziehungen zwischen den Konzerngesellschaften als Drittgeschäfte anerkannt werden müssen. Der gesamte konzerninterne Leistungsaustausch wird daher mit steuerlicher Wirkung erfasst, es kommt zu Kosten bei dem empfangenden und zu Gewinnen bei dem abgebenden Unternehmen. Zugleich wird damit die Möglichkeit eröffnet, die persönliche Zuordnung von Ergebnissen zu beeinflussen, wenn zum Beispiel die Hingabe von Finanzmitteln als förmliche Kapitalerhöhung, als verlorener Zuschuss oder als verzinsliches oder unverzinsliches Darlehen gestaltet werden kann – ganz abgesehen von der sich stets erweiternden Vielfalt hybrider Finanzierungsinstrumente. Diese Trennung der Gewinnermittlungseinheiten – neudeutsch: separate accounting – wird sogar im Recht der körperschaftsteuerlichen Organschaft praktiziert, bei der zwar das gesondert ermittelte Tochterergebnis der Muttergesellschaft zur Versteuerung zugewiesen, nicht aber die Gewinnermittlung insgesamt konsolidiert wird42. 41 42

EuGH v. 13.12.2005, Rs.C-446/03 (Marks & Spencer), IStR 2006, S. 19 ff. Herlinghaus in: Herzig (Hrsg.), Organschaft, 2003, S. 119 ff.

60

Wolfgang Schön

Die Anerkennung konzerninterner Schuldverträge hat zur natürlichen Folge, dass die Angemessenheit des Leistungsaustauschs überprüft werden muss. Im innerstaatlichen Recht gründet sich dies auf die Rechtsfigur der verdeckten Gewinnausschüttung43; im wichtigeren grenzüberschreitenden Fall treten die Regeln des Doppelbesteuerungsrechts über die Gewinnkorrektur nach dem Arm’s-Length-Maßstab hinzu44. Es ist heute weitgehend akzeptiert, dass die Leistungsfähigkeit dieser Prüfungsmaßstäbe immer weiter absinkt45. Hintergrund ist der Umstand, dass konzerninterne Wertbewegungen immer weniger marktgängige Güter – Rohstoffe, Halb- und Fertigprodukte oder typisierte Dienstleistungen – betreffen, bei denen Vergleichspreise ermittelt oder schlichte Kostenzu- oder -abschläge vorgenommen werden können. Im Vordergrund stehen vielmehr schwer bewertbare immaterielle Güter (weit über den rechtlich konturierten Bereich von Patenten oder anderen Schutzrechten hinaus) und individuelle Dienstleistungen 46. Hinzu kommt der Umstand, dass die Zuordnung von Gewinnen an einzelne Konzerngesellschaften nicht nur eine Bewertung von Leistungen erfordert, sondern zugleich die Übernahme von Risiken innerhalb des Konzernverbunds widerspiegelt47. Forschungs- und Entwicklungsgesellschaften oder Produktionsbetriebsstätten können sowohl auf eigene Rechnung als auch im Auftrag der Muttergesellschaft als auch in einem gesellschaftsähnlichen Kostenverbund arbeiten48. Schließlich gibt es konzernspezifische Vorteile – Synergieeffekte – die im Grundsatz keiner der beteiligten Gesellschaften a priori zustehen, sondern überhaupt erst im Zusammenwirken der Gesellschaften zur Entstehung gelangen49. 43 44

45

46

47

48

49

Darstellung bei Knobbe-Keuk., Bilanz- und Unternehmenssteuerrecht, 9. Aufl., 1993, § 19. Art.9 OECD-Musterabkommen; siehe dazu die Kommentierung bei Eigelshoven in: Vogel/ Lehner (Hrsg.), DBA-Kommentar, 4. Aufl., 2003. Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681, Rz. 362 ff.; Kahle, Wpg. 2007, S. 210 ff., 211 ff.; Oestreicher in: Endres u.a. (Hrsg.), Die internationale Unternehmensbesteuerung im Wandel – Symposium zum 65. Geburtstag von Otto H. Jacobs, 2005, S. 73 ff., 74 ff.; Verteidigung bei Owens, International Transfer Pricing Journal 2005, S. 99 ff. und Wassermeyer in: Endres u.a. (Hrsg.), Die internationale Unternehmensbesteuerung im Wandel – Symposium zum 65. Geburtstag von Otto H. Jacobs, 2005, S. 63 ff. Niznik, Tax Notes International v. 18.12.2006, S. 975 ff., 991 f.; Schreiber, StuW 2004, S. 212 ff., 218 f.; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 94; Spengel, Besteuerung von Einkommen – Aufgaben, Wirkungen und europäische Herausforderungen, Gutachten für den 66. Deutschen Juristentag, 2006, S. G 38 f. Oestreicher in: Endres u.a. (Hrsg.), Die internationale Unternehmensbesteuerung im Wandel – Symposium zum 65. Geburtstag von Otto H. Jacobs, 2005, S. 73 ff., 77 ff. Vann, Problems in international division of the business income tax base, Working Paper, 2006, Abschn.III C. Avi-Yonah/Clausing, A Proposal to Adopt Formulary Apportionment for Corporate Income Taxation – The Hamilton Project, Working Paper, 2006, S. 6, 14; Hellerstein, International Transfer Pricing Journal 2005, S. 103 ff., 107; McLure, European Taxation 2002, S. 586 ff., 586 f.; McLure/Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 248; Oestreicher in: Endres u.a. (Hrsg.), Die internationale Unternehmensbesteuerung im Wandel – Symposium zum 65. Geburtstag von Otto H. Jacobs, 2005, S. 73 ff., 76 ff.; ausführlich Schäfer/Spengel, The Impact of ICT on Profit Allocation within Multinational Groups: Arm’s Length Pricing of Formula Apportionment? ZEW Discussion Paper No. 0353, 2003, S. 2 ff., 8 ff.

B. Perspektiven der Konzernbesteuerung

61

Beispielhaft ist die jüngste cause célèbre des internationalen Steuerrechts, der Fall Glaxo Smithkline50. Die US-amerikanische Tochtergesellschaft eines britischen Pharmaherstellers vertrieb in den Jahren 1989 – 2000 erfolgreich ein Medikament auf dem amerikanischen Markt. Mit der US-amerikanischen Finanzverwaltung ergab sich ein Streit darüber, ob dieser Gewinn wesentlich auf die Forschungstätigkeit der Muttergesellschaft oder auf die Markterschließung des Tochterunternehmens zurückgeführt werden musste – mit entsprechenden Folgen für die Aufteilung des Gesamtgewinns auf die beteiligten Jurisdiktionen. Die US-Steuerverwaltung ordnete den wesentlichen Gewinnanteil der amerikanischen Vertriebsgesellschaft zu und gelangte zu einem steuerlichen Mehrbetrag von 4.6 Mrd. $. Vor wenigen Wochen hat sich GlaxoSmithkline mit den US-Behörden auf eine abschließende Zahlung in Höhe von 3.4 Mrd. $ geeinigt. Die britischen Finanzbehörden sind ihrerseits allerdings nicht bereit, eine entsprechende Gegenbuchung vorzunehmen, so dass der Konzern auf einer milliardenschweren Doppelbesteuerung „sitzen bleibt“.

3.4 Die zentrale Rolle von Finanzierungen Ganz unabhängig von der Prüfung der Angemessenheit von Leistung und Gegenleistung steht heute außer Frage, dass die Zuordnung von Gewinnen innerhalb von Konzernverbindungen weitgehend mit Hilfe von Finanzierungsmaßnahmen gestaltet werden kann51. Ausgangspunkt ist der Umstand, dass Entgelte für Fremdkapital – in erster Linie Zinsen – im Sitzstaat des Kapitalgebers besteuert werden, während Entgelte für Eigenkapital – die Gewinne der Tochtergesellschaften und daraus resultierende Dividenden – im Sitzstaat des Kapitalnehmers veranlagt werden. Die freie Wahl zwischen Fremd- und Eigenkapital führt aus der Sicht der Konzerne zu einer freien Wahl der steuerberechtigten Jurisdiktion. Dieser Umstand führt naturgemäß zu Reaktionen des Fiskus, die sich allerdings in der Vergangenheit immer weniger als zielführend erwiesen haben. Die deutschen Vorschriften über die steuerliche Nichtabzugsfähigkeit von Entgelten aus Gesellschafter-Fremdfinanzierung (§ 8a KStG) wurden vom Europäischen Gerichtshof 52 für diskriminierend erklärt – mit der Folge, dass seit 2002 auch für inländische Konzernverbindungen dieselben Regeln Anwendung finden und bisher unverdächtige Finanzierungsmaßnahmen in eine steuerliche Falle laufen lassen53. Trotz dieser erheblichen gesetzgeberischen und verwaltungstechnischen Anstrengungen gelingt es in der Praxis nicht hinreichend, grenzüberschreitende Konzerndarlehen sauber von echten Drittdarlehen zu unterscheiden, etwa bei Back to Back-Finanzierungen über neutrale Kreditinstitute54. Ebenfalls stark eingeschränkt sind Versuche, die Erträge ausländischer Finanzierungsgesellschaften unmittelbar der deutschen Konzernmutter zuzurechnen; die 50

51 52 53 54

Zum Sachverhalt und seinen Auswirkungen: Fris/Gonnet, Transfer Pricing International Tax Planning 2006, Heft 11, S. 9 ff.; Wright, Transfer Pricing International Tax Planning 2006, Heft 11, S. 3 ff. Kahle, Wpg. 2006, S. 1401 ff., 1402 f. EuGH v. 12.12.2002, Rs.C-324/00, IStR 2003, S. 55 ff. Fischer, DStR 2006, S. 2286 f.; Spengel/Braunagel, StuW 2006, S. 34 ff., 36 f. Herzig, DB 2007, S. 1 ff., 1.

62

Wolfgang Schön

wesentliche Grundlage hierfür – die §§ 7 ff. des deutschen Außensteuergesetzes – ist in ihren Grundzügen mit dem Gemeinschaftsrecht unvereinbar55. Das Bundesfinanzministerium hat vor wenigen Tagen in einem Erlass 56 die Anwendbarkeit dieser Vorschriften stark eingeschränkt. Den einzigen systematisch überzeugenden Ausweg, nämlich die Einführung einer deutschen Quellensteuer auf in das Ausland fließende Zinsen, hat sich der deutsche Gesetzgeber durch eine Vielzahl von Doppelbesteuerungsabkommen sowie im Juli 2003 durch seine Zustimmung zu der europäischen Richtlinie über die Besteuerung von Zins- und Lizenzgebühren im Konzern selbst verschlossen. Wenn heute im Rahmen der Unternehmenssteuerreform 2008 an einem Gesetzesentwurf gearbeitet wird, der die Abzugsfähigkeit von Zinsen in systemfremder Weise dramatisch einschränken wird57, so ist dies in erster Linie dem Umstand geschuldet, dass der systemkonforme Ansatz einer Quellensteuer nicht verwirklicht werden kann, solange sich nicht alle 27 Mitgliedstaaten der Europäischen Union einstimmig auf eine Aufhebung dieser Richtlinie verständigen (an deren Verabschiedung ihrerseits über Jahrzehnte gerungen worden ist). Eines muss klar sein: Ein Konzernsteuerrecht, welches an die Wahl zwischen Fremd- und Eigenkapitalfinanzierung zugleich die internationale Allokation des Besteuerungsrechts knüpft, ist langfristig nicht zu halten. Natürlich kann man dieses Problem auch bewältigen, indem das vor Jahrzehnten im US-Schrifttum entwickelte Modell einer Comprehensive Business Income Tax58 wählt und Zinsen weder beim Empfänger versteuert noch beim Schuldner für abzugsfähig erklärt. Doch setzt dies einen weitgehenden Konsens der Industrienationen voraus; die unilaterale Einführung durch Deutschland käme einem steuerpolitischen Selbstmord gleich.

3.5 Die Vermeidung von Gewinnrealisierungen Ein letzter Punkt betrifft Umstrukturierungen im Konzern59. Wenn und soweit diese betriebswirtschaftlich sinnvoll sind, sprechen gute Gründe dagegen, sie zum Anlass für eine prohibitive Gewinnrealisierung zu nehmen. Dies betrifft sowohl die Übertragung von Wirtschaftsgütern zwischen verschiedenen Steuerpflichtigen – etwa bei Verschmelzungen, Spaltungen und Einbringungen – als auch die Verlagerung von Wirtschaftsgütern zwischen verschiedenen Jurisdiktionen – etwa bei der Verbringung von Wirtschaftsgütern über die Grenze. Die Rechtsprechung des Europäischen Gerichtshofs60 gesteht zwar den Mitgliedstaaten zu, unter ihrer Fiskalhoheit entstan55 56 57

58

59

60

EuGH v. 12.9.2006, Rs.C-196/04 (Cadbury Schweppes), IStR 2006, S. 670 ff. BMF v. 8.1.2007 (IV B 4 – S 1351 – 1/07), DB 2007, S. 137. § 4h des Referentenentwurfs eines Unternehmenssteuerreformgesetzes v. 5.2.2007 (www.bmf. bund.de). Homburg in: Endres u.a. (Hrsg.), Die internationale Unternehmensbesteuerung im Wandel, Symposium für Otto H. Jacobs zum 65.Geburtstag, 2005, S. 14 ff., 23 ff. Fischer, DStR 2006, S. 2285 f.; Spengel/Braunagel, StuW 2006, S. 34 ff, 41 ff.; ausführlich Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681, Rz. 336 ff. Zuletzt EuGH v. 7.9.2006, Rs.C-470/04 (N), IStR 2006, S. 702 ff.

B. Perspektiven der Konzernbesteuerung

63

dene stille Reserven im Fall von Verlagerungen zu besteuern, verlangt aber eine diskriminierungsfreie Ausgestaltung, die insbesondere großzügige Stundungen und jahrelange Nacherhebungen einschließt61. Sowohl für die beteiligten Finanzverwaltungen als auch für die betroffenen Steuerpflichtigen ist dies ein Alptraum; es wäre deutlich sinnvoller, wenn eine weitergehende steuerliche Koordinierung von diesen Lasten befreien würde.

4. Zum Verhältnis Betriebsstätten / Tochtergesellschaften Die vorstehenden Ausführungen sollten verdeutlichen, dass die Sphärentrennung zwischen Konzerngesellschaften im internationalen Verkehr zu einer Fülle von Verwerfungen, Hindernissen und Gestaltungsoptionen führen, die weder den steuerpflichtigen Unternehmen noch den beteiligten Finanzverwaltungen dauerhaft zugemutet werden können. Bevor ich zu der Frage komme, ob und in welcher Weise das bisherige Konzernsteuerrecht durch eine stärker vereinheitlichende Betrachtung abgelöst werden sollte, ist allerdings auf eine internationale Entwicklung aufmerksam zu machen, die gleichsam in die Gegenrichtung läuft. Gemeint sind die Überlegungen innerhalb der OECD zur künftigen Gestalt der Betriebsstättenbesteuerung. In der internationalen Besteuerungspraxis hat sich in der Vergangenheit kein Konsens darüber entwickelt, ob und in welcher Weise inländische Betriebsstätten ausländischer Unternehmen und ausländische Betriebsstätten inländischer Unternehmen für die Gewinnabgrenzung zwischen Stammhaus und Betriebsstätte als verselbständigte Einheiten hypostasiert werden sollen62. Die OECD hat seit 2001 in umfangreich diskutierten Diskussionspapieren den Vorschlag unterbreitet, solche Betriebsstätten weitestgehend wie selbständige Tochtergesellschaften zu behandeln und auf diese Weise einen „Weltstandard“ zu formulieren. Der Bericht wurde im Dezember 2006 abgeschlossen63, die Bundesrepublik Deutschland hat bei der jüngsten Revision des Doppelbesteuerungsabkommens mit den Vereinigten Staaten diese Selbständigkeitsfiktion akzeptiert64. Man braucht nicht viel Phantasie, um sich vorzustellen, dass die Anwendung der oben geschilderten Rechtsfolgen der Konzernbesteuerung auf bloß fiktive Tochtergesellschaften zu einer Potenzierung der tatsächlichen und rechtlichen Problemstellungen führen wird65. Wenn man eine Gleichbehandlung von Tochtergesellschaft und Betriebsstätte anstrebt, dann bietet sich die Zusammenfassung wirtschaftlicher Einheiten – unabhängig vom rechtli61

62

63

64 65

Ausführliche Darstellung der Konsequenzen der EuGH-Rechtsprechung bei: Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Wegzugsbesteuerung und die Notwendigkeit einer Koordinierung der Steuerpolitiken der Mitgliedstaaten, 19.12.2006, KOM (2006) 825 endg. Rechtsvergleichende Darstellung bei Baker/Collier, The Attribution of Profits to Permanent Establishments, Cahiers de Droit Fiscal International, Bd.91b, 2006, S. 21 ff. mit Nationalberichten. OECD, Report on the Attribution of Profits to Permanent Establishements, Parts I (General Considerations), II (Banks) and III (Global Trading), 2006. Wolff/Eimermann, IStR 2006, S. 837 ff., 842 f. Ausführliche Diskussion bei Schön in: Lüdicke (Hrsg.), Besteuerung von Unternehmen im Wandel – Internationale Umwandlungen, Funktionsverlagerungen, Betriebsstätten, 2007, 71 ff.

64

Wolfgang Schön

chen Status der Unternehmensglieder – eher an als die Erstreckung separierender Modelle von der Tochtergesellschaft auf die Betriebsstätten66.

5. Das Konzept einer einheitlichen Konzernsteuerbemessungsgrundlage 5.1 Grundzüge und Stand des Konzepts Es ist diese prekäre Gesamtsituation67, welche den Europäischen Institutionen den Anlass gegeben hat, weiträumig über den Abbau von steuerlichen Hindernissen für die Gestaltung von Konzernunternehmen in der Gemeinschaft nachzudenken68. Auf der Grundlage einer umfangreichen Studie ihrer Dienststellen69 hat die Kommission in einer Mitteilung aus dem Jahre 200170 eine Reihe kurz- und langfristiger Koordinierungs- und Harmonisierungsmaßnahmen präsentiert, in deren Zentrum das Projekt einer einheitliche steuerlichen Bemessungsgrundlage für grenzüberschreitende Konzernunternehmen steht. In den Folgejahren haben in Abstimmung mit der Kommission Vertreter der Mitgliedstaaten in Arbeitsgruppen Einzelaspekte dieses Reformvorhabens diskutiert71; die Kommission berichtet laufend über die Fortschritte dieser Arbeiten72, die im Jahre 2008 in einen Richtlinienvorschlag münden sollen.

66 67

68

69

70

71

72

Schreiber, StuW 2004, S. 212 ff, 223 f. Zum Charakter der Rechtsunsicherheit als Investitionshindernis siehe Krister Andersson, A Common Consolidated Tax Base in Europe – A Way of Reducing Tax Uncertainty, 2005. Für eine „weltweite“ Lösung siehe Avi-Yonah/Clausing, A Proposal to Adopt Formulary Apportionment for Corporate Income Taxation – The Hamilton Project, Working Paper, 2006. Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681.; siehe dort auch zu „gezielten Maßnahmen“ im Vorfeld einer weitgehenden Harmonisierung Rz. 423 ff. Mitteilung der Kommission an den Rat, das Europäische Parlament und den Wirtschafts- und Sozialausschuss v. 23.10.2001, Ein Binnenmarkt ohne steuerliche Hindernisse: Strategie zur Schaffung einer konsolidierten Körperschaftsteuer-Bemessungsgrundlage für die grenzüberschreitende Unternehmenstätigkeit in der EU, KOM (2001) 582 endg.; siehe dazu bereits ausführlich Schön, European Taxation 2002, S. 276 ff. Zur Dokumentation dieser Arbeiten siehe http://ec.europa.eu/taxation_customs/taxation/ company_tax/common_tax_base/index_de.htm. Siehe unter anderem oben Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss: Ein Binnenmarkt ohne unternehmenssteuerliche Hindernisse, Ergebnisse, Initiativen, Herausforderungen, 24.11.2003, KOM (2003) 726 endg.; Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Umsetzung des Lissabon-Programms der Gemeinschaft: Bisherige Fortschritte und weitere Schritte zu einer gemeinsamen konsolidierten Körperschaftsteuer-Bemessungsgrundlage (GKKB), 5.4.2006, KOM (2006) 157 endg.

B. Perspektiven der Konzernbesteuerung

65

Das Reformkonzept73 soll im Wesentlichen aus vier Elementen bestehen: – Es werden europaweit einheitlich formulierte Gewinnermittlungsregeln eingeführt, die eine übereinstimmende steuerliche Bilanzierung in allen Konzerngesellschaften ermöglichen. – Diese Steuerbilanzen werden über die Grenzen der beteiligten Gesellschaften und über die Grenzen der beteiligten Staaten hinaus konsolidiert, d. h. es wird für die Gesamtheit aller Konzerngesellschaften und aller Betriebsstätten ein einheitliches Ergebnis ermittelt. – Dieses europaweit ermittelte Gesamtergebnis wird nach Maßgabe bestimmter Berechnungsfaktoren auf die Mitgliedstaaten verteilt. – Die Mitgliedstaaten wenden schließlich ihren jeweiligen Körperschaftsteuersatz auf den ihnen zugeordneten Anteil am Konzernergebnis an und erheben die anfallenden Steuern. Eine Harmonisierung der Steuersätze wird nicht angestrebt; vielmehr wird hier ein gesunder Wettbewerb zwischen den Mitgliedstaaten befürwortet. Eine solche einheitliche Konzernsteuerbemessungsgrundlage ist darauf angelegt, mit einem Schlag sämtliche oben geschilderten Probleme zu lösen74: – Eine doppelte Besteuerung von Tochtergewinnen bei der Muttergesellschaft kann bei einer Aufhebung der Grenzen zwischen den beteiligten Gesellschaften nicht mehr eintreten 75. – Die Frage der Zuordnung des Beteiligungsaufwands scheitert ebenfalls nicht mehr an der Sphärentrennung zwischen Mutterunternehmen und Tochterunternehmen; der Aufwand geht schlicht in das Gesamtergebnis des Konzerns ein. – Verluste können – jedenfalls bei Konsolidierung im Konzern – europaweit zusammengefasst werden76.

73

74

75 76

Grundzüge in: Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681, Rz. 507 ff. Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681, Rz. 518 ff.; Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Umsetzung des LissabonProgramms der Gemeinschaft: Bisherige Fortschritte und weitere Schritte zu einer gemeinsamen konsolidierten Körperschaftsteuer-Bemessungsgrundlage (GKKB), 5.4.2006, KOM (2006) 157 endg., Tz. 1; Spengel, Besteuerung von Einkommen – Aufgaben, Wirkungen und europäische Herausforderungen, Gutachten für den 66. Deutschen Juristentag, 2006, S. G 38 f., G 94 ff.; McLure, CESifo-Forum 2005, Heft 1, S. 32 ff., 36 f.; Niznik, Tax Notes International v. 18.12.2006, S. 975 ff., 991 f.; Westberg, European Taxation 2002, S. 322 ff., 325 f. Kahle, Wpg. 2006, S. 1401 ff., 1408. Schreiber, StuW 2004, S. 212 ff, 219; Spengel/Braunagel, StuW 2006, S. 34 ff, 47 f.; ausführlich Gérard/Weiner, Cross-Border Loss-Offset and Formulary Apportionment: How do they affect multijurisdictional firm investment spending and interjurisdictional tax competition?, CESifo Working Paper No. 1004, 2003.

66

Wolfgang Schön

– Die Problematik der Angemessenheitskontrolle von Schuldverhältnissen erledigt sich, denn Schuldverhältnisse innerhalb des Konzerns können in einer voll konsolidierten Steuerbilanz keine Gewinnwirksamkeit mehr entfalten77. – Umstrukturierungen im Konzern oder Verlagerungen von Wirtschaftsgütern können keine Gewinnrealisierung mehr hervorrufen78. – Vor allem: Die Wahl der Finanzierung (Eigenkapital oder Fremdkapital) zwischen Konzerngesellschaften besitzt keinen wesentlichen Einfluss auf die Zuordnung von Aufkommen zwischen den Mitgliedstaaten (wenn man vermeidet, dass Finanzkapital zu den relevanten Aufteilungsfaktoren gehört) 79. Dazu gehört auch, dass Sonderregeln über beherrschte Auslandsgesellschaften (Hinzurechnungsbesteuerung) überflüssig werden80. Mit anderen Worten: Die einheitliche Konzernsteuerbemessungsgrundlage bewältigt in eindrucksvoller Weise diejenigen Friktionen, die zurzeit an den Sollbruchstellen einer internationalen Konzernbesteuerung existieren. Hinzu kommen Vorteile, die gerade auf der europaweiten Einheitlichkeit der steuerlichen Gewinnermittlung beruhen, z. B. eine gewaltige Reduktion von Befolgungskosten (compliance costs), welche durch inhaltlichen Differenzen zwischen den nationalstaatlichen Gewinnermittlungssystemen hervorgerufen werden 81. Zugleich wirft ihre Einführung aber eine mindestens ebenso so große Zahl neuer Fragen und Schwierigkeiten auf, denen im Folgenden nachgegangen wird82.

5.2 Optionslösung oder Zwangsvereinheitlichung? Eine erste Grundsatzfrage ist darauf gerichtet, ob die einheitliche Konzernsteuerbemessungsgrundlage als Option angeboten oder mit Zwangscharakter ausgestattet werden soll. Diese Weichenstellung berührt die politischen Grundlinien der Steuerharmonisierung in Europa. Versteht man Gesetzgebung auf dem Gebiet des Konzernsteuerrechts vor allem als Maßnahme zur Erleichterung und Förderung grenzüberschreitender Unternehmensbildungen, so wird man sich für eine Optionslösung 77 78

79

80

81

82

Kahle, Wpg. 2006, S. 1401 ff., 1408; Schreiber, StuW 2004, S. 212 ff, 219. Arbeitsgruppe GKKB, Fragen im Zusammenhang mit der Umstrukturierung von Unternehmen, 28.7.2006 (CCCTB\WP\039\doc\de); UNICE, Comments on document CCCTB\WP\039 – Issues related to Business Reorganisations, 2006; Spengel/Braunagel, StuW 2006, S. 34 ff., 48. Kahle, Wpg. 2006, S. 1401 ff., 1408; Schreiber, StuW 2004, S. 212 ff., 219; Spengel/Braunagel, StuW 2006, S. 34 ff., 48. Avi-Yonah/Clausing, A Proposal to Adopt Formulary Apportionment for Corporate Income Taxation – The Hamilton Project, Working Paper, 2006, S. 6, 16; Kahle, Wpg. 2006, S. 1401 ff., 1408. Mitteilung der Kommission an den Rat, das Europäische Parlament und den Wirtschafts- und Sozialausschuss v. 23.10.2001, Ein Binnenmarkt ohne steuerliche Hindernisse: Strategie zur Schaffung einer konsolidierten Körperschaftsteuer-Bemessungsgrundlage für die grenzüberschreitende Unternehmenstätigkeit in der EU, KOM (2001) 582 endg., Abschn.5; Europäische Kommission, European Tax Survey, Taxation Papers, Working Paper 3/04, S. 21 ff.; Czakert, IStR 2006, S. 561 ff., 562; Herzig, StuW 2006, S. 156 ff., 156 f.; Mintz, 11 International Tax and Public Finance (2004), S. 221 ff.; Spengel/Braunagel, StuW 2006, S. 34 ff., 46 f. Zum Stand der Arbeiten Czakert, IStR 2006, S. 561 ff.

B. Perspektiven der Konzernbesteuerung

67

aussprechen83. Unternehmensgruppen hätten dann die Wahl, ob sie im klassischen Konzernsteuerrecht – einschließlich der dargestellten Hemmnisse und Gestaltungsspielräume – verbleiben oder sich der neuen einheitlichen Bemessungsgrundlage unterwerfen. Eine solche Optionslösung weist nicht nur den Vorteil auf, dass sie von den Unternehmen als deutliche Verbesserung der steuerlichen Rahmenbedingungen in Europa wahrgenommen würde; auch von den Mitgliedstaaten könnte dies als eher zurückhaltende Beeinträchtigung ihrer nationalen Steuersouveränität akzeptiert werden84. Zugleich würde sie die betroffenen Finanzverwaltungen und Unternehmenspraktiker nicht auf breiter Front einem grundstürzenden Revirement ihres Handlungsrahmens unterwerfen. Die Europäische Kommission hat sich daher entschieden für eine lediglich optionale Einführung der Einheitlichen Bemessungsgrundlage ausgesprochen. Diese Lösung wäre jedoch zugleich mit dem Nachteil behaftet, dass das Gestaltungspotential des Konzernsteuerrechts, seine Kompliziertheit und seine Manipulationsanfälligkeit eher gesteigert als reduziert würde85. International tätigen Konzernen würde eben eine weitere Handlungsalternative zur Verfügung gestellt, ohne eine einzige existierende Regelung aufzuheben. Wer Steuerpolitik im Konzernsteuerrecht mit dem Ziel betreibt, die privatautonome Verfügung über steuerliche Bemessungsgrundlagen zu reduzieren, muss daher für eine obligatorische Einheitsbesteuerung der Unternehmensgruppe plädieren. Man kann auch über vermittelnde Lösungen nachdenken, z. B. ein Obligatorium lediglich für grenzüberschreitende Konzerne oder für Konzerne einer bestimmten Mindestgröße oder nur für börsennotierte Muttergesellschaften anzuordnen und im Übrigen eine Option anzubieten. Entschließt man sich für die optionale Ausgestaltung der einheitlichen Bemessungsgrundlage, so sollte diese in jedem Fall nur für eine Unternehmensgruppe in ihrer Gesamtheit angeboten werden, nicht jedoch ein beliebiges „Opt-In“ oder „Opt-Out“ einzelner Konzerngesellschaften ermöglicht werden86.

5.3 Internationale und nationale Konzerne Mit der Frage nach dem Zwangscharakter einer Konzernbesteuerung ist zugleich die Frage aufgeworfen, ob die einheitliche Bemessungsgrundlage lediglich bei 83

84

85

86

Kommissar Kovacs, The European Commission’s Project for a CCCTB, CFE Conference, 28.9.2006; American Chamber of Commerce to the European Union, Position Paper on CCCTB, 2005, S. 1; European Business Initiative on Taxation (EBIT), Contribution to the Commission on the CCCTB, 2006, Rz. 8. Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Umsetzung des Lissabon-Programms der Gemeinschaft: Bisherige Fortschritte und weitere Schritte zu einer gemeinsamen konsolidierten KörperschaftsteuerBemessungsgrundlage (GKKB), 5.4.2006, KOM (2006) 157 endg., Tz. 3.4. Herzig, StuW 2006, S. 156 ff., 162; Mintz, CESifo-Forum 2002, Heft 1, S. 3 ff., 8; Mintz, 11 International Tax and Public Finance (2004), S. 221 ff, 226 ff.; Mintz/Weiner, 10 International Tax and Public Finance (2003), S. 695 ff., 700; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 102. So aber der Vorschlag von European Business Initiative on Taxation (EBIT), Contribution to the Commission on the CCCTB, 2006, Rz. 13.

68

Wolfgang Schön

internationalen Konzernen Anwendung finden oder zugleich auf rein inländische Unternehmensgruppen erweitert werden soll. Es ist einerseits offensichtlich, dass einige der geschilderten Probleme des Konzernsteuerrechts – etwa die Gewinnverlagerung zwischen verschiedenen Jurisdiktionen durch schuldrechtliche Vereinbarungen – im rein nationalen Konzern nicht auftreten können. Andererseits muss beachtet werden, dass die Technik der Vollkonsolidierung durch die Vermeidung von konzerninternen Gewinnrealisierungen, im Rahmen der Verlustverrechnung und für den Abzug von Beteiligungsaufwand auch in rein inländischen Situationen Vorteile zu bieten vermag. Wer steuerliche Wettbewerbsnachteile für kleine und mittlere Unternehmen vermeiden möchte, muss daher die Anwendung der einheitlichen Bemessungsgrundlage auch für den Inlandsfall ermöglichen87. In diesen Fällen muss allerdings der freiwillige Charakter einer solchen Einheitsbesteuerung betont werden.

5.4 Die Definition der Unternehmensgruppe Die geschilderten Fragestellungen führen auf eine Problematik zu, die für Gruppenbesteuerungen aller Art von zentraler Bedeutung ist: die Definition der Außengrenzen des Konzerns. Ausgangspunkt ist die Erkenntnis, dass eine Vereinheitlichung der steuerlichen Bemessungsgrundlage für Einzelgesellschaften zwar Vorteile bei dem Abbau von Befolgungskosten mit sich führen kann88, dass aber nur ein konzernübergreifender Ansatz die wirklichen Vorteile des Reformprojekts realisieren werden89. Es geht daher darum, welche Gesellschaften als gruppenzugehörig angesehen werden können und müssen, d. h. welche finanziellen Beteiligungsverhältnisse und welches Ausmaß wirtschaftlicher Integration für eine Einheitsbesteuerung vorausgesetzt werden müssen90. Beide Elemente werfen erhebliche Probleme auf.

87

88 89

90

Arbeitsgruppe „Gemeinsame Konsolidierte Körperschaftsteuer-Bemessungsgrundlage“ (AG GKKB), Bisherige Fortschritte und künftiges Programm für die GKKB, 20.11.2006 (CCCT\WP\046\doc\de), Tz. 33; Schön in: Herzig (Hrsg.), Organschaft, 2003, S. 612 ff., 616 f.; Schön, European Taxation 2002, S. 276 ff., 280 f.; Schreiber, StuW 2004, S. 212 ff, 224 ff. Herzig StuW 2006, S. 156 ff. Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Umsetzung des Lissabon-Programms der Gemeinschaft: Bisherige Fortschritte und weitere Schritte zu einer gemeinsamen konsolidierten KörperschaftsteuerBemessungsgrundlage (GKKB), 5.4.2006, KOM (2006) 157 endg., Tz. 3.3; Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No.9/06, Tz. II.1; Hellerstein, International Transfer Pricing Journal 2005, S. 103 ff., 107; McLure, European Taxation 2002, S. 586 ff., 591.; McLure/Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 251 f., 284 Zur entsprechenden Rechtslage in den USA siehe Hellerstein/McLure, IBFD-Bulletin 2004, S. 86 ff., 83; Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, European Commission, Taxation Papers, Working Paper No. 8/05, Tz. 2.1.2 ff.

B. Perspektiven der Konzernbesteuerung

69

5.4.1 Beteiligungsschwelle Beginnt man mit der Frage nach dem Mindestumfang, über den eine Beteiligung für die Anwendung der Einheitsbesteuerung verfügen sollte, so lässt sich aus dem Gesichtspunkt der einheitlichen Leitung das gesamte Spektrum von der 100prozentigen Tochtergesellschaft bis hin zur knapp kontrollierenden Mehrheit oder sogar zur gemeinsamen Beherrschung von Joint Ventures als Anknüpfungspunkt für die Gruppenbesteuerung begründen. Allerdings muss das Projekt einer Einheitsbesteuerung im Konzern bei Beteiligungsschwellen unterhalb von 100 % mit der Problematik fertig werden, dass an der abhängigen Gesellschaft außen stehende Gesellschafter beteiligt sind, deren Gewinnansprüche Berücksichtigung finden müssen. Dafür steht folgende Alternative zur Verfügung: Man kann versuchen, die gewünschte Konsolidierung des Ergebnisses der Tochtergesellschaft nur in Höhe der prozentualen Beteiligung der kontrollierenden Gesellschaft durchzuführen und es im Übrigen bei der traditionellen Gewinnermittlung belassen91. Eine solche gespaltene Gewinnermittlung wäre jedoch in hohem Maße kontraproduktiv, da sie für die Tochtergesellschaft sämtliche Ermittlungsprobleme des separate accounting mit allen Einzelfragen der Einheitsbesteuerung kumulieren würde. Es erscheint daher unumgänglich, auf der Ebene der Gewinnermittlung der abhängigen Gesellschaft die einheitliche Bemessungsgrundlage auch auf solche Gewinnanteile zu erstrecken, deren Ausschüttung den außen stehenden Gesellschaftern zugute kommt92. Dann lässt es sich allerdings nicht vermeiden, diesen Ergebnisanteil in die Gesamtberechnung des Konzerngewinns einzubeziehen, der nach allgemeinen Faktoren auf die beteiligten Gesellschaften und Jurisdiktionen verteilt wird. Aus der Sicht der außen stehenden Gesellschafter folgt daraus, dass seine individuellen Gewinnansprüche zwar kraft Gesellschaftsrechts auf das Ergebnis der Tochtergesellschaft beschränkt sind, die jeweilige körperschaftsteuerliche Vorbelastung seiner Dividende aber weitgehend aus dem Ergebnis anderer Konzerngesellschaften und sogar aus dem Zugriff ausländischer Fisci resultieren kann 93. Dieses Ergebnis spricht dafür, die einheitliche Konzernbesteuerung nur bei sehr hohen Beteiligungsquoten – etwa ab 95 % oder gar erst bei 100 % – zur Anwendung zu bringen94.

91

92

93

94

Darstellung bei Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. II.2.1. Zuletzt: Arbeitsgruppe „Gemeinsame Konsolidierte Körperschaftsteuer-Bemessungsgrundlage“ (AG GKKB), Bisherige Fortschritte und künftiges Programm für die GKKB, 20.11.2006 (CCCT\WP\046\doc\de), Tz. 51; Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 205. Westberg, European Taxation 2002, S. 322 ff., 328; Witt, Die Konzernbesteuerung, 2006, S. 19 f., 231 ff. Schön, European Taxation 2002, S. 276 ff., 281; Schreiber, StuW 2004, S. 212 ff., 232 f.; die Arbeitsgruppe „Gemeinsame Konsolidierte Körperschaftsteuer-Bemessungsgrundlage“ (AG GKKB), Bisherige Fortschritte und künftiges Programm für die GKKB, 20.11.2006 (CCCT\WP\046\doc\de), Tz. 45 spricht sich für eine Beteiligungsschwelle oberhalb von 75 % aus.

70

Wolfgang Schön

Die Festlegung einer besonders hohen Beteiligungsquote hat aber wiederum zur Folge, dass dem herrschenden Unternehmen das faktische Wahlrecht eingeräumt wird, durch leichte Reduktion seiner prozentualen Beteiligung über die Anwendbarkeit der geschilderten Einheitsbesteuerung zu disponieren. Man sollte daher eine sehr hohe Beteiligungsgrenze nur dann einführen, wenn man sich ohnehin dafür ausspricht, die einheitliche körperschaftsteuerliche Bemessungsgrundlage als „Option“ auszugestalten. Wer sich hingegen für eine zwangsweise Anwendung der Einheitsbesteuerung stark macht, wird mit der Beteiligungsgrenze so weit heruntergehen müssen, bis das herrschende Unternehmen mit einem weiteren Abbau der Beteiligungshöhe zugleich substantielle Einflussmöglichkeiten verliert, etwa bei 75 % oder bei 50 %. 5.4.2 Wirtschaftliche Integration der Tochtergesellschaft Eng verbunden mit der Frage nach der Ausgestaltung der Beteiligungsschwellen ist die Problematik der wirtschaftlichen Integration von Tochtergesellschaften. Vor dem Hintergrund, dass die einheitliche Bemessungsgrundlage in erster Linie die steuerlichen Folgelasten stark vernetzter Konzerngesellschaften bewältigen soll, wird man auf den ersten Blick eine weitgehende wirtschaftliche Verflechtung der einzelnen Konzernglieder verlangen wollen. Die Geschichte der deutschen Organschaft legt indessen ein beredtes Zeugnis davon ab, wie aus einem wirtschaftlich geprägten Leitbild ein von formellen Gesichtspunkten beherrschter Tatbestand werden kann95. Das Rechtsinstitut der Organschaft im deutschen Körperschaftsteuerrecht, Gewerbesteuerrecht und Umsatzsteuerrecht war ursprünglich vollständig beherrscht von dem Bild der Tochtergesellschaft als unselbständiger Betriebsabteilung, die sowohl finanziell als auch organisatorisch als auch in ihren wirtschaftlichen Funktionen vollständig in das Mutterunternehmen integriert war. Diese Elemente sind – jedenfalls im Körperschaftsteuerrecht und im Gewerbesteuerrecht – systematisch abgebaut worden, so dass heute die finanzielle Mehrheitsbeteiligung und die Ergebnisverrechnung über einen Gewinnabführungsvertrag den maßgeblichen Tatbestand bilden. Auch für eine auf europäischer Ebene eingeführte Einheitliche Bemessungsgrundlage muss die Alternative zwischen einer eher wirtschaftlichen oder einer eher formalen Betrachtungsweise geklärt werden96. Sieht man im Tatbestand des modernen integrierten Unternehmens die eigentliche Begründung für eine steuerliche Einheitsbetrachtung, wird man eher die wirtschaftliche Betrachtungsweise heranziehen97. Die historische Erfahrung des deutschen Organschaftsrechts, dass nämlich 95 96

97

Zur Entwicklung Witt, Die Konzernbesteuerung, 2006, S. 10 ff. Zum pro und contra siehe Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. II.1. Schäfer/Spengel, The Impact of ICT on Profit Allocation within Multinational Groups: Arm’s Length Pricing of Formula Apportionment? ZEW Discussion Paper No. 03-53, 2003, S. 2 ff., 21; Avi-Yonah/Clausing, A Proposal to Adopt Formulary Apportionment for Corporate Income Taxation – The Hamilton Project, Working Paper, 2006, S. 6, 24.

B. Perspektiven der Konzernbesteuerung

71

die diffusen Merkmale organisatorischer und wirtschaftlicher Integration sich als faktisch schwer zu kontrollieren und in der Praxis weitgehend gestaltbar erwiesen haben, lassen es indessen angeraten sein, von dem Tatbestand faktisch-wirtschaftlicher Integration Abstand zu nehmen. Vergleichbare Erkenntnisse werden aus den USA berichtet98. Unabhängig davon, ob man die Einheitliche Bemessungsgrundlage mit Zwangscharakter ausstatten oder als Option formulieren will, sollte für die beteiligten Unternehmen und Finanzverwaltungen Klarheit über den Tatbestand einer Unternehmensgruppe herrschen99. Daher sollte ausschließlich der Umfang der gesellschaftsrechtlichen Beteiligung an einem Unternehmen den Ausschlag für die Aufnahme in die einheitlich veranlagte Unternehmensgruppe geben. Auch dann werden Sonderfragen wie etwa die disquotale Ausgestaltung von Stimmrechten und Gewinnbeteiligung, die Qualifikation hybrider Finanzierungsinstrumente oder der Einsatz mittelbarer Beteiligungen noch genügend Abgrenzungsprobleme aufwerfen100. Angedeutet werden können an dieser Stelle nur die Probleme, die aus dem Eintritt oder dem Ausscheiden einzelner Konzernglieder für die steuerliche Gewinnermittlung resultieren. Bei dem Eintritt wird eine Überleitung aus den bisher verwendeten – nationalen – Bilanzkonventionen in das Recht der einheitlichen Bemessungsgrundlage erforderlich werden101. Bei einem Ausscheiden wird man umgekehrt verfahren müssen. Ein solches Eintreten oder Ausscheiden kann zugleich die Aufteilungsfaktoren beeinflussen und damit zu einer Verschiebung der Gewinnbeteiligung zwischen den Jurisdiktionen führen.

5.5 Die Definition des steuerpflichtigen Einkommens Den nächsten Schritt bei der Entwicklung einer gemeinsamen europäischen Bemessungsgrundlage bildet die Ausarbeitung eines einheitlichen Steuerbilanzrechts, in dem Aufwendungen und Erträge nach gemeinsamen Regeln für die gesamte Unternehmensgruppe ermittelt werden102. Hier stehen sich zwei grundsätzlich verschiedene Ansätze gegenüber. Ein erster Vorschlag geht dahin, von der Entwicklung genuin europäischer Gewinnermittlungsregeln abzusehen und 98

Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 203 f.; McLure/ Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 263 ff.; Weiner, European Taxation 2001, S. 380 ff., 383 f. 99 McLure, European Taxation 2002, S. 586 ff., 591; McLure, CESifo-Forum 2005, Heft 1, S. 32 ff., 35; Schreiber, StuW 2004, S. 212 ff., 222 f.; Spengel in: Herzig (Hrsg.), Besteuerung der Europäischen Aktiengesellschaft, 2004, S. 101 ff., 110; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 94; Witt, Die Konzernbesteuerung, 2006, S. 135 ff., S. 182 ff. 100 AG GKKB, Überblick über die wichtigsten in der zweiten Sitzung der Untergruppe „Konzernbesteuerung“ aufgetretenen Fragen (CCCTB\WP\048\doc\de), Tz. 5. 101 Schön, European Taxation 2002, S. 276 ff., 282 f.; Schön in: Herzig (Hrsg.), Organschaft, 2003, S. 612 ff., 618 f. 102 In den USA können die Staaten bei der Bemessung des Gewinns für die state corporation tax auf die einheitlichen Vorgaben der Federal Income Tax zurückgreifen Hellerstein/McLure, IBFDBulletin 2004, S. 86 ff., 91 ff.; Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 201; Niznik, Tax Notes International v. 18.12.2006, S. 975 ff., 988.

72

Wolfgang Schön

jeweils konzernweit das Gewinnermittlungsrecht zur Anwendung bringen, das im Sitzland der leitenden Gesellschaft gilt. Für die Konzernbesteuerung der BASF würde europaweit das deutsche Steuerbilanzrecht herangezogen, für die Veranlagung von Sanofi-Aventis das französische Steuerbilanzrecht. Dieses Konzept ist vor einigen Jahren unter dem Titel der „Home State Taxation“103 diskutiert worden; es besticht durch inhaltliche Flexibilität und den Verzicht auf aufwendige Harmonisierungsmaßnahmen auf europäischer Ebene. Es erfordert lediglich eine Technik „gegenseitiger Anerkennung“ von Gewinnermittlungsregeln und wird von der Kommission als „Pilotprojekt“ für kleine und mittlere Unternehmen ins Spiel gebracht104. Die Vorbehalte gegen das Modell überwiegen jedoch105. Es führt zu Wettbewerbsverzerrungen, wenn inländische Tochtergesellschaften ausländischer Konzerne je nach dem Sitzstaat des Mutterunternehmens einer günstigen oder ungünstigen Besteuerung unterliegen106, und es stimmt bedenklich, wenn in den Mitgliedstaaten der Europäischen Gemeinschaft das Rennen um die Ansässigkeit von Konzernleitungen dadurch verschärft würde, wenn ein günstiges Steuerbilanzrecht im Land der Konzernmutter europaweit auf Tochtergesellschaften und Betriebsstätten exportiert werden könnte107. Zur Wahrung der Stabilität und Ausgeglichenheit der europäischen Unternehmensbesteuerung bedarf es daher einer eigenständigen gemeinschaftsrechtlichen Formulierung der steuerlichen Gewinnermittlung. Die Entwicklung eines Konsenses über den Inhalt eines einheitlichen Steuerbilanzrechts für Konzerne ist Gegenstand der aktuell von der Kommission und einer Arbeitsgruppe der Mitgliedstaaten durchgeführten Konsultationen108. Diese 103

Lodin/Gammie, Home State Taxation, 2001; Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681, Rz. 505 ff. 104 Non-Paper der Kommission v. 7.7.2004 für die informelle Tagung des Rates „Wirtschaft und Finanzen“ am 10. und 11.9.2004; dazu näher Cerioni, European Taxation 2006, S. 375 ff.; Diemer/Mors, European Taxation 2006, S. 111 ff.; Thömmes/Rasch/Koch, European Taxation 2006, S. 117 ff. 105 Näher González Sánchez/Franch Fluxa, European Taxation 2006, S. 197 ff., 201 ff.; Herzig, StuW 2006, S. 156 ff., 158; Schön, European Taxation 2002, S. 276, 284 f.; Spengel in: Herzig (Hrsg.), Besteuerung der Europäischen Aktiengesellschaft, 2004, S. 101 ff., 106 ff.; Spengel/ Frebel, StuB 2003, S. 786 ff., 789. 106 Mintz/Weiner, 10 International Tax and Public Finance (2003), S. 695 ff., 701; Schön in: Herzig (Hrsg.), Organschaft, 2003, S. 612 ff., 615 ff.; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 101 f. 107 Niznik, Tax Notes International v. 18.12.2006, S. 975 ff., 989; Schön in: Herzig (Hrsg.), Organschaft, 2003, S. 612 ff., 615 f.; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 102; Westberg, European Taxation 2002, S. 322 ff., 329. 108 Czakert, IStR 2006, S. 561 ff., 562 ff.; Herzig, StuW 2006, S. 156 ff., 162 ff.; Herzig/Gellrich, IStR 2006, S. 757 ff.; ausführlich auch Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Umsetzung des LissabonProgramms der Gemeinschaft: Bisherige Fortschritte und weitere Schritte zu einer gemeinsamen konsolidierten Körperschaftsteuer-Bemessungsgrundlage (GKKB), 5.4.2006, KOM (2006) 157 endg., Annex 2.

B. Perspektiven der Konzernbesteuerung

73

Beratungen gestalten sich naturgemäß schwerfällig – nicht nur deshalb, weil die historische Entwicklung der steuerlichen Einkünfteermittlung in den Mitgliedstaaten zu sehr disparaten Techniken geführt hat, sondern auch, weil das Steuerbilanzrecht – nicht zuletzt in der Bundesrepublik Deutschland – seit Jahrzehnten in den Dienst staatlicher Wirtschaftslenkung gestellt wird. Eine Fülle positiver Lenkungsanreize – etwa durch Sonderabschreibungen – sowie negativer Lenkungsanreize – etwa auf dem Gebiet der Verlustverrechnungsschranken – beherrscht das deutsche und ausländische Steuerbilanzrecht. Eine gemeinschaftsweite Einigung auf einheitliche Gewinnermittlungsregeln würde den Gestaltungsspielraum des lenkenden Steuergesetzgebers auf dem Gebiet des Steuerbilanzrechts massiv einschränken – daran ist bereits im Jahre 1988 der sehr viel bescheidenere Versuch einer Vereinheitlichung der körperschaftsteuerlichen Bemessungsgrundlage gescheitert109. Das Konzept der einheitlichen Bemessungsgrundlage duldet allerdings im Kern keine Abweichung vom Ziel eines streng harmonisierten Steuerbilanzrechts110. Wenn nämlich den Mitgliedstaaten für den Ansatz oder die Bewertung von Bilanzpositionen Sonderrechte und Optionen eingeräumt werden, dann scheitert die beabsichtigte grenzüberschreitende Konsolidierung des Konzernergebnisses und auch die Aufteilung des Gesamtergebnisses nach allgemeinen Berechnungsfaktoren verliert ihre rechnerische Grundlage. Der Steuergesetzgeber muss sich vielmehr darauf beschränken, steuerliche Lenkung nicht im Bereich der Gewinnermittlung, sondern ausschließlich auf der Ebene der Steuerschuld – etwa durch tax credits für bestimmte Investitionen – anzusiedeln. Bei realistischer Betrachtung besteht die einzige Chance, zu einer einheitlichen Steuerbilanz auf europäischer Ebene zu gelangen, darin, sich weitgehend an die International Accounting Standards/International Financial Reporting Standards anzulehnen. Dabei stehen die Mitgliedstaaten und die herrschende Meinung auf dem Standpunkt, dass diese Anknüpfung sich in einer gemeinsamen Terminologie weitgehend erschöpft, die IAS/IFRS lediglich einen begrifflichen und keinen inhaltlichen Ausgangspunkt bilden können111. Demgegenüber ist eine stärkere materielle

109 110

111

Zu diesem Vorentwurf siehe Kreile, DB 1988, Beilage 18. Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Umsetzung des Lissabon-Programms der Gemeinschaft: Bisherige Fortschritte und weitere Schritte zu einer gemeinsamen konsolidierten KörperschaftsteuerBemessungsgrundlage (GKKB), 5.4.2006, KOM (2006) 157 endg., Tz. 3.1 und Tz. 4; McLure, European Taxation 2002, S. 586 ff., 597; Schreiber, StuW 2004, S. 212 ff, 218. Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Umsetzung des Lissabon-Programms der Gemeinschaft: Bisherige Fortschritte und weitere Schritte zu einer gemeinsamen konsolidierten KörperschaftsteuerBemessungsgrundlage (GKKB), 5.4.2006, KOM (2006) 157 endg., Tz. 2.1, Tz. 3.2; AG GKKB v. 15.11.2005, Bisherige Fortschritte und künftige Pläne (CCCTB\WP\020\doc\de), Abschn.V; Czakert, IStR 2006, S. 561 ff., 563; Herzig, StuW 2006, S. 156 ff., 159 f.; für eine stärkere Orientierung an Zahlungsströmen Kahle, Wpg. 2006, S. 1401 ff., 1402 f., 1406 f.; Schreiber, StuW 2004, S. 212 ff., 224 ff.

74

Wolfgang Schön

Anlehnung an die IAS/IFRS zu bevorzugen112. Bei aller Kritik am Inhalt dieser Standards und der Art und Weise ihres Zustandekommens ist es aus der Sicht der steuerlichen Gewinnermittlung ein Glücksfall, dass 7000 börsennotierte Unternehmen in Europa seit dem 1.1.2005 nach diesen einheitlichen Regeln ihre Konzernbilanzen aufstellen und publizieren müssen. Diese Gesellschaften – zu denen nahezu sämtliche global player der europäischen Wirtschaft gehören – müssen ein Interesse daran haben, ihre steuerliche Planung weitgehend an denselben Zahlen auszurichten, die sie auch ihrer internen Planung und externen Information zugrunde legen. Dabei sollten bestimmte steuerliche Vorbehalte – etwa ein Verzicht auf den steuerrelevanten Ausweis nicht realisierter Gewinne113 – berücksichtigt, aber nicht auf breiter Front Abweichungen von den IAS/IFRS formuliert werden. Je weiter sich die Kommission und die Mitgliedstaaten bei ihren Beratungen von materiellen Vorgaben den IAS/IFRS entfernen, umso schwieriger wird es für die Finanzverwaltungen sein, einen inhaltlichen Konsens zu finden und einen Pakt mit den betroffenen Unternehmen schließen zu können 114. Die derzeit gewählte Technik der europäischen Arbeitsgruppen, den IAS/IFRS lediglich die Fachterminologie zu entnehmen und in der Sache zu vielfach abweichenden Lösungen zu gelangen, wird die Tendenz der Mitgliedstaaten stärken, ihre jeweils eigene Tradition der steuerlichen Gewinnermittlung zu verteidigen.

5.6 Die Technik der Konsolidierung Alleine die Einführung eines einheitlichen Steuerbilanzrechts für grenzüberschreitende Unternehmen reicht nicht aus, um die geschilderten Probleme des separate accounting substantiell zu beheben. Entscheidend ist vielmehr die Konsolidation dieser Ergebnisse über die Grenzen der beteiligten Konzerngesellschaften und Betriebsstätten und über die Grenzen der beteiligten Mitgliedstaaten hinaus. Dies bleibt nicht bei einer bloßen Zurechnung von Ergebnissen stehen, sondern schließt die Gewinnwirksamkeit von Innentransaktionen grundsätzlich aus. Allerdings kann auf eine Dokumentation dieser Innentransaktionen nicht verzichtet werden, zumal diese Wertbewegungen auf die jeweiligen Faktorallokationen einen nicht geringen Einfluss haben dürften. Problematisch ist auch die in einer solchen Konsolidierung verborgene Verlagerung von stillen Reserven über nationale Grenzen hinweg. Wenn und soweit mit einer solchen gewinnneutralen Verlagerung von Wirtschaftsgütern 112

CEPS, Achieving a Common Consolidated Corporate Tax Base in the European Union, 2005, S. 23 ff.; Schön in: Schön (Hrsg.), Steuerliche Maßgeblichkeit in Deutschland und Europa, 2005, S. 1 ff., 112 ff.; Spengel in: Herzig (Hrsg.), Besteuerung der Europäischen Aktiengesellschaft, 2004, S. 101 ff., 111 ff. 113 CEPS, Achieving a Common Consolidated Corporate Tax Base in the European Union, 2005, S. 53 f.; Schön in: Schön (Hrsg.), Steuerliche Maßgeblichkeit in Deutschland und Europa, 2005, S. 1 ff., 36 ff. 114 Siehe auch Mitteilung der Kommission an den Rat, das Europäische Parlament und den Europäischen Wirtschafts- und Sozialausschuss, Umsetzung des Lissabon-Programms der Gemeinschaft: Bisherige Fortschritte und weitere Schritte zu einer gemeinsamen konsolidierten Körperschaftsteuer-Bemessungsgrundlage (GKKB), 5.4.2006, KOM (2006) 157 endg., Tz. 4.

B. Perspektiven der Konzernbesteuerung

75

über die Grenze zugleich die Faktoren der künftigen Gewinnverteilung im Konzern zum Nachteil des abgebenden Staates beeinflusst werden, verliert dieser bisherige Belegenheitsstaat substantielle Besteuerungsansprüche. Dies ist allerdings der Preis, der bezahlt werden muss, wenn man eine einheitliche Besteuerungsgrundlage in Europa anstrebt. Um weitere Abgrenzungsschwierigkeiten zu vermeiden, empfiehlt es sich auch, sämtliche Einkünfte einer Konzerngesellschaft in den Konsolidierungskreis einzubeziehen. Zwar könnte man bei bestimmten Einkunftsarten – etwa Einkünften aus der Vermietung von unbeweglichem Vermögen – andere Allokationsmechanismen vorziehen; doch würde dieser Ansatz die im Einkommensteuerrecht bekannten und kaum bewältigten Abgrenzungsprobleme zwischen gewerblichen und nichtgewerblichen Einkünften in das Körperschaftsteuerrecht transferieren. Auch die US-Erfahrungen empfehlen keine Trennung nach business income und non-business income115.

5.7 Die Behandlung von Drittstaaten Bei der Ausgestaltung der Einheitlichen Bemessungsgrundlage kommt besondere Bedeutung der Frage zu, in welchem Umfang Einkunftsquellen in Drittstaaten in der Konzernbesteuerung berücksichtigt werden116. Dies gilt einerseits für Betriebsstätten und Tochtergesellschaften von gemeinschaftsangehörigen Gesellschaften, die in nicht der EU belegen oder ansässig sind. Es gilt andererseits für Aktivitäten von Nicht-EU-Gesellschaften auf dem Gebiet der Gemeinschaft, wenn diese Gesellschaften in eine europäische Gruppe eingebunden sind. Ausgangspunkt muss die Erkenntnis sein, dass eine weltweite Erstreckung der einheitlichen europäischen Bemessungsgrundlage eine vollkommen unrealistische Perspektive darstellt117. Sie würde nicht nur den meisten Doppelbesteuerungsabkommen mit Nicht-EU-Staaten widersprechen; sie würde auch wegen der Einbeziehung von Auslandsaktivitäten von Auslandsgesellschaften allgemeine Regeln des Völkerrechts verletzten und in der praktischen Handhabung undurchführbar erscheinen. Damit ist vorprogrammiert, dass es für jede Unternehmensgruppe relevanter Größe wegen ihrer wirtschaftlichen Aktivitäten außerhalb der EU – dies 115

Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. II.2.2; Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 202 f.; McLure, European Taxation 2002, S. 586 ff., 592; Weiner, European Taxation 2001, S. 380 ff., 384. 116 Dazu AG GKKB, 17.2.2006, Räumlicher Anwendungsbereich der GKKB (CCCTB\WP\ 026\doc\de) sowie ausführlich Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. II.3. 117 Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 206; Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, European Commission, Taxation Papers, Working Paper No. 8/05, Tz. 4.3.1; anders wohl Avi-Yonah/Clausing, A Proposal to Adopt Formulary Apportionment for Corporate Income Taxation – The Hamilton Project, Working Paper, 2006, S. 6, 21.

76

Wolfgang Schön

beginnt bei der Schweiz und den USA und führt über Japan in sämtliche Entwicklungs- und Schwellenländer – doch wieder einen Bereich geben wird, für den die klassischen Regeln des separate accounting Geltung beanspruchen, d. h. nach Steuersubjekten gesondert Gewinne ermittelt und insbesondere konzerninterne Schuldverhältnisse anerkannt und auf ihre Angemessenheit überprüft werden müssen118. Man muss sich darüber im Klaren sein, dass diese Trennung zwischen EUGesellschaften (innerhalb der einheitlichen Gewinnermittlung) und Nicht-EUGesellschaften (außerhalb der einheitlichen Gewinnermittlung) nicht nur zu schwerwiegenden verwaltungstechnischen Störungen, sondern auch zu erheblichem Gestaltungspotential führen wird. Eine europäische Unternehmensgruppe mit einer Finanzierungsgesellschaft in der Schweiz kann eine Vielzahl von Zielen der Einheitlichen Bemessungsgrundlage letztlich vereiteln, indem doch wieder mit Hilfe von Fremdkapitalfinanzierungen Gewinne aus der Gemeinschaft heraus transferiert werden119. US-amerikanische oder japanische Konzernobergesellschaften können für ihre europäischen Aktivitäten die Vorteile der einheitlichen Bemessungsgrundlage in Anspruch nehmen und gleichzeitig über die Verrechnungspreisgestaltung zwischen den Kontinenten die Allokation von Gewinnen auf EU-Gesellschaften und Nicht-EU-Gesellschaften beeinflussen. Diese Gesellschaften leben gleichsam in der „besten aller Welten“, aber auch in der „schlechtesten aller Welten“, weil sie die Verpflichtungen und Fallstricke aus zwei grundverschiedenen Konzernsteuersystemen parallel beachten müssen. Hinzu treten rechtstechnische Probleme, etwa diejenigen der Anwendung von Doppelbesteuerungsabkommen auf die zwischen der EU und Drittstaaten fließenden Zahlungsströme, wenn – wie in der Vergangenheit – einerseits zwischen den einzelnen Mitgliedstaaten der EU und den Drittstaaten jeweils bilaterale Doppelbesteuerungsabkommen Anwendung finden und andererseits innerhalb der EU die territoriale und persönliche Zuordnung von Gewinnelementen innerhalb der Unternehmensgruppe gerade keine herausragende Rolle mehr spielen soll. Aus der Sicht der Doppelbesteuerungsabkommen wird man für den Abfluss von Zinsen und Dividenden in Drittstaaten doch wieder prüfen müssen, aus welchem Staat diese Zahlungsströme an Empfänger in Drittstaaten fließen. Bei Einkünften aus Drittstaaten muss je nach der Vereinnahmung durch einen Steuerpflichtigen aus einem bestimmten EU-Mitgliedstaat eine Freistellung oder eine bloße Steueranrechnung erfolgen. Dies wird zu Verwerfungen führen, weil die über einen Mitgliedstaat bezogenen Drittstaatseinnahmen im Rahmen der Konsolidierung auf sämtliche beteiligten 118

Arbeitsgruppe „Gemeinsame Konsolidierte Körperschaftsteuer-Bemessungsgrundlage“ (AG GKKB), Bisherige Fortschritte und künftiges Programm für die GKKB, 20.11.2006 (CCCT\WP\046\doc\de), Tz. 23 ff.; Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 206 f.; zum Transfer Pricing siehe AG GKKB v. 5.12.2006, Related Parties in CCCTB (CCCTB\WP\041\doc\en). 119 Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. II.3.2; Kahle, Wpg. 2006, S. 1401 ff., 1408; Schreiber, StuW 2004, S. 212 ff, 219; Weiner, European Taxation 2001, S. 380 ff., 384.

B. Perspektiven der Konzernbesteuerung

77

Jurisdiktionen verteilt werden und eben nicht (auch nicht für Zwecke der Anrechnung oder Freistellung) einem einzelnen Mitgliedstaat zugewiesen werden können120. Sinnvoll, wenn auch missbrauchsanfällig, wäre die völlige Freistellung von Einkünften aus Quellen außerhalb des Konsolidierungskreises. Die Gestaltung und Kontrolle von Verrechnungspreisen mit nicht beteiligten Staaten betrifft unter dem neuen Regime nicht lediglich einzelne Mitgliedstaaten im bilateralen Verhältnis, sondern hat Auswirkungen für alle der Einheitlichen Bemessungsgrundlage angeschlossenen Länder121. Verlangt werden müsste daher auch eine einheitliche Verrechnungspreiskontrolle für alle EU-Staaten gegenüber Drittstaaten. Schließlich wäre es nicht hilfreich, wenn im Bereich der Finanzierungsgeschäfte Missbrauchsregelungen gegenüber Drittstaaten unterschiedlich gefasst und gehandhabt werden122. Schließlich erscheint es rechtstechnisch kaum möglich, die Einkünfte europäischer Betriebsstätten von in Drittstaaten ansässigen Konzerngesellschaften in die einheitliche Gruppenbesteuerung einzubeziehen, weil für ein kohärentes Steuersystem dann auch die ausländischen Einkunftsteile dieser Konzernglieder verrechnet werden müssten123. Nicht ausgeschlossen ist hingegen eine freiwillige Einbeziehung von Gesellschaften aus Drittstaaten, wenn entsprechende völkerrechtliche Vereinbarungen mit den jeweiligen Ansässigkeitsstaaten geschlossen werden. Eine weitere Verkomplizierung wird dann eintreten, wenn nicht alle Mitgliedstaaten der EU sich der einheitlichen Bemessungsgrundlage anschließen sollten. Dann muss man mit einer Drittstaatenproblematik auch innerhalb der EU selbst rechnen, weil die Grenzlinie zwischen der Einheitsbetrachtung und dem traditionellen separate accounting dann zwischen den Mitgliedstaaten der EU verlaufen 120

Arbeitsgruppe „Gemeinsame Konsolidierte Körperschaftsteuer-Bemessungsgrundlage“ (AG GKKB), Bisherige Fortschritte und künftiges Programm für die GKKB, 20.11.2006 (CCCT\WP\046\doc\de), Tz. 18; Arbeitsgruppe GKKB v. 18.11.2005, Internationale Aspekte (CCCTB\WP\019\doc\de), Tz. 20; AG GKKB, 17.2.2006, Räumlicher Anwendungsbereich der GKKB (CCCTB\WP\026\doc\de), Tz. 15 ff.; Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. II.3.1; European Business Initiative on Taxation (EBIT), Contribution to the Commission on the CCCTB, 2006, Rz. 21. 121 Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 95; Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, European Commission, Taxation Papers, Working Paper No. 8/05, S. 19; American Chamber of Commerce to the European Union, Position Paper on CCCTB, 2005, S. 2. 122 Arbeitsgruppe GKKB v. 18.11.2005, Internationale Aspekte (CCCTB\WP\019\doc\de), Tz. 37; Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, European Commission, Taxation Papers, Working Paper No. 8/ 05, Tz. 4.3.3. 123 Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. II.3.2; für ein „Opt-In“-Modell EBIT European Business Initiative on Taxation (EBIT), Contribution to the Commission on the CCCTB, 2006, Rz. 19 f.

78

Wolfgang Schön

wird124. Die Besonderheit wird dann allerdings darin bestehen, dass sämtliche Gesellschaften aus EU-Mitgliedstaaten sowohl die Vorzüge des europäischen Sekundärrechts – etwa der Mutter-Tochter-RL oder der Richtlinie über Zinsen und Lizenzgebühren im Konzern – in Anspruch nehmen können als auch die Grundfreiheiten des EG-Vertrages. Internationale Konzerne haben es dann in der Hand, durch die Ansiedlung von Gruppengesellschaften in allen drei Jurisidiktionstypen – im engen Kreis der einheitlichen Bemessungsgrundlage, im weiteren Kreis der EUMitgliedstaaten und im äußeren Kreis der Nicht-EU-Staaten nahezu nach Belieben über die Anwendbarkeit rechtlicher Grundlagen für die Konzernbesteuerung zu befinden. Die Konzernsteuerplanung wird nicht ihr Ende finden, sondern zu neuen Höhenflügen aufbrechen.

5.8 Die Anwendung von Zuordnungsfaktoren Die entscheidende politische Neuorientierung, die mit dem Konzept einer einheitlichen Konzernsteuerbemessungsgrundlage verbunden ist, liegt in einer neuen Methode für die Aufteilung des Konzerngewinns zwischen den beteiligten Mitgliedstaaten125. Der bisherige transaktionsbezogene Ansatz, der in einem ersten Schritt von den handelsrechtlichen Ergebnissen der jeweiligen Konzerngesellschaften ausgeht und im zweiten Schritt die Angemessenheit der Leistungsverhältnisse zwischen den Konzerngesellschaften am Maßstab des arm’s-lengthGrundsatzes prüft, soll entfallen. An deren Stelle würden grob geschnittene Verteilungsfaktoren treten, z. B. der im jeweiligen Staat anfallende Personalaufwand, das dort belegene Vermögen oder das Maß der jeweiligen Außenumsätze126. Auf diese Weise soll versucht werden, die Gewinne dort zu allozieren, wo die wesentlichen wirtschaftlichen Grundlagen für ihre Erzielung angesiedelt sind. Dabei muss die Auwahl und Gewichtung der Faktoren für alle beteiligten Mitgliedstaaten einheitlich vorgenommen werden, um Doppel- oder Nichtbesteuerungen zu vermeiden127. Wer das Projekt einer solchen Aufteilungsmethode von vornherein für fehlgeleitet hält, mag aus zwei Gründen innehalten: Zum einen werden derartige Formelaufteilungen (formulary apportionment) in einigen großen Industriestaaten schon praktiziert, z. B. in den USA oder in Kanada für die Zuordnung von Körperschaftsteuern

124

Zur Problematik Arbeitsgruppe GKKB v. 18.11.2005, Internationale Aspekte (CCCTB\WP\ 019\doc\de), Tz. 1; Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. II.3.1 (i); Cerioni, European Taxation 2006, S. 375, 191 f.; European Business Initiative on Taxation (EBIT), Contribution to the Commission on the CCCTB, 2006, Rz. 23. 125 Hey, StuW 2004, S.193 ff., 205 f. 126 Wellisch, StuW 2004, S. 267 ff., 268 ff.; Wellisch, 60 Finanzarchiv (2004), S. 24 ff., 25 ff. 127 Arbeitsgruppe GKKB v. 17.11.2006, Mechanismus zur Aufteilung der GKKB (CCCTB\WP\ 047\doc\de), Tz. 14; Weiner, European Taxation 2001, S. 380 ff., 384 f.; Wellisch, StuW 2004, S. 267 ff., 270 f.

B. Perspektiven der Konzernbesteuerung

79

an die Einzelstaaten und Provinzen128, und selbst in Deutschland wird das Gewerbesteueraufkommen zwischen den Gemeinden schlicht nach Maßgabe der jeweiligen Lohnsumme verteilt. Zum anderen haben sich bei der Anwendung der traditionellen arm’s-length-Regel im der Praxis des internationalen Steuerrechts auf Dienstleistungen und Immaterialgüter wegen des häufigen Fehlens von Vergleichspreisen für konzerninterne Wertverschiebungen so genannte Gewinnaufteilungsmethoden entwickelt, die letztlich ebenfalls darauf abzielen, den Konzerngewinn nach Maßgabe der ergebnisrelevanten Produktionsfaktoren in grober Weise aufzuspalten129. Die entscheidende Frage zielt natürlich auf die Auswahl der Zuordnungsfaktoren für den Konzerngewinn. Vor dem Hintergrund, dass es dazu keine „wissenschaftlich eindeutige“ Lösung gibt130, hat sich zu diesen Fragen schon vor Jahrzehnten insbesondere in den USA eine ausgedehnte Diskussion entwickelt, die in den letzten Jahren auch die europäische Fachwissenschaft erreicht hat. Versucht man, die Problematik auf einen Nenner zu bringen, so geht es darum, einen Zuordnungsfaktor zu finden, der einerseits von der Finanzverwaltung ohne erheblichen Aufwand ermittelt werden und vom Steuerpflichtige nicht manipuliert werden kann, der andererseits aber nicht seinerseits zu einer Verzerrung der Allokation von Ressourcen führt. Schließlich muss der Aufteilungsfaktor auch den Ansprüchen der nationalen Steuerrechtsordnung an die Belastungsgleichheit und die Grundregeln des internationalen Steuerrechts über die Anknüpfung der Besteuerungshoheit genügen. Von vornherein problematisch unter dem Gesichtspunkt der Steuergerechtigkeit sind daher Vorschläge einer Aufteilung des individuellen Konzerngewinns einer Unternehmensgruppe nach makroökonomischen Faktoren, etwa dem Verhältnis der Bruttosozialprodukte der betroffenen Staaten131 oder der Gewichte der vertretenen Industrien132. Logisch vertretbar, aber anfechtbar unter dem Gesichtspunkt der internationalen Steuerhoheit wäre wohl auch eine marktbezogene Anknüpfung nach dem Sitz des Empfängers von Lieferungen und Leistungen, weil dies zu einer Faktorzuordnung an Staaten führen würde, in denen der Konzern weder über eine 128

Darstellung bei Hellerstein, International Transfer Pricing Journal 2005, S. 103 ff., 110 f.; Hellerstein/McLure, IBFD-Bulletin 2004, S. 86 ff., 94 f.; McLure, European Taxation 2002, S. 586 ff., 589; Weiner, European Taxation 2001, S. 380 ff., 381 ff.; Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, European Commission, Taxation Papers, Working Paper No. 8/05, Tz. 2.1.2 ff. 129 Avi-Yonah/Clausing, A Proposal to Adopt Formulary Apportionment for Corporate Income Taxation – The Hamilton Project, Working Paper, 2006, S. 6, 30; McLure McLure/Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 246 f.; Schäfer/Spengel, The Impact of ICT on Profit Allocation within Multinational Groups: Arm’s Length Pricing of Formula Apportionment?, ZEW Discussion Paper No. 03-53, 2003, S. 2 ff., 17 f.; Spengel/Braunagel, StuW 2006, S. 34 ff., 48 f. 130 McLure, European Taxation 2002, S. 586 ff., 587; McLure/Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 246 f. 131 McLure, European Taxation 2002, S. 586 ff., 596; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 96. 132 Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 211; Wellisch, StuW 2004, S. 267 ff., 271 f.

80

Wolfgang Schön

ansässige Gesellschaft noch eine Betriebsstätte, sondern lediglich über eine Kundenbasis verfügt133. Vor diesem Hintergrund kommen im Wesentlichen drei Anknüpfungsfaktoren in den Blick: die Lohnsumme, das Vermögen und die Umsätze. Jede Faktorzuordnung hat zunächst den Nachteil, dass sie wie eine Steuer auf die jeweiligen Produktionsfaktoren wirkt und daher Ausweicheffekte zur Folge haben kann134. Auch erscheint es problematisch, für alle Produktionsfaktoren in Abhängigkeit von den ihnen zugeordneten Kosten gleiche Produktivität und gleiche Erträge zu unterstellen 135. Dies hat z. B. zur Folge, dass im Verhältnis von Hochlohn- und Niedriglohnländern bei gleicher Arbeitsproduktivität das Hochlohnland den höheren Steueranteil generiert – obwohl de facto dort ein niedrigerer Unternehmensgewinn festzustellen ist136. Soweit Unternehmen über die Zuordnungsfaktoren verfügen können, ist daher immer mit Verlagerungseffekten – und damit einer neuen Form der Steuerplanung – zu rechnen137. Im Übrigen weisen die genannten Faktoren jeweils positive und negative Merkmale auf. Legt man die Lohnsumme der Zuordnung des Einheitsgewinns zugrunde, so handelt es sich um eine einfach zu ermittelnde und weitgehend manipulationsresistente Größe138. Allerdings wirkt diese Zuordnung wie eine Zusatzsteuer auf den Lohn; sie kann je nach dem Verhältnis der Steuersätze und des Lohnniveaus Arbeitsplatzverlagerungen zur Folge haben. Bedenkt man weiterhin, dass die Körperschaftsteuer im Grundsatz die Rendite auf das eingesetzte Kapital abschöpfen soll, während der Einsatz des Faktors Arbeit von der Lohnsteuer erfasst werden soll, bietet sich eher die Wahl des jeweiligen Kapitaleinsatzes als Anknüpfungsfaktor an139. Eine Zuordnung nach Maßgabe der Belegenheit von Vermögenswerten ist 133

McLure, European Taxation 2002, S. 586 ff., 590; Schäfer/Spengel, The Impact of ICT on Profit Allocation within Multinational Groups: Arm’s Length Pricing of Formula Apportionment?, ZEW Discussion Paper No. 03-53, 2003, S. 2 ff., 7; Schreiber, StuW 2004, S. 212 ff, 221; zu der Notwendigkeit einer „Throwback“-Regel für den Fall extraterritorialer Verkaufsfälle siehe Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 212. 134 McLure, European Taxation 2002, S. 586 ff., 592; McLure/Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 278; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 95 f.; Weiner, European Taxation 2001, S. 380 ff., 382 f.; Wellisch, 60 Finanzarchiv (2004), S. 24 ff., 16 (mit der Hervorhebung, dass diese Form indirekter Besteuerung durch ihre Intransparenz erhöht problematisch erscheint). 135 McLure, European Taxation 2002, S. 586 ff., 587; McLure/Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 258; Mintz/Weiner, 10 International Tax and Public Finance (2003), S. 695 ff., 698; P. Musgrave in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 46 ff., 56. 136 Arbeitsgruppe GKKB v. 17.11.2006, Mechanismus zur Aufteilung der GKKB (CCCTB\WP\ 047\doc\de), Tz. 15. 137 Schreiber, StuW 2004, S. 212 ff., 220; Wellisch, 60 Finanzarchiv (2004), S. 24 ff., 25 f. 138 Schreiber, StuW 2004, S. 212 ff., 222. 139 Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. IV.2.2.1 (i); Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 209; P. Musgrave in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S.46 ff., 56; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 96 f.

B. Perspektiven der Konzernbesteuerung

81

daher systematisch überzeugender, weist in der praktischen Durchführung jedoch größere Schwierigkeiten auf. Dabei empfiehlt es sich zunächst, in einem ersten Schritt das Finanzkapital und seine Allokation vollständig aus der Faktorbestimmung herauszurechnen, da in der Wahl zwischen Fremd –und Eigenkapital im geltenden System die größten steuerpolitischen Probleme liegen und diese Gestaltungsmöglichkeiten nicht in das neue System übertragen werden dürfen140. Betrachtet man konsequent lediglich die Aktivseite der Bilanz, das Anlage- und das Umlaufvermögen, so wird man zwar für Grundstücke und den wesentlichen Teil beweglicher Sachen keine Schwierigkeiten mit der territorialen Zuordnung haben, aber schon bei den heute wesentlichen Immaterialgütern – und den werttreibenden Verbundeffekten – erweist sich die territoriale Aufteilung der maßgeblichen Güter als problematisch141. Schwer zu bewältigen sind schließlich die Probleme der Bewertung dieses Zuordnungsfaktors. Legt man die Bilanzwerte des Vermögens zugrunde, so können hochwertige Produktionsfaktoren – etwa selbstgeschaffene immaterielle Wirtschaftsgüter – außer Blick geraten – dadurch würden Fabrikationsstätten gegenüber Forschungs- und Entwicklungseinrichtungen oder Zentralfunktionen deutlich an Gewicht zunehmen142. Legt man hingegen die Verkehrswerte des Vermögens zugrunde, so bedarf es parallel zur bilanziellen Gewinnermittlung einer laufenden Wertfortschreibung für Zwecke der Faktoraufteilung. Eine gewisse Erleichterung kann vielleicht die Bezugnahme auf fair values aus den handelsrechtlichen IAS/IFRS-Bilanzen bieten, doch sind die rechtlichen und tatsächlichen Grundlagen in diesem Bereich noch weitgehend im Fluss143. Eher leicht festzustellen sind wiederum umsatzorientierte Faktoren, etwa das Maß der entgeltlichen Lieferungen und Leistungen. Aus US-amerikanischer Sicht wird vielfach eine an Verkäufen im Bestimmungsstaat ausgerichtete Zuordnung diskutiert, weil diese die Nutzung des jeweiligen Marktes widerspiegelt und der Kun140

McLure, European Taxation 2002, S. 586 ff., 595. Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. IV.2.2.1 (i); Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 211 f.; McLure, European Taxation 2002, S. 586 ff., 595; McLure, Tax Notes International v. 10.3.1997, S. 859 ff.; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 97; Schäfer/Spengel, The Impact of ICT on Profit Allocation within Multinational Groups: Arm’s Length Pricing of Formula Apportionment?, ZEW Discussion Paper No. 03-53, 2003, S. 2 ff., 24. 142 Gleiches gilt, wenn Immaterialgüter – wie vielfach in den USA – von vornherein außer Betracht bleiben; näher Arbeitsgruppe GKKB v. 17.11.2006, Mechanismus zur Aufteilung der GKKB (CCCTB\WP\047\doc\de), Tz. 16; Herzig, StuW 2006, S. 156 ff., 161; McLure, Tax Notes International v. 10.3.1997, S. 865 f.; McLure, European Taxation 2002, S. 586 ff., 594; McLure/ Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 269 f.; Schäfer/Spengel, The Impact of ICT on Profit Allocation within Multinational Groups: Arm’s Length Pricing of Formula Apportionment? ZEW Discussion Paper No. 03-53, 2003, S. 2 ff., 23; Schreiber, StuW 2004, S. 212 ff., 221; Wellisch, StuW 2004, S. 267 ff., 273. 143 Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. IV.2.2.1 (i). 141

82

Wolfgang Schön

denstamm wenig Mobilität aufweist und daher auch nicht leicht manipuliert werden kann144. Aus europäischer Sicht überwiegen jedoch die Bedenken. Zunächst wird innerhalb der Europäischen Union die Nutzung der Nachfragekraft von Konsumenten bereits hinreichend durch die harmonisierte Umsatzsteuer erfasst. Die Körperschaftsteuer soll daher verstärkt den Beitrag des Produktionsstandorts und die Rendite des dort eingesetzten Kapitals in den Blick nehmen145. Natürlich kann man die Besteuerung auch danach ausrichten, von welcher Betriebsstätte oder Tochtergesellschaft Umsätze ausgeführt werden. Die Problematik liegt dann in der Gestaltbarkeit des Leistungsortes146. Darf es einen Unterschied machen, ob der französische Markt unmittelbar von einer deutschen Produktionsstätte, von einer französischen Vertriebsgesellschaft oder einem Standort in Luxemburg mit Gütern versorgt wird? Diese jeweiligen Schwierigkeiten haben in der bisherigen Diskussion zu der Schlussfolgerung geführt, dass nur eine Mischformel – etwa aus gleichgewichtiger Heranziehung von Lohnsumme, Vermögenswerten und Umsatzgrößen – der Formelaufteilung dienen soll. Mit der Vermehrung der Zuordnungsfaktoren erhöht sich naturgemäß die Komplexität des Systems sowie seine Fehler- und Streitanfälligkeit. Gleiches gilt für Überlegungen, branchenspezifische Formeln vorzusehen – etwa für Produktions- Handels- und Finanzdienstleistungsunternehmen. Vielfach wird auch gemeint, dass es letztlich nur ein maßgebendes Kriterium für die Auswahl der Verteilungsfaktoren geben könne, nämlich der politische Konsens zwischen den Mitgliedstaaten. Speziell für den europäischen Bereich wird schließlich darüber nachgedacht, eine Zuordnung nach dem umsatzsteuerlichen Mehrwert vorzunehmen, d. h. dass der nach der 6. USt-Richtlinie festgestellte Gesamtumsatz des Konzerns auf die einzelnen Mitgliedstaaten verteilt und entsprechend auch der Gewinn zugeordnet wird147. Dies hätte den Vorteil einer Verfeinerung der Faktorallokation, wäre aber wiederum den Einzelproblemen einer individualisierenden Berechnung ausgesetzt. Zwar würde die Finanzstruktur bei einer solchen mehrwertorientierten Betrachtung 144

Avi-Yonah/Clausing, A Proposal to Adopt Formulary Apportionment for Corporate Income Taxation – The Hamilton Project, Working Paper, 2006, S. 6, 12; McLure, CESifo-Forum 2005, Heft 1, S. 32 ff., 35; Mintz/Weiner, 10 International Tax and Public Finance (2003), S. 695 ff., 702 f. 145 Wellisch, StuW 2004, S. 267 ff., 272; eine sinnvolle Aufteilung der „Beiträge“ des Produktionsstaates und des Konsumtionsstaates ist ohne hin nicht zu leisten (Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 209 f.; McLure, European Taxation 2002, S. 586 ff., 593; Schäfer/Spengel, The Impact of ICT on Profit Allocation within Multinational Groups: Arm’s Length Pricing of Formula Apportionment? ZEW Discussion Paper No. 03-53, 2003, S. 2 ff., 6 f.; Schreiber, StuW 2004, S. 212 ff., 221. 146 Arbeitsgruppe GKKB v. 17.11.2006, Mechanismus zur Aufteilung der GKKB (CCCTB\WP\ 047\doc\de), Tz. 17; Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 213; McLure/Weiner, in: Cnossen (Hrsg.), Taxing Capital Income in the European Union, 2000, S. 243 ff., 270; Weiner, European Taxation 2001, S. 380 ff., 385. 147 Ausführliche Diskussion bei Agúndez-García, The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, European Commission, Taxations Papers, Working Paper No. 9/06, Tz. IV.2.2.2.

B. Perspektiven der Konzernbesteuerung

83

außer Acht bleiben, doch müssten für die Innenumsätze zwischen den Konzerngesellschaften Feststellungen und Angemessenheitsprüfungen durchgeführt werden, die wieder in die Problematik der Verrechnungspreise hineinführen148.

5.9 Institutionelle Fragen Die Einführung einer einheitlichen steuerlichen Bemessungsgrundlage wirft schließlich im Bereich der institutionellen Verwirklichung eine Fülle von Fragen auf, etwa die nach der richtigen Rechtsgrundlage – Richtlinie, Verordnung? – und nach der Einrichtung von begleitenden Gremien, die inzwischen im europäischen Bilanz- und Gesellschaftsrecht eine wichtige Rolle spielen149. Nimmt man die Verwirklichungsperspektiven in den Blick, so ist damit zu rechnen, dass allenfalls eine Kerngruppe europäischer Mitgliedstaaten – Deutschland, Frankreich, Österreich, Niederlande etc. – bereit sein werden, die für die Einführung einer solchen Bemessungsgrundlage erforderliche Koordination und den damit einhergehenden Souveränitätsverzicht zu leisten. Für Rechtssetzung im Rahmen einer solchen Kerngruppe wird bereits die Anwendung der Regeln über „verstärkte Zusammenarbeit“ ins Auge gefasst, die durch den Vertrag von Nizza in das Recht der Europäischen Union aufgenommen worden sind150. Sie würden es einer Gruppe von mindestens acht teilnehmenden Staaten ermöglichen, untereinander verbindliche Rechtssätze zu schaffen. Andere Staaten könnten sich – wenn die einheitliche Bemessungsgrundlage sich als Erfolgsmodell herausstellt – zu einem späteren Zeitpunkt anschließen. Die Erfahrungen mit der Schwerfälligkeit europäischer Rechtssetzung – etwa die Einschränkungen der mitgliedstaatlichen Gesetzgebung durch die Mutter-Tochter-RL oder durch die Richtlinie über Zinsen und Lizenzgebühren im Konzern – legen es aber auch nahe, den teilnehmenden Mitgliedstaaten von vornherein ein „Ausstiegsszenario“ zu präsentieren. Die Unwägbarkeiten, welche die Einführung einer einheitlichen Bemessungsgrundlage für Unternehmen und Finanzverwaltungen mit sich bringen wird, sind enorm. Die Vereinheitlichung der Bemessungsgrundlage wird nicht nur den Steuerwettbewerb auf der Ebene der Steuersätze und damit auch das Problem der Standortverlagerungen anfeuern. Hinzu treten die Sorgen um unvorhersehbare Gestaltungsoptionen und Steuerausfälle. Es wäre daher 148

Hellerstein/McLure, 11 International Tax and Public Finance (2004), S. 199 ff., 214 ff.; McLure, CESifo-Forum 2005, Heft 1, S. 32 ff., 35 f.; McLure, European Taxation 2002, S. 586 ff., 594; Mintz/Weiner, 10 International Tax and Public Finance (2003), S. 695 ff., 699; Niznik, Tax Notes International v. 18.12.2006, S. 975 ff., 992; Schön, European Taxation 2002, S. 276 ff., 284; Sørensen, 11 International Tax and Public Finance (2004), S. 91 ff., 97; Weiner, Formulary Apportionment and Group Taxation in the European Union: Insights from the United States and Canada, European Commission, Taxation Papers, Working Paper No. 8/05, Tz. 5.5.2; Wellisch, StuW 2004, S. 267 ff., 273 f.; Westberg, European Taxation 2002, S. 322 ff., 326 f. 149 Schön, European Taxation 2004, S. 426 ff. 150 Unternehmensbesteuerung im Binnenmarkt, Arbeitsdokument der Dienststellen der Kommission, SEK (2001) 1681, Rz. 538 ff.; European Commission, Briefing Note: Enhanced Cooperation in the Taxation Area v. 4.4.2006; ausführlich Cerioni, European Taxation 2006, S. 375 ff., 187 ff.; McLure, Legislative, Judicial, Soft Law, and Cooperative Approaches to Harmonizing Corporate Income Taxes in the US and the EU, 2006.

84

Wolfgang Schön

von vornherein sinnvoll, den Beitritt zu einer solchen einheitlichen Bemessungsgrundlage im Rahmen der verstärkten Zusammenarbeit mit einem „Kündigungsrecht“ der Mitgliedstaaten zu versehen. Dies würde nicht nur in der nachfolgenden Anwendungspraxis der beteiligten Staaten zu einer gewissen Ausgewogenheit und wechselseitigen Rücksichtnahme führen, sondern es könnte zugleich vielen Mitgliedstaaten den Beitritt erleichtern und damit von vornherein für größere Akzeptanz und breitflächige Durchsetzung sorgen. Nur erwähnt werden können in diesem Zusammenhang die verwaltungstechnische Bewältigung der Einheitsbesteuerung. Nicht vorgesehen und auch vollständig unrealistisch ist eine Harmonisierung des nationalen Steuerverwaltungsrechts. Benötigt wird allerdings eine Verstärkung der Zusammenarbeit über die bisherigen Richtlinien zur Amtshilfe und Beitreibung hinaus. Eine deutliche Vereinfachung würde es mit sich führen, wenn eine zentrale Zuständigkeit der Steuerverwaltung am Ort der Konzernleitung geschaffen würde. Dies setzt aber ein erhebliches Vertrauen der übrigen Finanzverwaltungen voraus, nicht nur im Hinblick auf die Gewinnermittlung des Konzerns, sondern auch und vor allem im Hinblick auf die Feststellung und Bewertung der jeweiligen Verteilungsfaktoren.

5.10 Schlusswort Es mag sein, dass die geschilderten Überlegungen zu einer einheitlichen Bemessungsgrundlage im Konzern für die Körperschaftsbesteuerung in Europa eine blanke Utopie darstellt. Die Europäische Kommission ist in der Vergangenheit mit sehr viel bescheideneren Projekten gescheitert. Die Größe des Vorhabens ist aber letztlich nur an der Größe der Probleme orientiert, mit denen das Körperschaftsteuerrecht in Europa konfrontiert ist.

C. Taxable Corporate Income – Common Structural Elements

85

C. Taxable Corporate Income – Common Structural Elements I. CCCTB: Der steuerpflichtige Unternehmensgewinn – gemeinsame Strukturelemente Gunter Mayr

1. Einleitung Die CCCTB (Common Consolidated Corporate Tax Base) setzt die Beantwortung von folgenden drei Fragen voraus: – Wie sind die Gewinnermittlungsvorschriften der gemeinsamen Bemessungsgrundlage ausgestaltet? – Wie erfolgt die Konsolidierung? – Nach welchen Faktoren erfolgt die Aufteilung des konsolidierten Ergebnisses auf die Mitgliedstaaten? – Wie sieht der Aufteilungsschlüssel aus? Mit der Beantwortung jeder dieser drei Fragen würde die CCCTB an sich funktionieren. Zudem wirft die CCCTB natürlich noch weitere wichtige Fragen auf (zB: Soll die CCCTB optional sein?1), doch dabei geht es weniger um das grundsätzliche Funktionieren der CCCTB, sondern gewissermaßen um die „Einführungsmodalitäten“. Der Beitrag beschäftigt sich mit der ersten Fragestellung für die CCCTB: Es geht um die konkrete „steuertechnische“ Ausgestaltung der gemeinsamen Bemessungsgrundlage und damit um die Regeln der steuerlichen Gewinnermittlung. Es werden hier einerseits die wichtigsten Arbeitsergebnisse der Unterarbeitsgruppe 1 in Berlin (Wirtschaftsgüter und steuerliche Abschreibungen, Veräußerungsgewinne und -verluste), der Unterarbeitsgruppe 2 in Rom (Rücklagen, Rückstellungen und Verbindlichkeiten) und der Unterarbeitsgruppe 3 in Paris (Ermittlung des steuerlichen Einkommens) zusammengefasst,2 andererseits wichtige und intensiv diskutierte Punkte sowie Grundsätze herausgegriffen und stärker beleuchtet.

2. Problemstellung Das Finden und Konkretisieren einer gemeinsamen Bemessungsgrundlage ist eine Kernaufgabe für die CCCTB, wobei der Zweck einer gemeinsamen Bemessungsgrundlage sogleich erkennbar ist: Da die direkten Steuern in der EU nicht harmonisiert sind, weichen auch die Gewinnermittlungsvorschriften in den Mitgliedstaaten 1 2

Dazu unten Punkt 5. Vgl. dazu auch Czakert, IStR 2006, 561.

86

Gunter Mayr

stark voneinander ab, was für grenzüberschreitend tätige Unternehmen mit erheblichen Mühen verbunden ist. Auch lassen sich die in den einzelnen Mitgliedstaaten erzielten Gewinne nicht seriös miteinander vergleichen. Entsprechend ungenau sind auch Studien und Statistiken zu den effektiven KSt-Steuersätzen in der EU, was man auch daran erkennt, dass sich je nach zugrunde liegender Berechnungstechnik unterschiedliche Ergebnisse ergeben. Grundsätzlich bringt bereits eine gemeinsame Bemessungsgrundlage den grenzüberschreitend tätigen Unternehmen erhebliche Erleichterung, weil dann die Gewinne nicht mehr nach einer Vielzahl unterschiedlicher nationaler, sondern nur nach einer gemeinsamen Bemessungsgrundlage zu ermitteln sind. Bereits damit wären steuerliche Hemmnisse für grenzüberschreitend tätige Unternehmen reduziert, die Befolgungs- und Bürokratiekosten erheblich verringert; denn statt einer Vielzahl von unterschiedlichen steuerlichen Gewinnermittlungssystemen wäre nur mehr die gemeinsame Bemessungsgrundlage zu beachten. Die Europäische Kommission steckt sich dennoch das Ziel, eine konsolidierte Bemessungsgrundlage – optimalerweise – schon von Beginn weg einzuführen.3 Dies ist zweifellos ein ehrgeiziges Ziel. In Hinblick auf die damit verbundenen Fragestellungen und Probleme spricht sehr viel dafür, zunächst nur eine gemeinsame Bemessungsgrundlage (ohne Konsolidierung) zu schaffen. Denn nicht nur eine sachgerechte Konsolidierung wirft schwierige Fragen auf,4 sondern insbesondere auch die Aufteilung des konsolidierten Ergebnisses auf die einzelnen Mitgliedstaaten. Dazu müsste ein sachgerechter Aufteilungsmechanismus gefunden werden (anknüpfend an Größen wie zB Lohnkosten und/oder Umsätze), der zudem ökonomisch unerwünschte Verhaltensweisen ausschließt (wie die gezielte Verlagerung von Geschäftszweigen). Im Ergebnis wird wohl kein Mitgliedstaat Aufkommensverluste hinnehmen wollen und „Meinungsdifferenzen“ scheinen vorprogrammiert. Auf die komplexen administrativen Fragestellungen im Zusammenhang mit einer grenzüberschreitenden Konsolidierung kann hier nur verwiesen werden.5 Dies alles spricht dafür, zunächst auf die Konsolidierung zu verzichten. Die gemeinsame Bemessungsgrundlage selbst sollte von ihrer grundsätzlichen Ausrichtung effizient, transparent, bestimmt und – wie die Kommission betont – „einfach“6 sein. „Einheitlichkeit und Vereinfachung“ sind nach der Kommission die „Leitprinzipien“ der gemeinsamen Bemessungsgrundlage.7 Das Leitprinzip der Vereinfachung muss aber auch im Lichte einer „sachgerechten Besteuerung“ gesehen werden. Zwischen Sachgerechtigkeit und Vereinfachung muss eine Relation gefunden werden. Vereinfachung alleine ist zu wenig.

3 4 5 6 7

Mitteilung der Kommission vom 2.5.2007, Kom(2007),223 endgültig. Auch zB ihr Verhältnis zu den OECD-Grundsätzen. Siehe dazu den Beitrag von Czakert. Vgl. Anhang 2 des Fortschrittsberichts vom 5.4.2006, 15 f. Mitteilung der Kommission vom 2.5.2007, Kom(2007),223 endgültig.

C. Taxable Corporate Income – Common Structural Elements

87

3. Ausgangssituation Eine gemeinsame Bemessungsgrundlage erfordert es, die Mitgliedstaaten auf einen „gemeinsamen Nenner“ zu bringen. Dieses Unterfangen ist nicht ganz einfach, muss sich doch jeder Mitgliedstaat von seinen nationalen und vertrauten Gewinnermittlungsvorschriften (etwas) lösen – wie auch die Kommission betont8. Dieses notwendige Loslösen und Akzeptieren einer gemeinsamen Bemessungsgrundlage ist der erste wichtige Schritt für die Mitgliedstaaten. Ausgangspunkt der fachlichen Diskussionen zur Ermittlung der gemeinsamen Bemessungsgrundlage sind die IAS/IFRS. Dies hat sich deshalb als zweckmäßig erwiesen, weil mit den IAS/IFRS bereits ein internationales Regelwerk vorliegt, das es den Vertretern der einzelnen Mitgliedsstaaten ermöglicht, eine einheitliche bilanzrechtliche Sprache zu sprechen und mit Fachbegriffen dasselbe zu bezeichnen. Deshalb hat man sich auch sogleich von den einzelnen nationalen Handelsoder Zivilrechten gelöst, weil diese in den Mitgliedstaaten nicht nur inhaltlich, sondern auch terminologisch unterschiedlich ausgestaltet sind. Die IAS/IFRS sind aber für steuerliche Zwecke zu adaptieren. Denn nicht nur die Zielsetzung einer Steuerbilanz (Besteuerung nach der „Leistungsfähigkeit“) weicht von der Informationsfunktion der internationalen Rechnungslegungsstandards ab, sondern eine zB formelle Anknüpfung an die IAS/IFRS brächte das Problem mit sich, dass dann die steuerliche Bemessungsgrundlage von privaten Standard-Settern abhängen würde. Eine solche „Delegation“ an private Standard-Setter ist abzulehnen.9 Eine formelle Anknüpfung an die IAS/IFRS erfolgt daher nicht, die IAS/IFRS bilden den Ausgangspunkt für die gemeinsame Bemessungsgrundlage und von den IAS/IFRS wird immer dann abgewichen, wenn die Zielsetzung einer Steuerbilanz ein Abweichen erfordert.

4. Besteuerungsgrundsätze 4.1 Realisationsprinzip Die gemeinsame Bemessungsgrundlage soll sich an einer Besteuerung nach der wirtschaftlichen Leistungsfähigkeit ausrichten. Ausfluss einer Besteuerung nach der wirtschaftlichen Leistungsfähigkeit ist das Realisationsprinzip. Das Realisationsprinzip soll auch zu den tragenden Grundsätzen der einheitlichen Bemessungsgrundlage gehören. Damit unterliegen nur am Markt realisierte Gewinne der Besteuerung;10 noch nicht realisierte Wertsteigerungen im Unternehmen sollen noch 8 9

10

vgl. Mitteilung der Kommission vom 2.5.2007, Kom(2007),223 endgültig (Punkt 3). In Deutschland und Österreich wurde in Hinblick auf die Maßgeblichkeit der Handelsbilanz für die Steuerbilanz darauf hingewiesen, dass eine (nationale) Zulässigkeit von IAS/IFRS für den Einzelabschluss das Ende der Maßgeblichkeit bedeuten würde, weil es verfassungsrechtlich nicht zulässig wäre, die inhaltliche Festlegung des Besteuerungsumfanges letztlich privaten Gremien zu übertragen; vgl. dazu Zorn, in Bertl/Egger/Gassner/Lang/Nowotny, Die Maßgeblichkeit der handelsrechtlichen Gewinnermittlung für das Steuerrecht, Wien 2003, 179 (195 f.); Hennrichs, in Ebling, Besteuerung von Einkommen, DStJG 24, 301 (306). Vgl. dazu zB Tipke/Lang, Steuerrecht16, § 4, Tz. 13 ff. und 81 ff.; Mayr, Gewinnrealisierung, Wien 2001, 2 ff.

88

Gunter Mayr

keiner Besteuerung zugeführt werden. Eine „fair value“-Bewertung von Wirtschaftsgütern hat nicht zu erfolgen, weil eine Besteuerung von nicht realisierten Gewinnen – von bloßen Gewinnerwartungen – nicht der (aktuellen) Leistungsfähigkeit von Unternehmen entspricht. Gewinnerwartungen haben sich noch nicht realisiert, sie sind noch nicht sicher und am Markt noch nicht bestätigt. Zudem stehen einem Unternehmen erst bei Realisierung die für die Steuerlast notwendigen Mittel zur Verfügung, zuvor fehlt es an der notwendigen „Zahlungsfähigkeit“ (Liquidität). Das Realisationsprinzip verbindet folgerichtig den steuerlichen Markteinkommensbegriff (nur das am „Markt“ realisierte Einkommen unterliegt der Besteuerung) mit der Gewinnermittlung. Der Zeitpunkt der Gewinnrealisierung tritt mit dem Umsatzakt – also mit der Leistungserbringung – ein. Mit Leistungserbringung hat der zur Leistung (bzw Lieferung) Verpflichtete alles getan, sodass sein Anspruch auf Gegenleistung sicher ist. Das Realisationsprinzip zählt zu den wichtigsten Grundsätzen einer gemeinsamen Bemessungsgrundlage. Als Ausfluss des Realisationsprinzips sind zB die einzelnen Wirtschaftsgüter höchstens zu den Anschaffungs- oder Herstellungskosten zu bewerten. Die bloße unrealisierte Wertsteigerung im Betrieb genügt für eine Besteuerung nicht. Kalkulatorische Kosten wie zB eine kalkulatorische Miete oder ein kalkulatorischer Unternehmerlohn gehören nicht zu den steuerlichen Anschaffungs- oder Herstellungskosten.

4.2 Imparitätsprinzip Unterschiedliche Auffassungen bestehen beim Imparitätsprinzip. Es geht dabei um die Frage, ob unter bestimmten Umständen auch ein noch nicht realisierter Verlust steuerlich berücksichtigt werden soll. Ein Teil der Mitgliedstaaten vertritt eine „symmetrische Behandlung“ von Gewinnen und Verlusten und lehnt daher eine Verlustantizipation ab. Auch wenn eine symmetrische Behandlung von Gewinnen und Verlusten vom „logischen“ Standpunkt viel für sich hat, würden sich ergebnisbezogen dadurch Probleme ergeben. Die bilanztechnischen Instrumente einer Verlustantizipation sind Teilwertabschreibungen und Verlustrückstellungen. Lehnt man Teilwertabschreibungen ab, würde das dazu führen, dass zB unverkäufliche Handelsware („Ladenhüter“), „schlechte“ (uneinbringliche) Forderungen oder Beteiligungen an in eine Krise geratenen Unternehmen zu ihren Anschaffungs- oder Herstellungskosten fortgeführt werden müssten; erst der Abgang (die Veräußerung) brächte für das Unternehmen den Verlust. Es stellt sich die Frage, ob eine solche Besteuerung verhältnismäßig ist.11 Denn die Unternehmen wären gezwungen, Wirtschaftsgüter, deren Teilwerte gesunken sind, zu veräußern, um den Veräußerungsverlust (bzw. die Wertminderung) lukrieren zu können. Unverkäufliche Ware unverändert mit den Anschaffungskosten zu bilanzieren, erzeugt bei einigen Mitgliedstaaten ein bilanzrechtliches Unbehagen. Daher würde der andere Teil der Mitgliedstaaten unter bestimmten Umständen auch Teilwertabschreibungen und Verlustrückstellungen 11

Dazu zB Mayr, Rückstellungen, Wien 2004, 219 ff.

C. Taxable Corporate Income – Common Structural Elements

89

zulassen. Für diese Sichtweise sprechen gute Gründe; es muss dabei aber sichergestellt sein, dass der drohende Verlust ausreichend konkretisiert und wahrscheinlich ist.

4.3 Zurechnung von Wirtschaftsgütern und Leasing Bei der Frage der Zurechnung von Wirtschaftsgütern drehte sich die Diskussion um das wirtschaftliche und zivilrechtliche Eigentum. Die Mehrheit der Mitgliedstaaten knüpft die Zurechnung an das zivilrechtliche Eigentum, wobei in Ausnahmefälle die Zurechnung auch bei einem anderen – dem wirtschaftlichen Eigentümer – erfolgen kann. Im Ergebnis lagen die Mitgliedstaaten bei der Zurechnung nicht allzu weit auseinander. Eine generelle Anknüpfung an das zivilrechtliche Eigentum bei der gemeinsamen Bemessungsgrundlage brächte aber das Problem mit sich, dass die nationalen Zivilrechte in den Mitgliedstaaten nicht einheitlich ausgestaltet sind und sich dadurch bei grenzüberschreitenden Tätigkeiten bei der Zurechnung Unklarheiten ergeben könnten. Deshalb verständigte man sich auf folgende Definition: „Ein Vermögenswert ist dem Unternehmen zuzuordnen, welches die ausschließlichen Rechte und Pflichten zum Besitz, zum Gebrauch, zur Übertragung, zur Nutzung der Früchte und das Risiko der Beschädigung oder des Untergangs des Vermögenswertes zu tragen hat.“ Die Zurechnung von Wirtschaftsgütern wurde insbesondere auch im Zusammenhang mit Leasing diskutiert, weil hier in einigen Mitgliedstaaten Sonderzurechnungsregelungen bestehen. Ausgehend von der allgemeinen Zurechnungsdefinition soll auch beim Leasing die Zurechnung zu demjenigen erfolgen, bei dem die Chancen und Risken liegen, wobei man sich an IAS 17 orientieren möchte. Die Kriterien von IAS 17.10 für die Zurechnung zum Leasingnehmer erweisen sich als ausgezeichnete Diskussionsgrundlage, wobei die Diskussion noch fortzusetzen ist.

4.4 Einzel- oder Gruppenabschreibung? Hinsichtlich der Abschreibungsmethode bestehen noch unterschiedliche Auffassungen. Ein Teil der Mitgliedstaaten tritt für eine Einzelabschreibung ein, bei der allen Wirtschaftsgütern (zB anhand von Abschreibungstabellen) eine individuelle Nutzungsdauer zugewiesen wird. Der andere Teil zieht eine Gruppenabschreibung („Pooling-Methode“) vor, bei der die Wirtschaftsgüter einer Gruppe zugeordnet werden. Bei der Gruppenabschreibung werden verschiedene Wirtschaftsgüter zusammengefasst und mit demselben Satz abgeschrieben. Die Gruppenabschreibung (Zusammenfassung von Wirtschaftsgütern) ist für sich zweifellos einfacher, führt allerdings auch zu einer starken Typisierung. Auf Grund der Vereinfachung präferiert die Kommission offenkundig eine Gruppenabschreibung mit mehreren Abschreibungsgruppen.12 Mehrere (zB 4-5) Ab12

Vgl. Anhang 2 des Fortschrittsberichts vom 5.4.2006, 16; nach der Mitteilung vom 2.5.2007, Kom(2007),223 endgültig, Anhang 2, ist diese Frage für die Kommission noch offen.

90

Gunter Mayr

schreibungsgruppen könnten tatsächlich einen für alle Mitgliedstaaten vertretbaren Kompromiss darstellen. Überlegenswert wäre auch, Wirtschaftsgüter mit einer Nutzungsdauer von bis zu zehn Jahren in (2-3) Abschreibungsgruppen zusammenzufassen, bei Wirtschaftsgütern mit einer Nutzungsdauer von über zehn Jahren (zB Gebäuden) hingegen eine Einzelabschreibung vorzusehen. Das Hauptargument für eine Gruppenabschreibung liegt in ihrer Einfachheit. Allerdings sollte noch näher untersucht werden, ob eine Gruppenabschreibung tatsächlich eine Vereinfachung bringt. Denn in fast allen Mitgliedstaaten müssen die Unternehmen bereits nach nationalem Handelsrecht oder nach IAS/IFRS eine Einzelabschreibung durchführen, sodass eine steuerliche Gruppenabschreibung gerade nicht zu einer Vereinfachung, sondern zu höherem Verwaltungsaufwand führen würde. Zudem scheint auch noch nicht geklärt, ob und wie Teilwertabschreibungen im Falle von Abschreibungsgruppen vorzunehmen sind.

4.5 Rückstellungen Die Bildung von Verbindlichkeitsrückstellungen sollte grundsätzlich anerkannt werden. Bei Verbindlichkeitsrückstellungen handelt es sich um dem Grunde und/ oder der Höhe nach ungewisse Verbindlichkeiten. Die Mehrheit der Mitgliedstaaten und die Kommission präferieren eine allgemeine Definition von Rückstellungen, steuerlich nicht zu berücksichtigende Rückstellungen würden dann in einer „Negativliste“ aufgelistet werden. Auch Rückstellungen, denen keine Außenverpflichtung zugrunde liegt (Aufwandsrückstellungen), dürfen nicht gebildet werden. Den für die Bildung einer Rückstellung notwendigen Grad der Gewissheit („Wahrscheinlichkeit der Inanspruchnahme“) legen die Mitgliedstaaten derzeit recht unterschiedlich fest, was insbesondere die Diskussion bei Garantierückstellungen gezeigt hat. Da die Mitgliedstaaten überein gekommen sind, dass sich die Rückstellungsdefinition an IAS 37 ausrichten sollte, würde es sich anbieten, das Wahrscheinlichkeitskriterium von IAS 37 zu übernehmen. Der Eintritt einer Verpflichtung gilt nach IAS 37.23 dann als „wahrscheinlich“, wenn der Ressourcenabfluss mit einer Wahrscheinlichkeit von über 50% eintritt („wenn mehr dafür als dagegen spricht“). Da die IAS/IFRS ganz grundsätzlich der Ausgangspunkt für die gemeinsame Bemessungsgrundlage sind und von den IAS/IFRS dann abgewichen wird, wenn sie nicht den steuerlichen Zielen entsprechen, bietet es sich beim Wahrscheinlichkeitskriterium an, an die IAS/IFRS anzuknüpfen. Gewisse praktische Unterschiede bei den Rückstellungen lassen sich in den Mitliedstaaten wohl nicht vermeiden. Denn es besteht Übereinstimmung, dass Rückstellungen auf Grund gesetzlicher Verpflichtungen in einem Mitgliedstaat anzuerkennen sind. Da zB die Umweltstandards in den einzelnen Mitgliedstaaten voneinander abweichen, könnte es in der Praxis vorkommen, dass für dieselbe Aktivität („Umweltbeeinträchtigung“) in einem Mitgliedstaat mit hohen Umweltstandards eine Rückstellung zu bilden ist, in einem anderen Mitgliedstaat hingegen nicht. Für die gemeinsame Bemessungsgrundlage stellt das an sich kein Problem dar; im Falle einer Konsolidierung wäre damit eine gewisse Verzerrung (Quer-

C. Taxable Corporate Income – Common Structural Elements

91

subventionierung) verbunden, was auch gegen eine Konsolidierung sprechen könnte.13 Die Frage der Zulässigkeit von Verlustrückstellungen (Rückstellungen für drohende Verluste aus schwebenden Geschäften) hängt mit dem Imparitätsprinzip und mit Teilwertabschreibungen zusammen. Befürwortet man Teilwertabschreibungen, muss man konsequenterweise auch für Verlustrückstellungen eintreten, weil beide Instrumente einen noch unrealisierten Verlust vorwegnehmen und gleichermaßen dem Imparitätsprinzip folgen. Hat zB ein Unternehmen Handelswaren bestellt, deren Marktwert sodann verfallen ist, macht es für das Unternehmen wirtschaftlich betrachtet keinen Unterschied, ob das (Beschaffungs)Geschäft noch schwebt (Verlustrückstellung) oder die Ware bereits auf Lager liegt (Teilwertabschreibung). Teilwertabschreibung und Verlustrückstellung sind konsequenterweise gleich zu behandeln.

4.6 Betriebsausgaben Bei den abzugsfähigen Betriebsausgaben besteht unter den Mitgliedstaaten hinsichtlich der grundsätzlichen Vorgehensweise weitgehend Übereinstimmung: Grundsätzlich mindern alle betrieblich veranlassten Aufwendungen als Betriebsausgaben den Gewinn („Veranlassungsprinzip“, Kausalität). Nicht abzugsfähige Aufwendungen sollen in einer Liste angeführt werden. Bei Geld- und Vertragsstrafen, bei Bußgeldern, bei Luxusaufwendungen und bestimmten Repräsentationsaufwendungen tendiert die Mehrheit dazu, sie den nicht abzugsfähigen Aufwendungen zuzuordnen. Auch zB die Körperschaftsteuer sollte (als „Personensteuer“) nicht abzugsfähig sein. Bei Transaktionen unter Nahestehenden wird eine Angemessenheitsprüfung im Sinne eines Fremdvergleiches angedacht.

4.7 Eigene Steuerbilanz? Ob die gemeinsame Bemessungsgrundlage im Ergebnis auch zu einer eigenen Steuerbilanz führt, hängt wohl auch davon ab, wie sehr die gemeinsame Bemessungsgrundlage von IAS/IFRS abweicht. Derzeit werden in vielen Mitgliedstaaten keine eigenständigen Steuerbilanzen, sondern „Mehr-Weniger-Rechnungen“ von den Handelsbilanzen gemacht. Eine von Handelsbilanzen abgeleitete „Mehr-Weniger-Rechnung“ mag auf den ersten Blick zwar praktikabel erscheinen, würde aber dazu führen, dass man sich bei grenzüberschreitenden Tätigkeiten auch mit den zugrunde liegenden nationalen Handelsrechten auseinandersetzen müsste. Dies erscheint weder praktikabel noch mit dem Sinn und Zweck einer gemeinsamen Bemessungsgrundlage vereinbar. Eine „Mehr-Weniger-Rechnung“ auf Basis der IAS/IFRS würde das Problem in sich tragen, dass derzeit in der EU viele Körperschaften keinen IAS/IFRSAbschluss erstellen; so ist zB in Deutschland, Frankreich, Spanien oder Österreich 13

Dieselbe Frage stellt sich bei der Abzugsfähigkeit von allen „spezifisch inländischen“ Sachverhalten wie zB Sozialbeiträgen oder Steuern (ausgenommen der Körperschaftsteuer); im Falle einer Konsolidierung wäre dies wohl immer mit gewissen Verzerrungen verbunden.

92

Gunter Mayr

der Einzelabschluss nach nationalem Handelsrecht zu erstellen.14 Viele Unternehmen müssten sodann, ohne handelsrechtlich verpflichtet zu sein, einen IAS/IFRSAbschluss erstellen. Dies liefe ebenfalls dem Vereinfachungsgedanken der gemeinsamen Bemessungsgrundlage zuwider. Zudem geht bei „Mehr-Weniger-Rechnungen“ oftmals auf Grund der erforderlichen Nebenrechnungen die Übersichtlichkeit verloren. Auch die von der Kommission offenkundig präferierte Gewinn- und Verlustrechnung würde mE umfangreiche Nebenrechnungen erfordern (zB im Bereich der Bewertung, Abschreibung, Wertaufholung, Rückstellungen); eine Bilanzmethode beinhaltet bereits diese Nebenrechnung und ist übersichtlicher. Gerade bei grenzüberschreitender Tätigkeit führt wohl nur eine eigenständige Steuerbilanz zu verständlichen, übersichtlichen und überprüfbaren Ergebnissen.

5. Ergebnis und Ausblick Die Arbeit der Mitgliedstaaten in den technischen Unterarbeitsgruppen zeigt, dass man bei den Strukturelementen der gemeinsamen Bemessungsgrundlage schon sehr weit gekommen ist, grundsätzliche Fragen konnten ebenso beantwortet werden wie wichtige Detailfragen. Auch wenn noch einige Detailfragen offen sind, zeichnet sich ab, dass letztlich ein geschlossenes, stimmiges und für die Mitgliedstaaten akzeptables sowie attraktives Gewinnermittlungssystem vorgelegt werden kann. Die gemeinsame Bemessungsgrundlage sollte nicht nur für Großkonzerne neue Möglichkeiten und Vereinfachungen bringen, sondern insbesondere auch neue Chancen für kleinere und mittlere Unternehmen bieten, die bereits grenzüberschreitend tätig sind, dies künftig sein wollen oder in Grenzregionen angesiedelt sind. Die gemeinsame Bemessungsgrundlage ist eine Chance für den Binnenmarkt und für Europa, sie soll nicht bürokratisch, sondern möglichst einfach ausgestaltet sein und daher Vorteile für alle Unternehmen bieten. Es wird wohl davon auszugehen sein, dass eine gemeinsame Körperschaftsteuer-Bemessungsgrundlage auch auf die nationalen Einkommensteuer-Bemessungsgrundlagen ausstrahlen wird, weil zB die Abschreibungsregeln für Körperschaften und Einzelunternehmen bzw Personengesellschaften nicht grundsätzlich voneinander abweichen können. Das Funktionieren und die Vorteile der gemeinsamen Bemessungsgrundlage hängen auch noch von grundsätzlichen Entscheidungen ab. So wird derzeit diskutiert, ob die gemeinsame Bemessungsgrundlage für Unternehmen optional oder verpflichtend eingeführt werden soll. Die Kommission tritt für Optionalität ein,15 um 14

15

Art 5 IAS-Verordnung, EG 1606/2002, räumt den Mitgliedstaaten ein Wahlrecht zur Anwendung von IAS/IFRS auch für den Einzelabschluss ein, die erwähnten Mitgliedstaaten haben von diesem Wahlrecht aber nicht Gebrauch gemacht, in Deutschland kann zB ein IAS/IFRSAbschluss nur zu Informationszwecken aufgestellt werden. Der unlängst präsentierte Entwurf für IFRS für Klein- und Mittelbetriebe wird in Deutschland und Österreich bereits vom Ansatz her kritisiert; so erscheinen IAS/IFRS für den Mittelstand zu aufwendig und machen auch wenig Sinn; vgl. dazu zB Niehus, DB 2006, 2529; Baetge, BB 2005, Heft 30, Die Erste Seite; Kalina-Kerschbaum/Steggewentz, DStR 2007, 215; Kahle/Dahlke, DStR 2007, 313. Mitteilung der Kommission vom 2.5.2007; Kom(2007),223 endgültig, (Seite 7).

C. Taxable Corporate Income – Common Structural Elements

93

offenkundig eine größere Akzeptanz für die CCCTB in den Mitgliedstaaten und in der Wirtschaft zu finden. Ob Optionalität der richtige Weg ist, muss allerdings bezweifelt werden. In der Mitteilung der Kommission vom 2.5.2007 finden sich als „Leitprinzipien“ für die CCCTB „Einheitlichkeit und Vereinfachung“. Diese beiden Begriffe umschreiben den Grundgedanken der CCCTB ausgezeichnet. Optionalität passt jedoch nicht zu den Leitprinzipien der Kommission; denn Optionalität kann nicht zu einer Vereinfachung führen. Zudem ist interessant und der Kommission beizupflichten, dass die Bemessungsgrundlage „eher breiter als schmaler“ mit „weniger steuerlichen Anreizen und Ausnahmen“ ausgestaltet wird. Fraglich ist nur, ob eine solche breite gemeinsame Bemessungsgrundlage im Wettbewerb mit den nationalen Bemessungsgrundlagen bestehen kann, wenn die gemeinsame Bemessungsgrundlage nur optional sein soll. Denn bei Optionalität würde ein solcher Wettbewerb bestehen bleiben. Abgesehen davon erscheint es nicht zweckmäßig und mit dem Grundgedanken der CCCTB vereinbar, den steuerlichen Wettbewerb der nationalen Systeme fortzusetzen.16 Eine optionale Lösung lässt sich mit dem Sinn und Zweck einer gemeinsamen Bemessungsgrundlage und den Leitprinzipien „Einheitlichkeit und Vereinfachung“ kaum vereinbaren; Optionalität bringt gerade nicht die angepeilte Vereinfachung, sondern führt – im Gegenteil – zur Verkomplizierung: Je mehr Bemessungsgrundlagen zur Wahl stehen, desto komplizierter und verwaltungsaufwendiger wird es für die Unternehmen. Es liegt auf der Hand, dass Unternehmen Vergleichsrechnungen anstellen lassen, um zu ermitteln, welche Bemessungsgrundlage für sie steuerlich am günstigsten ist. Insbesondere für kleinere Unternehmen kann dies eine erhebliche finanzielle Belastung darstellen, die sie von einer grenzüberschreitenden Tätigkeit abhält. Mit den Leitprinzipien der CCCTB „Einheitlichkeit und Vereinfachung“ lässt sich Optionalität nicht vereinbaren.

16

So aber zB Krister Andersson im folgenden Beitrag.

94

Krister Andersson

II. An optional and competitive CCCTB – a comprehensive approach to reach the Lisbon objectives – Krister Andersson

Thank you very much for giving me the opportunity to discuss the Common Consolidated Corporate Tax Base at this meeting in Berlin. We have heard this morning many reasons why we embark on this journey of designing a new corporate tax base for Europe. I think it is important to keep in mind that the present corporate tax systems of Europe have evolved over a hundred years of time and were designed in a completely different setting than the one we have today. The European economies, as well as jurisdictions outside Europe, were to a large extent closed economies and the requirements on the corporate tax system were very different. Nevertheless, I think it is important to consider the history when we assess what we are trying to accomplish now and what the factors are that need to be changed to achieve our objectives.

1. Aims of the CCCTB We heard from the German Finance Minister and the European Commissioner for Taxation today about the need to reform the European economy and to achieve the objectives agreed upon in the so called Lisbon agenda namely to become the most competitive and dynamic knowledge-based economy in the world. This has equally been underlined in statements by governments in Member States of the European Union. One vital element of this agenda is of course to have a level playing field and to extract the full benefit of the Internal Market. Here tax obstacles are a considerable problem, particularly regarding conflicting tax claims arising from the present 27 corporate tax systems in Europe. Europe’s future tax policy must reinforce EU economic, innovations, health and consumer protection, sustainable development, environmental and energy policies, and support the modernization of the European Social Model. There is also the objective of a durable reduction in the overall tax burden in the Union. Tax obstacles hampering growth and employment must be eliminated or at least substantially reduced. If we return to the problems addressed in the lecture by professor Spengel, the European Commission embarked on the two-folded agenda with short term measures and more long term comprehensive solutions to these problems. Some of the most serious problems are the costs associated with transfer pricing, exit taxation, lack of cross border loss consolidation and CFC regimes. These problem areas materialize in numerous cases of double taxation within the European Union and of course also in relation to Third Countries. The design of the present 27 tax systems is to a large extent not taking into account the ways businesses operate today when national borders increasingly do not coincide with a competitive business structure. The result is very large administrative costs for businesses and therefore for the wel-

C. Taxable Corporate Income – Common Structural Elements

95

fare of the citizens of Europe. In addition tax uncertainty lead to suboptimal investment levels. Correctly designed, the CCCTB offers great potential in resolving some immediate tax problems currently facing European businesses and would enhance economic efficiency and welfare. Apart from minimizing the risk of double taxation, it would potentially remove transfer pricing problems within the EU, reduce compliance costs, simplify EU corporate restructuring and create a level playing field both within the EU and outside. Such a comprehensive measure would resolve the transfer pricing issues and I would say that this is the main reason why the European business community in general is very positive to this work and to the idea of developing a model for businesses being able to opt for a common consolidated corporate tax base. The costs and the problems of transfer pricing issues are very large and with the administrative costs on top of this, the European business community really sees the need for change and for something new.

2. Underlying tax principles When designing a new tax system, it is useful to go back and see what systems exist and how different tax principles have been applied and adhered to. This is exactly what the European Commission Services has embarked upon by going back to fundamental principles. What are key elements in a sound and equitable, efficient tax system, promoting growth as agreed on in Lisbon? How should it be structured and which key features should be included? One of the key features I would like to stress, and which I think can not be stressed enough, is the principle of net taxation. That is, the profits should be taxed but they should only be taxed once. Income should be taxable and costs should be recognised and fully deductible in a timely manner. Most governments and countries once adhered to that principle but we seem largely to have forgotten the principle of net taxation. Often we treat costs in an ad hoc way; sometimes they are not deductible, only a fraction of the total costs is deductible depending on circumstances, or other factors. Often losses may not be offset against profits. This creates severe distortion in the European economy. The comprehensive measure of a common consolidated corporate tax base must address these very fundamental questions. Of course, we also find other standard principles, like taxation based upon realization, predictability and transparency. If we look at other principles that are very important, I would like to stress competitiveness and simplicity. It will impact the view on optionality and administrative aspects. I will come back to these issues shortly. Another panel will discuss the “one stop shop”, i.e. the possibility of filing one single tax return in one single location for an entire group. The main objective is to promote economic efficiency and therefore the common consolidated corporate tax base should be based on capital import neutrality. That is a key feature of the system. Using the terminology of professor Spengel, the CCCTB should to a very large extent be based on the source principle rather than the

96

Krister Andersson

residence principle. Capital import neutrality basically will result in the same tax burden irrespective of the location of the owner of the capital, or of the source for financing if it is within the European Union. This is a principle we typically adhered to when we designed our national tax systems. Most countries apply the same tax rule for domestic companies irrespectively of where in the country the owner is located. As a matter of fact, many tax reforms have tried to enhance equal treatment between different sources of finances and investor types. There can still be local differences but on the national level taxes are typically designed based on capital import neutrality within our national borders. Now we need to extent that framework to the new, larger market – the new internal market of Europe. It sometimes involves rather drastic and even painful changes of the tax structure in national tax regimes. In this respect, I may add that a CCCTB based on capital import neutrality is fundamentally different from the concept of the home state taxation (HST). There, one would basically extend the national systems beyond national borders. There would still be a national capital export neutrality concept as the overriding tax principle. Such a concept inevitably creates distortions and unequal treatment in the single market, based on national residence. The principle in the CCCTB must be capital import neutrality and this will have implications for how foreign source income should be taxed. This will also involve many other aspects of the design of the system. This panel addresses the design of taxable income, but I think it is important to start with the principles. I would then like to come back to some of the other issues.

3. Important features of the CCCTB Let me say something about optionality. From a business perspective, it is very important for the support of the CCCTB that the system is optional. As the Member States inherently can choose to join the regime or not, optionality in this respect refers to the possibility for businesses to opt in or out. In accordance with this, the Commission has reiterated that the CCCTB should be optional for business. There are several reasons for this. First, an optional CCCTB would minimize the risk of a non-competitive tax system. If the CCCTB would be compulsory from the outset, there is an apparent risk that much of the needed political pressure to find consensus on truly competitive solutions would disappear as companies would be forced to use the system whether good or bad. Such a development would hardly help Europe at all. If the CCCTB does not prove to be a competitive system relative to the current domestic regimes, it is important that business can stick to current tax rules, despite their transfer pricing problems, lack of consolidation etc. If not, there is a risk that the CCCTB will put Europe in a worse position than today, something that obviously would go against the very objective of the project. Furthermore, a change from domestic treatment to CCCTB taxation will impose significant costs which have to be borne by the business community. In particular for small and medium sized companies, these costs can be burdensome and occasionally outweigh the benefits of the CCCTB. A compulsory CCCTB could prove to

C. Taxable Corporate Income – Common Structural Elements

97

be detrimental in this respect. It is also hard to see the rationale behind forcing really small and medium size enterprises that are acting only on the domestic market into a complete new set of rules to calculate their taxable profit. However, optionality does not include cherry-picking for businesses, including some entities and leaving others outside the CCCTB. There is also a need for continuity and rules may have to be implemented preventing businesses from shifting in and out of the CCCTB every other year. Finally, an optional CCCTB would give room for continued competition between national tax systems, a feature that governments and businesses see as important. As a matter of fact, it would introduce a completely new element of tax competition since firms would not need to relocate to be taxed under an alternative system. The competition between the CCCTB and the national tax systems may be too much for some Finance Ministers. They may not find it appealing to meet with competition from yet another tax system. But if Europe is to have an effective and dynamic economic development, it is crucial for the system to be optional to keep incentives for governments to resolve problems. Let me say something about the need for full intra-group consolidation from the very start as a key factor and a key element in the structure of a common tax base. Here the Commission has reiterated that it should be accompanied with consolidation in order to reap the benefits. Consolidation is a key element in the sense that it is like the glue in the system that keeps it together. If we just have a common base, countries will start to change it and it will evolve over time and we will soon have 27 not 1, new CCCTB-Systems. We will be sitting here in 10 or 15 years time discussing how to change the mess of 54 or even more systems. Therefore, it is important that governments share revenues and that the CCCTB is consolidated from the very beginning. It is also one of the key factors to ensure net taxation, to remove double taxation, transfer pricing problems, and it is, as I mentioned before, a way to make sure that we get only one extra corporate tax system. On the tax rate issue it goes without saying that there should be no harmonization, a matter that is not even discussed very much. Countries need to retain their sovereign right to impose the corporate tax rate they see fit. This would maintain a sound tax competition within the EU and ensure flexible and competitive tax levels vis á vis non-European countries. I think it is also important to know that the tax rate applicable to the fraction of profit allocated to a country does not necessarily have to be the same as the corporate tax rate in the national corporate tax system. The base will be different and countries should of course be allowed to set the rate for the CCCTB just as they are free to set the rate for the national corporate tax system. Let me turn to what the European Commission said in its communication on simplification and uniformity. It must be one system, it has to be uniform and it has to use a comprehensive and exclusive broad tax base contributing to a simpler system. I believe that those objectives are similar to those expressed when national tax systems were contemplated and enacted. Europe now has a good opportunity to turn back to good old principles, in a way that may not be so easy for national governments to follow. It would of course be possible for national governments to undertake similar initiatives as the European Commission has, eliminating transfer pric-

98

Krister Andersson

ing problems and allowing for consolidation and net taxation, but so far we have seen very few, if any, initiatives in this direction. From a business perspective it is essential that the CCCTB will adhere to these fundamental structural elements. The importance of adopting well functioning administrative rules for the CCCTB cannot be overemphasized. Correctly implemented, it will promote simplicity, reduce compliance costs and ensure a streamlined and common application of the CCCTB in all countries. Improperly designed, however, it could lead to increased complexity and compliance costs. For the CCCTB to have any significant impact on the compliance burden, it must allow for a single compliance in a single location. If the system were to require the corporate group to continuously deal with the tax administration in each Member State, the fact that there is a common set of rules would not significantly reduce compliance costs. It is therefore important that the CCCTB be developed around a one-stop-shop approach where a single Member State is responsible for handling all tax affairs of a given group. Notably, this approach would mean that the taxpayer has to file one tax return only in one country for the entire group. Similarly, the approach would mean that the ‘one-stop-shop’ authority is solely responsible for the audit of the books. It must be recognized that under the CCCTB, the computation of the tax base in no longer a local matter with respect to the individual entity. Consequently, the tax audit of a separate company is not to be carried out by the individual local authority in each member state where the group is active. Obviously, the tax administration of the parent state will sometimes lack the capacity needed to carry out the audits in other Member States. Thus, where required, the local authorities may, on the request of the ‘one-stop-shop’ authority, need to offer support in the execution of the tasks at hand. Several ways of structuring such an interaction can be envisaged. To lower compliance costs, however, the crucial point is to allow the taxpayer to have to interact with one administration only. The ‘one-stop-shop’ approach must apply equally to the apportionment of the calculated profit or loss. For obvious reasons, it is not acceptable that an apportionment decision made by the appointed tax authority (i.e. the one in the parent state) can be ignored or overruled by the tax administration of another Member State. This is absolutely crucial if the CCCTB is to resolve the problem of double taxation. The CCCTB must not endorse a system where several tax authorities must be consulted to come to a common agreement. Considering the current experiences in the Transfer Pricing area, such an approach would be time-consuming and often result in double taxation. It therefore appears that the only reasonable approach is to give the ‘one-stop-shop’ authority the power of making an apportionment ruling which is authoritative for the taxpayer (i.e. which the taxpayer can rely on). Any disagreement between the Member States must be resolved at a common and single point of authority. By providing for a ‘one-stop-shop’ regime with one audit in one state, the problem of having different applications of the CCCTB due to different interpretation and approaches in the various Member States would largely be resolved, at least

C. Taxable Corporate Income – Common Structural Elements

99

with respect to the individual entities in the group. All entities in a group would be audited according to the same approach regardless of where they were situated. Indeed, the issue would potentially still be relevant with respect to the relation between different groups. This seems very hard to avoid, however, and can only be mitigated by providing for a clear and simple CCCTB-legislation. It should also be noted that different application of the same tax rules is not a unique problem for the CCCTB. It is currently often the case with respect to different authorities in the same Member State. Regarding uniformity and simplification, I share the views expressed by the European Commission in its communication: “Uniformity and simplification are […] guiding principles of the work to date and the new base should be ‘broad’ rather than ‘narrow’.”1 17

Accordingly, the CCCTB must repalce current 27 parallel systems with one common system which is comprehensive, exclusive (no domestic deviations), broad, and simple. Let me go into some detail on the definition of taxable income. I will illustrate this with an example of depreciation of fixed assets. It has been a topic that has been discussed quite extensively, particularly in Germany but also in other countries. Many countries presently have very detailed depreciation schemes for assets. It certainly adds to complexity and it is time to evaluate and assess these systems. The expected economic life of an asset is the basis used for depreciation purposes. However, between similar individual assets we know that the real, actual life time may vary greatly. A machine may break down long before it is fully depreciated. It may, on the other hand, run much longer than we thought, being economically viable much longer than we had anticipated. Therefore, the ex ante assumption, which is the basis for tax depreciation, is a crude measure of the ex post actual life time of an individual asset. This is a strong argument against having individual depreciation schemes for assets. On the other hand, if you pool the assets, there will be miscalculating in one direction for one asset and in another direction for another asset. On average, the depreciation allowance may reflect actual depreciation more accurately and it may be more precise than if you had very specific assets, each with a very detailed depreciation scheme. It is like owning individual stocks compared to investing in a mutual fund. If we have individual stocks, we all know that they can go up and they can go down. The risk is very large owning just one stock. But if you have a portfolio, a mutual fund of stocks, the risk in that portfolio is considerably lower than for an individual asset. It is the same when it comes to depreciation of fixed assets. Furthermore, when we designed the present depreciation schemes, inflation rates were high in many countries. Hopefully, that is history and price stability is here to stay. The need for detailed depreciation allowances is therefore lower. The most important factor is that the cost is recognized to assure net taxation. The timing is not quite as important anymore since the inflation rate is low. 17

Commission COM(2007) 223.

100

Krister Andersson

If we look to the countries that have used pooling as a method, it has worked very well in a number of countries that uses that method. My own country, Sweden, is one of them. It is a simple system and I think it is therefore reassuring to read in the Commission’s communication of May 2, 2007, that developing the pooling method offers significant advantages. It does not have to be only one pool. Also a system with a few pools for assets could offer important benefits relative to current individualized regimes.

4. Conclusion Let me round off this presentation by concluding that developing this new system by going back to good old principles, rethinking tax elements and timing, is a very welcome step toward achieving a well functioning internal market for Europe. It is a necessary step if we want to gain efficiency and enhance welfare for our citizens. Correctly implemented it will certainly reduce tax obstacles as professor Spengel showed in his presentation. Only a comprehensive measure can resolve all of the issues with transfer pricing, loss compensation and an equitable allocation of profits, and make countries willing to share good times and bad. If it is correctly implemented, it could be a significant step forward in Europe. We don’t know exactly how the system will look. It is too early to tell. The jury is still out and any final call on the CCCTB must wait until a concrete proposal is presented. It looks very promising and the work undertaken so far by the Commission is encouraging. It is a great work done by the European Commission and it gets support from the European business community. When it comes to the outcome it will depend on the actual specification of the system, how simple it is and how far countries are actually willing to rethink their strategy in corporate tax policy. To merit the support from the business community at the end of the day, it is a key factor that the system is optional for businesses; that it is consolidated from the start to insure net taxation and that it is based on the concept of a one-stop-shop.

C. Taxable Corporate Income – Common Structural Elements

101

III. Panel Statements 1. Martina Baumgärtel

Meine sehr verehrten Damen und Herren. Die Besteuerung von Finanzinstrumenten oder Financial Institutions ist natürlich sehr eng verbunden mit der Besteuerung und der Einbeziehung von Financial Institutions in die gemeinsame Bemessungsgrundlage. Die grundlegende Entscheidung ist eigentlich klar: Wenn es denn eine gemeinsame konsolidierte Bemessungsgrundlage gibt, dann muss diese auch Versicherungsunternehmen und Banken, also Financial Institutions mit umfassen. Allerdings ist auch klar, dass hier Besonderheiten wegen der Financial Instruments zu Tage treten und damit die Frage aufgeworfen wird, wie eine sinnvolle Besteuerung aussehen soll. Hier möchte ich ein ganz grundsätzliches Statement abgeben, weil ich aus einem Konzern komme, der sowohl im Versicherungssektor als auch im Bankensektor tätig ist. Man kann diese Frage nicht einheitlich für alle Financial Institutions beantworten. Man muss immer auf die jeweilige Branche, in der man sich befindet, Rücksicht nehmen. Denn die Besonderheiten der jeweiligen Branche sind zu unterschiedlich, als dass ein einheitliches „Set of Rules“ für beide Branchen Anwendung finden könnte. Ausgangsbasis sollte aber immer eine gemeinsame konsolidierte Bemessungsgrundlage sein, wie sie für alle anderen Branchen auch gilt. Branchenbesonderheiten können nur insoweit Platz greifen, als es dafür tatsächlich eine Rechtfertigung gibt. Grundlage sollte aus unserer Sicht immer die Besteuerung nach der Leistungsfähigkeit sein. Dies bedeutet grundsätzliche Beachtung des Realisationsprinzips, das heute bereits mehrfach angesprochen wurde. Und hierzu gehört unverrückbar auch die Besteuerung des Nettoeinkommens, also die Tatsache, dass Betriebsausgaben und Verluste stets abzugsfähig sein müssen. Was heißt das nun konkret für Versicherungen? Bei Versicherungen liegt der Schwerpunkt auf den Passivposten der Bilanz, nämlich auf den Rückstellungen. Und hier ist ganz klar: Wenn man das Realisationsprinzip ernst nimmt, dann kann man die versicherungstechnischen Rückstellungen auf das Realisationsprinzip zurückführen. Das gilt sowohl für die Schadenrückstellungen als auch letztlich für die Schwankungsrückstellungen (equalisation reserves). Auch diese Rückstellungsart, die versicherungsspezifisch ist, dient dazu, Verpflichtungen gegenüber Dritten und nicht gegenüber dem eigenen Unternehmen abzudecken. Sie ist damit keine „Aufwandsrückstellung“. Und daher ist ihr Ansatz dringend notwendig. Das Gleiche gilt im Grundsatz auch für die Großrisikenrückstellungen (catastrophy reserves). Allerdings hat man hier im Gegensatz zu den beiden anderen erwähnten Rückstellungsarten, den Schaden- und Schwankungsrückstellungen, keine Erfahrungen der Vergangenheit und insofern ist das letztlich keine Frage der Rückstellungsbildung dem Grunde nach, sondern eine Frage der Bewertung dieser Rückstellungen.

102

Martina Baumgärtel

Betrachtet man nun die Seite der Banken, dann gewinnt hier natürlich bei der Problematik der Financial Instruments die Bewertung von derivativen Instrumenten besondere Bedeutung, konkret auch die Frage der Anwendung des IAS 39. Und hier ist entscheidend, ob es zu einer fair market Bewertung für Finanzinstrumente kommen muss oder nicht. Meines Erachtens liegt hierin auch ein essenzieller Unterschied zu den Versicherungsunternehmen. Bei Versicherungsunternehmen sind die Investments longterm investments, also langfristige Investments. Dagegen beinhalten bei Banken die Tradingbooks ja in der Regel Positionen, die innerhalb kürzester Zeit, in der Regel bis zu sechs Monaten, umgeschlagen werden. Insoweit kann ich mir durchaus vorstellen, dass ein Ansatz nach dem Realisationsprinzip einer Marktwertbewertung nicht entgegensteht, da die Werte relativ nah liegen dürften. Allerdings gilt dies nur unter der Voraussetzung nicht zu volatiler Kapitalmärkte. Der zweite Punkt, der mit Financial Instruments verbunden ist, ist die Frage der Veräußerungsgewinnbesteuerung und der Dividendenbesteuerung. Und hier, lassen Sie es mich ganz grundsätzlich sagen, gehen wir von einer konsolidierten Bemessungsgrundlage aus. Das heißt, die Frage, wie Dividenden und Veräußerungsgewinne bei Beteiligungen oder Shareholdings zu versteuern sind, stellt sich nur gegenüber Unternehmen, die eben nicht in den Konsolidierungskreis einbezogen werden, konkret Minderheitsbeteiligungen und Streubesitz und generell auch gegenüber Drittstaaten, für welche die CCCTB keine Anwendung findet. Bei Dividenden stellt sich zunächst die grundsätzliche Frage, ob das Anrechnungsverfahren oder die Freistellungsmethode Anwendung findet. Hier kann man durchaus geteilter Meinung sein. Ich meine eigentlich, dass die Freistellungsmethode, die heute in zahlreichen Mitgliedstaaten angewendet wird, die wettbewerbsneutralere Methode ist, da der Wettbewerb der nationalen Steuerrechtsordnungen aufrecht erhalten wird. Angenommen, Sie halten eine Beteiligung an einem USUnternehmen. Sie konkurrieren mit einem US-Unternehmen im US-Markt. Nur die Freistellungsmethode kann letztlich erreichen, dass dieses Unternehmen unter Wettbewerbsgesichtspunkten nicht benachteiligt wird. Allerdings, dass gebe ich zu, kann es natürlich auch zu einer Ausnutzung der Freistellungsmethode kommen und insofern ist gegenüber Drittstaaten eine CFC-Legislation – eine Abwehrgesetzgebung notwendig. Bei Anwendung der Anrechnungsmethode würden sich aber auch im Rahmen der gemeinsamen Bemessungsgrundlage grundsätzliche Probleme technischer Art ergeben, beispielsweise auf welches Steuerniveau angerechnet werden soll, wenn die Steuersätze nicht harmonisiert werden. Wir haben ja dann ein gemeinsames Unternehmen innerhalb Europas mit Aufteilung des Gewinns auf verschiedene Länder. Der Shareholder ist rein zufällig in einem der Mitgliedsstaaten. Auf welche Steuer soll also angerechnet werden und mit welchem Steuerniveau? Diese Komplexität vermeidet die Freistellungsmethode. Was Veräußerungsgewinne anbelangt, handelt es sich letztlich hier auch um Schüttungspotential. Entweder bereits entstandenes Schüttungspotential in Form von offenen Rücklagen oder stillen Reserven oder von künftigen Gewinnerwartungen. Wenn man dieses Konzept verfolgt, dann müssen auch Veräußerungsgewinne aus Beteiligungen steuerfreigestellt werden, um sicher zu gehen, dass Gewinne nur

C. Taxable Corporate Income – Common Structural Elements

103

einer einmaligen Besteuerung unterliegen, um das Konzept der Einmalbesteuerung von Gewinnen durchgängig zu verwirklichen. Bei Veräußerungsverlusten könnte man die Frage stellen, ob das nun auch heißt, dass Veräußerungsverluste bei Beteiligungen nicht abzugsfähig sein sollten. Ich meine nein. Dies hängt aber entscheidend davon ab, wie Verluste steuerlich behandelt werden. Wenn Verluste generell immer beim Unternehmen, das die Verluste erwirtschaftet, vollständig genutzt werden können, und zwar auch nach einem etwaigen Wechsel des Anteilseigners, dann kann man auch darauf verzichten, einen Veräußerungsverlust, wenn die Beteiligung aufgegeben wird, in den Händen des Shareholders steuerlich abzugsfähig zu stellen. Wenn dies aber nicht der Fall ist und es irgendwelche Restriktionen für Verluste gibt, dann muss sicherlich gewährleistet sein, dass sich der Verlust einmal im System, nämlich spätestens beim Anteilseigner auswirkt. Ganz kurz noch zum Betriebsausgabenabzug im Zusammenhang mit Beteiligungen. Die Steuerfreistellung bei Beteiligungen, für die ich mich hier ausspreche, führt zu einer Einmalbesteuerung. Dabei handelt es sich um eine technische Steuerfreistellung (ein „außer Ansatz Lassen“) zur Vermeidung von wirtschaftlicher Doppelbesteuerung und nicht um eine Steuerfreistellung in dem Sinne, dass das Einkommen generell nicht der Besteuerung unterliegt. Damit ist zwangsläufig verbunden, dass auch Ausgaben, die mit der Beteiligung im Zusammenhang stehen, natürlich steuerlich abzugsfähig sein müssen, weil sie ja eben gerade durch die wirtschaftliche Betätigung, dem Halten der Beteiligung verursacht sind. Insofern ist eine Einschränkung des Betriebsausgabenabzugs im Zusammenhang mit Beteiligungen nicht gerechtfertigt, selbst dann nicht, wenn Dividenden und Veräußerungsgewinne steuerfrei sind.

104

Christian Comolet-Tirman

2. Christian Comolet-Tirman

Merci Monsieur le Président, Vous m’invitez à parler des mesures incitatives. De quoi s’agit-il ? De toutes les mesures spécifiques qui visent à favoriser soit certains secteurs, soit certaines zones géographiques, soit certains types d’activités. Ces mesures peuvent prendre la forme d’une déductibilité particulièrement généreuse des dépenses, ou d’une taxation allégée des produits ou encore d’une réduction de la cotisation d’impôt ellemême. Cela peut paraître paradoxal de donner un coup de projecteur particulier sur le sujet des mesures incitatives au sein de l’ACCIS, alors que ces dernières n’occupent, à ce stade, qu’une place limitée dans le projet ACCIS. Mais, précisément, il est intéressant de s’interroger sur le pourquoi d’une telle situation. Pour ce faire, je voudrais rappeler d’abord les objectifs qui ont guidé nos travaux et qui conduisent assez naturellement à ne réserver aux dispositifs incitatifs qu’une place circonscrite, montrer ensuite que cette orientation générale s’est répercutée aussi bien sur les discussions qui ont eu lieu concernant les charges que sur celles relatives aux produits, avant d’évoquer les questions qui restent en suspens et appellent des réflexions complémentaires. Les quatre objectifs qui ont guidé nos discussions conduisent assez naturellement à réserver une place limitée aux dispositifs incitatifs. Le premier objectif, c’est que l’ACCIS doit être suffisamment attractif pour susciter l’adhésion des entreprises. Le deuxième objectif, c’est que le nouveau système doit garantir aux États membres un montant adéquat de recettes et éviter de les faire contribuer au financement des politiques menées par leurs voisins. Cela implique que l’assiette soit relativement large pour préserver pleinement la capacité des États à fixer librement leurs taux en fonction de leurs besoins. Cela implique également qu’un éventuel aménagement des règles communes par un Etat doit rester sans incidence sur les recettes redistribuées aux autres États. Un troisième objectif, c’est que le nouveau système doit être suffisamment simple et uniforme pour améliorer effectivement le fonctionnement du marché intérieur. Cela implique de ne pas recourir systématiquement au principe de la reconnaissance mutuelle. A défaut, on ne parvient pas à une base vraiment commune. Cela implique également de ne pas laisser aux États, après l’application des clefs de répartition, toute latitude pour procéder à des aménagements au niveau national. Le quatrième objectif, qui va de soi mais qu’il est néanmoins bon de rappeler, c’est que le nouveau régime doit être compatible avec la réglementation afférente aux aides d’État et avec les réglementations afférentes au commerce international. Lorsqu’on les combine, on arrive à la conclusion que seule une assiette large, avec des dispositifs incitatifs en nombre limité, permet de satisfaire au mieux ces différents objectifs. Une assiette large est de nature à satisfaire les administrations soucieuses de garantir la couverture des besoins financiers de l’État, d’améliorer la transparence

C. Taxable Corporate Income – Common Structural Elements

105

des conditions de concurrence et d’avoir un système relativement simple à gérer. Une assiette large répond également aux attentes en matière de fonctionnement du marche intérieur, puisqu’elle est « lisible » et limite les distorsions. En apparence, elle est moins favorable aux entreprises, mais ces dernières sont également soucieuses de simplicité et elles peuvent trouver bien d’autres avantages au projet : harmonisation de l’assiette, simplification des modalités déclaratives, imputation des pertes… C’est ce qui conduit la Commission, dans sa communication du mois de mai, à souligner qu’ « il y aura moins d’incitations et d’exceptions » et que « les déductions et incitations spécifiques devront être limitées en nombre, convenablement ciblées et dûment justifiées ». Cette orientation générale est illustrée aussi bien par les travaux ayant trait à la déduction des charges qu’à ceux afférents à la taxation des produits. S’agissant des produits, il est prévu que tous les produits courants réalisés soient taxés, à l’exception des dividendes (pour éviter les doubles impositions). De la même façon, il est prévu que toutes les plus-values réalisées soient imposées comme des revenus ordinaires, à l’exception des plus-values de cession de titres de participation (là encore, pour éviter une potentielle double imposition). S’agissant des charges, les charges courantes engagées dans l’intérêt de l’entreprise ont vocation à être déductibles, à l’exception d’un petit nombre d’entre elles : certains impôts, certaines contributions sociales volontaires, ainsi qu’une quote-part des dépenses engagées pour les besoins de la perception de revenus exonérés ne seraient pas déductibles. Le débat sur les amortissements – qui s’est focalisé sur le choix de la méthode (par composants ou par catégories/« pooling ») – n’a pas conclu à la nécessité d’adopter une durée d’amortissement particulièrement courte : à ce stade, c’est un amortissement sur la durée d’utilité qui est privilégié. De même, un principe général de déductibilité des provisions sera sans doute posé et assorti d’une liste de provisions non déductibles, sans que soit opéré un choix tranché en faveur d’un régime de provisionnement particulièrement avantageux. Au total, et comme je l’indiquais tout à l’heure, la compétitivité de l’ACCIS résultera pour l’essentiel de caractéristiques autres que celles liées à l’importance des mesures incitatives. Plusieurs questions restent néanmoins en suspens. Certaines sont afférentes au traitement de secteurs spécifiques comme cela vient d’être évoqué par Madame Baumgärtel, notamment le secteur financier. Mais, la question principale est celle de la recherche–développement. Dans sa communication de novembre 2006, la Commission a elle-même préconisé un traitement favorable des dépenses de recherche-développement. C’est évidemment souhaitable eu égard aux objectifs de Lisbonne qui on été rappelés tout à l’heure. A ce stade, a été prévue la possibilité de passer en charges les dépenses occasionnées par le développement d’un brevet en interne et d’inscrire à l’actif les dépenses liées à l’acquisition d’un brevet. Mais deux questions au moins sont encore en suspens : le concessionnaire d’un brevet peut-il inscrire à l’actif les dépenses correspondan-

106

2. Christian Comolet-Tirman

tes ; dans les hypothèses d’inscription à l’actif, sur quelle durée doit avoir lieu l’amortissement ? A cet égard, il n’est pas aberrant d’envisager un amortissement sur une durée plus courte que celle des avantages économiques futurs. Trois points en conclusion : l’attractivité d’ACCIS résultera de caractéristiques autres que celle qui aurait consisté à multiplier les mesures incitatives ; les travaux sur ACCIS peuvent contribuer à la réflexion sur la pertinence des dispositifs incitatifs dans les différents États membres ; il subsiste une incertitude liée au caractère optionnel de l’ACCIS évoqué tout à l’heure : ce caractère optionnel n’introduit-il pas un risque de concurrence entre certains impôts sur les sociétés nationaux qui comporteraient beaucoup de mesures incitatives et un régime commun ACCIS qui en comporterait moins ?

C. Taxable Corporate Income – Common Structural Elements

107

3. Peter Essers

Thank you very much Mr. Chairman. I would like to make some comments on what we’ve heard this morning and this afternoon. First, the closer one looks at the different relationships between tax accounting and financial accounting systems in the Member States, the more difficult it proves to be to categorize the different systems. There are no systems with a full linkage between tax and financial accounts. Neither are there systems with no dependence at all between tax accounts and financial accounts. Reality has many colours. The interesting thing is that the more the financial accounting systems grow towards each other under the influence of IAS and IFRS the more the fiscal accounting systems converge. In more or less independency countries like the United Kingdom and the Netherlands you see a movement towards more alignment, whereas in traditional dependency countries like Germany, France, Austria, Spain, Belgium, you see a tendency to introduce more and more autonomous fiscal profit accounting elements. In my opinion, the main question is no longer whether or not linkage should be applied but rather to what extent an autonomous concept of fiscal profits can be influenced by financial accounting rules like IAS and IFRS. The fact that IAS and IFRS exist is an enormous advantage for the development of CCCTB. At the beginning of the process towards the development of CCCTB the European Commission acknowledged this by saying that IFRS should be the starting point for a system of CCCTB. However, my impression is that this has changed. Now, IFRS are no longer starting point for the CCCTB but only a tool to guide and to inform the discussions by providing a common vocabulary. In the last communication of the Commission IFRS is not even mentioned any longer. Looking at the admirable work already done by the CCCTB working group my impression is that the Commission is aiming at a set of rules and standards not directly seeing IFRS as a starting point but leading to international tax accounting standards which seem to be rule based instead of principle based. This might be a dangerous development because then Member States will have many arguments to say no to proposals which might be very detailed. Also, the development of such a set of tax accounting standards might prove to be very difficult and time-consuming. One of the reasons why the Commission and the CCCTB working group might be driving away from IFRS could be the fact that IFRS applies the fair market value concept, which is, as we all know, in conflict with the realisation principle as one of the main tax principles. Nevertheless, I think we should realize that IFRS allows or prescribes fair value accounting only in specific cases, such as investment real estate, hedge accounting and other passive investments. If necessary, we could find separate tax solutions for these kinds of topics. This argument becomes even more relevant if you realize that in certain cases fair value accounting is not necessarily against tax accounting principles. If assets, like securities, shares or obligations, are

108

Peter Essers

directly realizable there is no need to wait for actual realization to take profits fiscally into account. My last point – Mr. Chairman – would be that introducing a legislative proposal on a set of international tax accounting standards already in 2008 might be bit too early because in the Member States autonomous developments of the introduction of IFRS still have to crystallize out, both from a tax perspective and from the perspective of IFRS. In my opinion, the proposals in 2008 should focus mainly on common tax accounting principles. Moreover, they should identify specific areas where in the near future consensus has to be reached because IAS/IFRS does not lead to a solution in these areas. Most of all the proposals should create a format for the further rulemaking process and for future dispute settlements. I don’t think we can have a system which will be there for eternity. What we need is a dynamic concept which can be changed. IFRS have not been made in one, two or three years. So I would like to make a plea for an interactive dynamic concept of fiscal profit, shaped in a continuous dialogue between the different stakeholders in the concept of fiscal profit: business, tax advisors and tax administrations. The concept of fiscal profit fitting within the legal framework of CCCTB must be flexible enough to be adapted to changing insights and circumstances, e.g., in the science of business economics, financial accounting, IAS/IFRS, the introduction of new financial instruments or technical developments. Also, participating companies must have the possibility to opt for several tax accounting systems, all fitting within the accepted CCCTB framework. Finally: although IAS/IFRS are not decisive for tax accounting within the framework of the CCCTB, they should be explicitly acknowledged as a starting point. Thus, a system that is in line with IAS/IFRS should – as a rule – also be regarded as in line with the CCCTB framework, unless this system is against the agreed principles of the CCCTB framework or against a specific rule formulated within the CCCTB legal framework.

C. Taxable Corporate Income – Common Structural Elements

109

4. Malcolm Gammie

Thank you, Mr. Chairman, for your introduction. When you have spent many years considering and talking about a subject, you wonder whether you have anything new to tell the audience. I do not think that I can add to what I have said before. I have spoken and written on this subject many times before and the issues do not change. However, let me put the issue a different way. It is said that a camel is a horse that has been designed by committee. The issue is whether the committee that is working to devise a CCCTB will produce a thoroughbred from the 27 varieties of corporate tax horse that we find in the European Union at present or whether it will produce a camel, and perhaps a monstrous camel with three humps, five legs and two heads. I listened to what Professor Andersson had to say about tax competition and the business demand for the CCCTB to be an optional system. If we want tax competition, then I recommend that we retain 27 different corporate tax systems and do not trouble with a CCCTB. We might like to see the European Court of Justice recover some of the vigour that it has lost over the last couple of years, to put pressure on Member States to ensure that their tax systems comply with the requirements of the EC treaty. With the ECJ’s assistance the Union might eventually have 27 level corporate playing fields. It is likely that each of those playing fields would be a different height, shape and size but each would at least be level and there would be no barriers to moving from one to play on another. Twenty-seven different systems of company tax certainly appear incompatible with a single market. The solution from a single market perspective is quite straight forward. We should have one corporate tax system only. Not all single markets, however, have one system. The United Kingdom does but England, Scotland, Wales and Northern Ireland have no taxing powers to speak of. The American single market on the other hand manages with a single Federal system and a variety of State Systems based on the Federal system. The USA started from a different point to Europe and reached its single market solution in a different way. In particular, the single market came first and corporate taxation followed. By comparison, as Professor Anderson noted, the majority of the 27 systems in the European Union have started from a closed economy model that has been adapted to deal with cross-border transactions. We are familiar with the international features of corporate tax systems: residence, exemption of foreign income or credit for foreign taxes, the attribution of income and arm’s length transfer pricing of cross-border transactions. In short, the single market has to develop against the background of and despite such national tax systems. A single market imposes a number of practical constraints on the form that the corporate tax can take. As the single market integrates it drives tax systems in a particular direction and towards the adoption of particular features. The international features of national tax systems – residence, attribution, credit for foreign tax, arm’s length transfer pricing – are essentially incompatible with a single market. In effect,

110

Malcolm Gammie

the taxation of intra-Community trade must be compatible with the taxation of domestic or national trade. The forces of tax competition and of the single market and the work of the European Court of Justice will all serve to drive tax systems in a common direction if not to one uniform outcome. I do not address the bigger issue that the European Union is not an island. Even with a CCCTB the EU is only one player in a global economy, subject to similar business, economic and practical administrative pressures with longer term implications for the shape and future of corporate taxation. With those considerations in mind, what should we be striving to achieve within the European Union? Different options or choices lie at the heart of the formulation of the CCCTB. To arrive at any agreement on the very basic question of how to compute profits under a CCCTB, we will have to produce uniform answers to a number of questions that arise from the process of computing profits. That is, after all, why we have 27 different corporate tax systems in the first place, because there is no single answer to the many questions that computing profits pose. It is why different States adopt different forms and rates of depreciation, different provisions for liabilities and different rules for recognising income. We should not under-estimate the task of producing a uniform set of rules of common application in 27 member States. It would be no solution for the CCCTB to allow Member States the several options for dealing with particular computational issues, so that in the end there is no common base. The CCCTB must present a uniform solution of common application. The practical outcome of optionality, however, is to ensure that there is no single solution. There will be the common base and there will be national rules: 28 systems rather than twenty-seven. Inevitably, different businesses may require different tax solutions. The obvious distinction is between small and large business. In the United Kingdom we have several different rules for computing unincorporated business profits and corporate profits. Small corporate businesses raise different issues to large corporate businesses. The taxation of small business must be integrated more closely with the personal taxation of their owner-managers than in the case of large business. That aspect of business taxation is difficult to avoid. The focus of the CCCTB is to produce a system for large pan-European businesses and there seems little sense in allowing large business the luxury to choose which suits them better – the CCCTB or national tax systems – or to play one off against the other. Although the option is intended to be irrevocable and will not allow corporate groups freely to move into and out of the CCCTB, nevertheless if member States can agree the form of CCCTB despite the difficulties, there seems little sense in making it an optional system. It suggests perhaps that European business has made optionality an issue because it fears that they will get a camel and not a thoroughbred. If that is the outcome then we should abandon the project. The project will only have value if it produces a sensible proposal and not a unique animal. A sensible proposal should not need to be optional. Even with a non-optional system for European business operating in participating member States, optionality will be unavoidable to some extent. In common with

C. Taxable Corporate Income – Common Structural Elements

111

current directives I assume that CCCTB will apply to some but not to all legal business forms. Accordingly there will be choices as to which business form to adopt. If there is a common base but no consolidation, cross border losses, transfer pricing and the like will be continuing issues and significantly reduce the value of the proposal. If there is consolidation, a decision will be needed on the definition of a group and the requirements that will bring companies within the group; whether all companies within a particular ownership or set of definitions must be brought within the common consolidated system or whether some can opt out. One need only look at the variety of consolidation regimes across Europe and the World to see the scope for planning which entities are in and which are outside the group. Then there is the option of whether the entity or business is within the Union or outside because any solution that produces a CCCTB for Europe must have its borders at the European Union. In today’s World business always has the option to stand outside those borders instead of within it. None of this sounds particularly optimistic in terms of the progress that we can make and how satisfactory it will be. The work on the CCCTB is nevertheless necessary work, if only to decide whether it is possible to design a thoroughbred or whether we will inevitably end up with a camel. If the answer is a camel I hope we will be able to see it as such and recognise that a camel may do European competitiveness little good in a worldwide horse race.

D. Consolidation, Allocation and International Aspects

113

D. Consolidation, Allocation and International Aspects I. Evaluating the Common Consolidated Corporate Tax Base Ulrich Schreiber

1. Introduction The European Commission proposes a common consolidated corporate tax base (CCCTB)1. The first step to a CCCTB is a harmonised tax base which replaces the different tax accounting rules of 27 member states. Given the harmonised tax base, the CCCTB is designed to replace separate entity accounting and the arm’s length standard with consolidation and formula apportionment of a European group’s profit. The Commission does not propose to harmonise the corporate income tax rate. The member states are still free to choose the corporate income tax rate. Thus, the Commission’s proposal rests on the source principle. Prevailing separate entity accounting makes use of market prices and ignores the fact that the group has replaced markets by hierarchy. By contrast, formula apportionment is based on the fact that the group is a single economic unit. Thus, from an economic point of view, formula apportionment seems to be the appropriate method to tax a group. However, it is impossible to allocate the group’s profit according to certain factors that have caused that profit. Consequently, both separate entity accounting and formula apportionment can not be regarded as mechanisms that allocate profits fairly according to the input involved. Profit allocation rules have to be judged on their economic consequences rather than on their economic correctness. This paper discusses source based formula apportionment (section 2) and then proceeds to residence based separate entity accounting (RSA) as an alternative to the European Commission’s proposal (section 3). Section 4 highlights possible ways to a European group taxation scheme. Section 5 concludes.

1

See Commission of the European Communities, Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee, Towards an Internal Market without tax obstacles, Brussels, COM (2001) 582 final; European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG), Progress to date and future plans for the CCCTB, CCCTB\WP\046\doc\en, Brussels, 20 November 2006.

114

Ulrich Schreiber

2. The Commission’s Proposal: Formula Apportionment and Source Based Taxation 2.1 Consolidation and Apportionment The Commission’s proposal of a CCCTB aims at abolishing the separate entity standard. The group’s profit is instead assessed by means of consolidating the profit and loss of the group’s companies and permanent establishments. Once the group’s profit is determined, it is allocated on the basis of an agreed formula to the member states where the group does business (has taxable nexus). The CCCTB tackles some of the most prominent tax obstacles in the common market: – Cross-border loss compensation. – Book-profit shifting. – Profits from intra-group transactions. These obstacles have their common origin in the source based separate entity accounting approach of international group taxation. Intra-group transactions have to be priced according to the arm’s length standard in order to allocate the group’s profit to the member states where the group has established subsidiaries or permanent establishments. Since each subsidiary is taxed as a separate entity, profits from intra-group trading are taxed before ultimate realisation, which does not occur until the group sells the respective goods to a third party. In principle, cross border loss compensation is not possible since each company of the group is a legally distinct part of the group. Moreover, multinational groups experience an incentive to shift profits to low tax member states by manipulating transfer prices or by using intragroup debt. The group might also concentrate external debt in a high tax member state. The CCCTB has the potential to overcome the above mentioned problems of separate entity accounting. Full consolidation eliminates profits from intra-company trading and effects an immediate cross-border loss offset. Book profit shifting via intra-group trading ceases to be and book profit shifting via intra-group financing is not possible because expenses (e.g. interest expenses) und revenues (e.g. interest income) are matched. Formula apportionment substitutes arm’s length prices, which is expected to substantially reduce the compliance and administrative costs of profit allocation within the group2.

2.2 Investment Decisions Since member states tax their share of the group’s profit according to their corporate income tax rate, the CCCTB basically rests on the source principle. Therefore, as under separate entity accounting, taxes still have an impact on investment decisions. The apportionment mechanism determines the average tax rate of a European group, as is illustrated by the following simple example. The example sets aside the defini2

See Mintz, J., Corporate Tax Harmonization in Europe: It’s All About Compliance, International Tax and Public Finance 2004, 221-234.

D. Consolidation, Allocation and International Aspects

115

tion of the group and assumes that profits and losses of the group’s companies are fully consolidated. Moreover, common tax accounting rules and a harmonised apportionment mechanism are assumed. Two companies T and P are located in the member states T and P, own the capital CT and C P , and earn a profit PT and PP respectively. The consolidated profit amounts to PG = PT + PP . The apportionment formula is based on the capital invested. The underlying assumptions may either be that capital invested reflects the group’s value-creating activities in a certain location or that the jurisdiction where the capital is invested has the right to tax the corresponding profits. Then the tax rate of the group amounts to C

C

P T (1) τ G = τ T ⋅ C + τ P ⋅ C G

G

with CG = CT + C P denoting the group’s total capital, and τ T , τ P the corporate income tax rates in member state T and P. The effective tax rate in a single member state depends on the statutory tax rate and the capital ratio. The total taxes to be paid by the group (TG = τ G ⋅ PG ) are influenced by the capital ratio and the consolidation process. Under formula apportionment the corporate income tax is in effect a tax on the respective factors that enter the formula3. As a result, formula apportionment gives room to international tax planning. With tax rate differences in the member states, the tax burden of the group depends on how the group’s profit is apportioned to the member states. Given that the group’s profit is allocated to the member states on the basis of micro factors (such as capital, wages and sales), the group experiences a strong incentive to move factors that enter the formula to low tax member states. Some argue that capital ownership neutrality is far more important than tax neutrality with respect to investment decisions. The argument rests on the observation that a significant part of direct investments are not green field investments but mergers and acquisitions. Capital ownership neutrality is preserved if taxes do not distort capital ownership patterns. If all countries adopt the source principle and exempt foreign profits from taxation, the tax treatment of foreign investment is the same for all potential owners4. Assets are allocated to their most productive owners. Within the Union, however, one can not expect that source based formula apportionment does not distort ownership patterns. The reason is that the formula implies the ratio of profits and the economic factors which enter the formula to be the same for all companies of the group. Thus, the formula allocates profits from profitable companies to less profitable companies. Consequently, the acquisition of a company that becomes part of the group has an impact on the tax burden of the purchaser. If the acquisition reduces the group’s tax burden because more profits are allocated to

3

4

See McLure, C., The State Corporate Income Tax: Lambs in Wolves’ Clothing, in H. Aaron and M. Boslin (eds.), The Economics of State Taxation, 327-346. See Desai, M./Hines, J., Evaluating International Tax Reform, National Tax Journal 2003, 487502.

116

Ulrich Schreiber

the acquired company, the purchaser saves taxes and thus the value of the target company increases. To see this effect in more detail, assume again that only capital enters the formula. The target company T pays taxes of the amount τ T ⋅ PT . The buyer, company PB, pays taxes τ B ⋅ PB . Before the acquisition the two companies pay taxes of the total amount (2) T = τ T ⋅ PT + τ B ⋅ PB since they are taxed separately. After the acquisition of company T according to equation (1) total taxes of the newly-formed group G amount to ⎛

(3) TG = τ G ⋅ PG = ⎜⎜τ T ⋅ ⎝

CT C ⎞ + τ B ⋅ B ⎟⎟ ⋅ PG CG CG ⎠

with C G = CT + C B and PG = PT + PB. The acquisition does not change the total taxes paid and the equation T = TG holds if the profitability of both companies and the group are the same5: (4)

PG PT P = = B CG CT C B

In the case of the purchaser exhibiting a higher profitability than the target company, a part of the purchasing company’s profit is allocated to the target company by the formula. With a lower tax rate in the country of residence of the target company, the group’s tax burden is lower than the tax burden of the separately taxed companies. Given this effect, profitable companies in high tax countries will hold the assets of low taxed companies. As a consequence, formula apportionment distorts investment decisions as well as ownership patterns.

2.3 Tax Revenue Effects Member states will be interested in preserving their tax revenue when the CCCTB is introduced. Since formula apportionment rests on the basic idea to allocate profits to member states where the group carries on its business, member states will most likely restrain from using only macro factors (e.g. GDP or VAT base). However, a formula that rests on micro factors like capital, wages and sales might not fit into the fiscal policy of many member states. According to an empirical study carried out to predict the consequences of a change from the existing system to formula apportionment (on the basis of capital, wages and sales and with cross-border loss compensation), virtually all economies would lose tax revenue6. Most probably, it will be very 5 6

Using (4) in (3) one determines TG = τ T ⋅ (PT C T ) ⋅ C T + τ B ⋅ (PB C B ) ⋅ C B = τ T ⋅ PT + τ B ⋅ PB = T . See Fuest, C./Hemmelgarn,T./Ramb, F., International Tax and Public Finance 2007, 627-648; the data are restricted to German multinationals and the estimation of the tax revenue effects does not take into account the tax planning responses of the companies.

D. Consolidation, Allocation and International Aspects

117

difficult for the member states to agree on a common formula7. The negotiations may lead to a more complicated formula designed not to change the status quo of the corporate income tax allocation within the European Union. Even a complicated formula might not be regarded as sufficient to protect the tax revenue of member states. Formula apportionment creates a close link between economic activity in a member state, as measured by the firm-specific factors that enter the formula, and the profits attributed to the respective member state. Consequently, tax planning will concentrate on the group’s average tax rate. Member states will have to take into account the companies’ behavioural responses to the agreed formula. To give an example, assume that capital enters the formula. For administrative reasons, member states may decide to measure capital on the basis of book values. Most of the group’s valuable intangibles are then not regarded as capital for the purpose of profit allocation. A company that owns intangibles and earns an economic rent faces the incentive to transfer tangible capital to low tax member states in order to increase the part of the group’s profit to be attributed to the respective member state. As a result, the group is able to transfer at least a part of its economic rent to low tax member states. High tax member states might suffer from both the loss in tax revenue and the loss in real economic activity. One might argue that this problem could be overcome by implementing a formula that rests on more than one firm-specific factor or by implementing a formula that also includes macro factors. Tax planning will be more difficult if it has to take into account multiple micro factors. As far as macro factors are used, tax planning ceases to be. However, to the extent the factors are designed to reflect real economic activity (micro factors), it might not be easy to tackle tax planning and to preserve the tax revenue. The formula is e.g. not less exposed to tax planning if wages and sales are added as a second and third micro factor to capital. The CCCTB eliminates the arm’s length standard and the groups are able to price intra-group transactions as they wish8. Consequently, assets could be transferred at book value from one member state to another 9. Member states would have to give up their tax claim on capital gains (the difference between the fair value and the book value of an asset). Of course profits from the use of the assets as well as profits from selling the assets to a third party increase the common tax base. Therefore, a member state that does not tax capital gains in the case of intra-group transactions, exchanges its full tax claim on the respective gains with its share of these gains 7

8

9

See Agúndez-Garcia, A., The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, Taxation Papers, working paper, 2006, 32 and Martens-Weiner, J., Company Tax Reform in the European Union, New York 2006, 47 for an overview of the apportionment mechanism. See European Commission, Common Consolidated Corporate Tax Base Working Group (CCCTB WG), An overview of the main issues that emerged at the third meeting of the subgroup on group taxation (SG5 – 05 & 06 February 2007), CCCTB\WP\053\doc\en, Brussels, 1st March 2007, 7, no. 33. Note that valuable firm-specific assets (like technical or economical know how and good-will) are usually not capitalised for tax purposes.

118

Ulrich Schreiber

according to the formula. The capital gains leave that member state completely untaxed if the group loses its tax nexus in the respective member state upon the transfer of assets. Since the CCCTB basically rests on the elimination of intra-group results, the member states might be inclined to accept assets being moved at book value from one jurisdiction to another, once the CCCTB is introduced. From this perspective, capital gains taxation seems to be only a transitional problem when member states are reluctant to give up their tax claims on capital gains that existed before the introduction of the CCCTB. This is, however, not the case as the scope of the group may change over time. Member states have to agree on common rules regarding capital gains relating to assets of companies that leave or enter the group. Capital gains taxation is necessary in order to avoid international tax planning. Without capital gains taxation, companies that leave the group could channel capital gains to a single member state, and companies that join the group could channel capital gains to the group. In order to prevent that type of tax planning, capital gains have to be taxed on the occasion of a company entering or leaving the group. Capital gains taxation reintroduces the need to assess arm’s length prices for transferred assets. What seems to be necessary to preserve the tax revenue of the member states is burdensome for the companies because restructuring a European group could trigger substantial amounts of capital gains taxes.

2.4 Third Country-Tax Policy The CCCTB will be restricted to the European Union (water’s edge restriction)10. Thus formula apportionment will interfere with separate entity accounting if the group owns subsidiaries in third countries (non-EU countries). The member states follow different tax policies with respect to third countries. Some member states apply the worldwide principle, while others rely on both the source principle and the worldwide principle. Under a CCCTB member states have to make their choice: Income from non-EU sources can be either (i) included in the common tax base or (ii) completely excluded from the common tax base. In case (i), it is necessary to agree on a common tax policy with respect to income from non-EU sources. Case (ii) might be regarded as unsatisfactory from an economic point of view but as administratively appealing. The choice might not be easy. If member states agree on a worldwide concept, there are further choices to make. Suppose member states decide that interest and license payments from non-EU affiliates increase the group’s income. Will they also agree that dividends, capital gains upon the disposal of shares and income from nonEU permanent establishments are to be included in the common tax base? Member states that strictly apply the worldwide principle could probably agree. Member states that rely on source based taxation in this respect and thus exempt foreign divi10

See Agúndez-Garcia, A., The Delineation and Apportionment of an EU Consolidated Tax Base for Multi-Jurisdictional Corporate Income Taxation: A Review of Issues and Options, Taxation Papers, working paper, 2006, 20 for the territorial scope of the consolidated tax base.

D. Consolidation, Allocation and International Aspects

119

dends, capital gains relating to foreign shareholdings and income from foreign permanent establishments would have to alter their foreign tax policy and thus may be reluctant to agree on worldwide group taxation. Given that member states are not in the position to agree on a common tax policy in respect to non-EU countries, it could be regarded as a first important step to restrict European group taxation to the European part of the business. However, there is a risk that the CCCTB could not smoothly function in that case. If the tax base is divided into EU-sourced income and non-EU-sourced income, it is necessary to allocate expenses to non-EU-revenue. Suppose a company belonging to a European group invests in shares of a non-EU corporation and finances the acquisition with debt. If dividends and capital gains are excluded from the common tax base, the same must be true for the interest payments connected with debt financing of the acquisition. The allocation of third country revenues and expenses has an impact on the common tax base and thus member states must agree on common rules with respect to the assignment of third country income. Moreover, if some member states apply worldwide taxation while others rely on the source principle there is an incentive for the European group to invest in low taxed non-EU permanent establishments via a subsidiary located in a member state that exempts foreign profits. Foreign shareholdings could be attributed to companies located in member states that exempt dividends and capital gains. If interest income or license fees from non-EU sources are excluded from the common tax base, the group has an incentive to channel the respective payments to low tax member states. Cherry picking with respect to foreign income undermines the idea of a having a common tax base. The common tax base is, however, not only put at risk when non-EU income is excluded but also because a European group could shift profits and investments to non-EU countries. Since with respect to non-EU countries separate entity accounting and the arm’s length standard apply, the group concentrates its tax planning activities on investment shifting within the Union, whereas vis-à-vis non-EU countries the group is able to engage in both investment and profit shifting. The incentive to shift profits outside the Union depends on the average European corporate income tax rate (equation 1) and the corporate income tax rate in the respective nonEU country. The average EU tax rate decreases when the group shifts EU-investments to low tax member states. Consequently, the incentive to shift profits outside the Union decreases as well. Given this interaction of intra-European tax planning and tax planning with respect to non-EU countries, high tax rates in some member states may have an impact on the tax revenue of low tax member states. A high tax rate of a member state where the group invests increases the average European tax rate that the group faces. If the group therefore decides to shift profits outside the Union, the common European tax base shrinks and low tax member states suffer from the loss in tax revenue as well as high tax member states. This effect might be important if it is more cost-effective to shift profits instead of investments to other countries. A group might shift profits outside the European Union by investing in financial assets in non-EU countries or by financing European investments by intra-group

120

Ulrich Schreiber

loans from outside the Union. In both cases the common tax base decreases to the detriment of all member states. Given the possibility to reduce the common tax base, there is a need to harmonise anti-avoidance tax legislation that deals with third countries. Without harmonisation of the respective tax rules, a European group could concentrate its tax planning activities with respect to non-EU countries in member states with weak anti-avoidance tax legislation. Typical examples are tax deferral by use of low taxed foreign controlled subsidiaries (CFC legislation) and debt financing in high tax countries (thin-capitalisation rules and other restrictions on debt). Thus, from a fiscal point of view common action of the member states is needed. Member states, however, might not agree on a common tax policy with respect to restrictions on debt because some of them accept the loss in tax revenue in favour of better investment conditions for foreign capital in terms of taxes. In summary, the CCCTB calls for far reaching harmonisation of international tax policy. Member states must agree on a common approach with respect to income earned in non-EU countries as well as on common rules with respect to profit shifting to non-EU countries.

3. An Alternative to the Commission’s Proposal: Separate Entity Accounting and Residence Based Taxation 3.1 Separate Entity Accounting Given the tax rate differences in the European Union, the distortions of investment decisions caused by source-based CCCTB might be significant. In order to avoid these distortions, European group taxation could move towards residence based taxation of the group’s profit while preserving separate entity accounting and the arm’s length standard. Under residence based separate entity accounting (RSA), the profits and losses of subsidiaries and permanent establishments are attributed to the group’s parent. The information needed to assess foreign income in the taxpayer’s state of residence represents a well known difficulty of residence based taxation. Within the European Union, however, the member state of residence could resort to common information exchange rules as well as to common tax accounting rules. The member state where the parent is located taxes the group’s income and grants a credit for taxes paid by the subsidiaries. RSA displays many of the advantages of formula apportionment: – Profit and loss consolidation is effected. – Book-profit shifting does not save taxes. – Taxes on profits from intra-group trading can be deferred. RSA represents a major change for member states that exempt foreign profits from domestic taxation. Even member states that rely on worldwide income taxation do not, in principle, tax retained profits of foreign corporations. However, even those member states that exempt foreign profits usually tax other foreign capital income, like dividends, capital gains, interest income or license fees. Moreover, many member states tax retained profits of foreign corporations in case of tax avoidance.

D. Consolidation, Allocation and International Aspects

121

3.2 Investment Decisions It is well known that residence based taxation eliminates the impact of taxes on the decision where to invest (capital export neutrality) 11. Capital export neutrality is based on equal treatment of domestic and foreign income and conforms to both the legal definition of the common market as an internal market characterised by the abolition of the obstacles to the free movement of goods, persons, services and capital (Art. 3 Sec. 1 lit. c EC) and to the principles of non-discrimination and nonrestriction established by case law of the European Court of Justice (ECJ). Under a system of residence based taxation, the state of residence must fully credit foreign corporate income taxes in order to preserve capital export neutrality. An unrestricted credit, however, is not feasible from a fiscal point of view. The ordinary credit is usually restricted to domestic taxes that are apportioned to the foreign income. In case foreign corporate income taxes exceed domestic corporate income taxes (excess credit situation), foreign taxes determine the tax burden of the group and the group is effectively taxed as under the source principle. Consequently, there should be no limitations on the ordinary credit in order to avoid an excess credit situation as far as possible. Residence based taxation conforms to capital ownership neutrality, if all investors face the same tax burden on domestic and foreign investments. The total tax burden may vary between investors, but every investor will prefer the investment with the highest pre-tax return. However, in order to ensure both capital export neutrality and ownership neutrality, profits and other capital income have to be taxed at the same rate in each member state, while the member states are free to choose the respective tax rate. Given these conditions, under residence based taxation neither green field investments nor mergers and acquisitions change the tax burden of the group. In a setting where some countries tax foreign profits while other countries exempt foreign profits from taxation, exemption reduces the tax burden of foreign investment and improves the competitiveness of resident companies in foreign markets. If a single high tax country taxes foreign profits and grants a foreign tax credit, investors located in exemption countries would hold low taxed assets while investors located in the credit country would hold high taxed assets. In the presence of international tax distortions of both investments and ownership patterns, it might be advisable for the member states to restrict residence based taxation to income derived from European sources and to exempt profits from non-EU countries from taxation in order to improve the competitiveness of EU companies in the world market.

3.3 Tax Revenue Effects Residence based taxation does not substantially alter the rules that govern the allocation of corporate tax revenue within the European Union. Since the arm’s length

11

See Homburg, S., Competition and Co-ordination in International Capital Income Taxation, FinanzArchiv 1999, 1-17.

122

Ulrich Schreiber

standard is not replaced with formula apportionment, member states directly control their tax revenue. The tax revenue of the member state where the group’s parent is located might increase because that member state taxes the total European income of the group. On the other hand, losses of subsidiaries and permanent establishments reduce the parent’s taxable income. The resulting loss in tax revenue, however, is offset by future profits of the subsidiaries that increase the parent’s tax base. Residence based taxation reduces the incentive to shift income from one member state to another. Thus, European groups are no longer inclined to manipulate the arm’s length price of intra-firm transactions. Transfer prices continue to determine the tax revenue of the member states, but they tend to be irrelevant for the group’s tax burden. Since transfer pricing is less relevant in terms of tax planning, the administrative and compliance costs of the arm’s length principle might decrease within Europe. Due to restrictions on the foreign tax credit, RSA may cause distortions of investment decisions. However, under separate entity accounting the link between economic activity in a member state and the taxes paid to that member state is weaker than under formula apportionment. A high tax member state may lose tax revenue under separate entity accounting because the group succeeds in shifting bookprofits to a low tax member state. Given this mobility of the tax base, the high tax member state does not necessarily lose real economic activity. Thus, member states might find themselves in a position that is better under separate entity accounting than under formula apportionment. From a fiscal point of view, a serious point of concern with residence based taxation is the incentive to move the tax residence to a low tax country. Resident companies might transfer capital gains to low tax countries. The former residence country loses the tax revenue as well as the benefits from the headquarter activities of resident multinationals. It is unclear how strong the incentive to move would be. Firstly, it is unclear how member states would choose their corporate income tax rate under a system of residence based taxation, given the fact that higher taxes could generate additional tax revenue without deterring EU-inbound investment. Secondly, the non-tax cost of moving the tax residence might be substantial and a European group might prefer to shift profits to non-EU countries instead of moving its tax residence. Lastly, exit taxation may prevent companies from moving. The merger directive12 prescribes a tax deferral for all assets that are attributed to a permanent establishment located in the former state of residence. Assets that are attributed to the moving company are subject to immediate exit taxation. Those assets are probably valuable firm specific intangibles. The liquidity effect of exit taxation might thus be very burdensome. It is, however, an open question if applicable exit taxation rules conform to European law and therefore unclear if the transfer of the parent’s tax residence puts the tax revenue of the respective member state at risk.

12

See EC Directive 2005/19/EC amending Directive 90/434/EEC 1990 on the common system of taxation applicable to mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States (mergers directive), OJ, L 58, 19.

D. Consolidation, Allocation and International Aspects

123

3.4 Third Country-Tax Policy Residence based taxation of a European group is not designed to ensure worldwide capital export neutrality and worldwide capital ownership neutrality. Thus, each member state applies its respective rules for non-EU income of a company including the rules of a double taxation treaty. All income from third countries should be excluded from residence based European group taxation. Only EU-income of an EU-subsidiary or permanent establishment is attributed to the parent. Since RSA avoids a common tax base, there is no need to harmonise tax legislation that aims at preserving this common tax base. Each single member state decides on tax rules targeted at international tax planning. If a member state does not tackle a certain type of international tax planning (e.g. does not introduce CFC legislation or restrictions on debt), it has no direct impact on the tax revenue of another member state. Of course the profit to be attributed to the parent might be reduced but the same is true for the foreign taxes to be credited.

4. The Way to European Group Taxation 4.1 Common Corporate Income Tax Policy? Common tax accounting rules do not substantially contribute to the avoidance of tax distortions with respect to decisions on where to locate investments and companies in the Common Market. However, harmonised tax accounting rules represent a seizable advantage for companies, particularly for small and medium sized companies that suffer from high compliance costs when making use of the Common Market. The advantage of reduced compliance and administrative costs should be decisive when member states consider that policy option. As opposed to RSA, the CCCTB requires far reaching harmonisation not only with respect to the elements of the CCCTB (group definition and formula) but also in regard to the tax policy vis-à-vis third countries. Moreover, the CCCTB changes the structure of the member states’ tax claim on the companies’ profits. Up to now, formula apportionment works as an element of the revenue sharing mechanism between political subdivisions of a nation-state13. By contrast, European formula allocation would be the only revenue sharing mechanism with respect to the corporate income tax collected from European groups. It is, therefore, conceivable that member states would like to choose a rather complex formula that works to some extent like a fiscal equalisation mechanism between member states.

13

In Germany e.g. the corporate income tax revenue a state (Bundesland) is collecting from a resident company is allocated by formula on the basis of wages to the states where the company does business (Art. 107 para. 1 Basic Constitutional Law, Grundgesetz, § 2 Apportionment Tax Act, Zerlegungsgesetz). The corporate income tax is then divided (Art. 106 para. 3 Basic Constitutional Law) between the states and the federation (Bund). Finally, wealthier states transfer a part of their tax revenue to less wealthier states (Art. 107 para. 2 Basic Constitutional Law, § 4 Financial Equalisation Act, Finanzausgleichgesetz).

124

Ulrich Schreiber

The CCCTB requires new structures to administer the corporate income tax. Given that the member state where the group’s parent is located assesses the group’s profit and attributes that profit to the other member states according to the formula, there must be close cooperation between the member states’ tax administrations. Member states have to make sure that the tax administrations across the Union uniformly interpret and apply the common tax rules. The ECJ would be the highest level of jurisdiction. However, it might be advisable to have a lower European court to decide day-to-day cases that arise under the CCCTB. It is an open question if the member states are ready for such a close cooperation in the field of corporate income taxation. Member states compete with each other over the corporate income tax rates 14. Neither the move to CCCTB nor the move to RSA would reduce that type of tax competition. In fact, the reduction of tax obstacles to cross-border investment effected by CCCTB or RSA intensifies tax rate competition. In both cases member states may fear that headquarters of European groups could move. Under RSA the group experiences the incentive to take advantage of low corporate income tax rates for the group’s profit, and under CCCTB the group might wish to take advantage of favourable administrative procedures concerning the determination and allocation of the group’s profit in the member state that is responsible for the assessment. Given the importance of the corporate income tax rate, member states might be reluctant to proceed to any European group taxation scheme without an agreement on the corporate income tax rate, even if they should accept common rules to determine taxable profit. The issue of a harmonised corporate income tax rate raises the well known questions if tax competition is necessary to constrain the budgetary policy of member states, or if tax competition is harmful because it restricts redistribution policies and damages the tax systems of the member states 15. Some member states might be of the opinion that unbridled tax competition is harmful. Other member states might see tax competition as a useful instrument of national economic policy. Given the different views on tax competition, an agreement on the corporate income tax rate represents a major problem for a common corporate tax policy. With an agreement on a common corporate income tax rate, however, both CCCTB and RSA would reduce economic distortions and, from this perspective, could be candidates for a European group taxation scheme16.

4.2 Single-Stage Solution or Step-by-Step Solutions? European groups are often not able to fully exploit the advantages of the Common Market due to national tax law that hampers cross border investments. On the other 14

15

16

See European Economic Advisory Group at CESifo, Report on the European Economy 2007, Munich, 2007, 121-132. See Sørensen, P. B., Company Tax Reform in the European Union, International Tax and Public Finance 2004, 91-115. See Wissenschaftlicher Beirat beim Bundesministerium der Finanzen (Advisory Board to the Federal Ministry of Finance), Einheitliche Bemessungsgrundlage der Körperschaftsteuer in der Europäischen Union, Berlin 2007, 75-76.

D. Consolidation, Allocation and International Aspects

125

hand, diverging tax law in the member states gives rise to international tax planning which in turn threatens the member states’ tax revenue. The member states’ tax legislation targeted at preventing cross border tax planning of European groups often violates European law. Given both the tax obstacles and the tax planning opportunities, there is an urgent need for group taxation schemes that are compatible with the Common Market. The European Commission’s proposal of a common consolidated corporate tax base (CCCTB) is a remarkable step in that direction. It displays the main advantage to abolish the arm’s length standard of profit allocation within a European group. Since it is an arduous and costly task to assess arm’s length prices, there are good reasons to be in favour of the Commissions proposal17. The CCCTB is, however, a very ambitious undertaking. It is a single-stage solution, i.e. all elements of the CCCTB have to be commonly introduced. The fact that the CCCTB completely changes the taxation of European groups hinders the introduction of the new group taxation scheme. In order to overcome this obstacle, a possible scenario could be that the companies are granted an option18 to be taxed under the CCCTB. The option would allow companies to choose whether to consolidate or not. Though an option increases the companies’ tax planning cost, it works in favour of companies. Only companies that expect to gain from the CCCTB would opt for consolidation, while other companies would decide to remain under the separate entity standard. An option, however, threatens the tax revenue of the member states and it forces the member states to administer a parallel system of group taxation within the European Union. The introduction of the CCCTB might be facilitated if an option is granted, but in order to avoid the tax planning opportunities of an option, member states might be in favour of the CCCTB to be compulsory. Given these difficulties with respect to the introduction of the CCCTB, another scenario is that member states refrain from a single-stage solution and choose a stepby-step strategy to adapt their group taxation schemes to European Law and the needs of the Common Market by addressing the most pressing problems with targeted solutions. It is very likely that the ECJ’s Marks & Spencer ruling19 forces member states to open their group taxation regimes for European losses or to completely abolish group taxation rules. The abolishment of group taxation rules would damage the economy of the member states as well as the European economy. Therefore, the 17

18

19

See Martens-Weiner, J., Company Tax Reform in the European Union, New York 2006, 105107. See Commission of the European Communities, Communication from the Commission to the Council, the European Parliament and the European Economic and Social Committee, Implementing the Community Programme for improved growth and employment and the enhanced competitiveness of EU business: Further progress during 2006 and next steps towards a proposal on the Common Consolidated Corporate Tax Base (CCCTB), Brussels, COM (2007) 223 final, 6. European Court of Justice (13 December 2005), Case C-446/03 (Marks & Spencer), ECR I 2005, 10837. Great Britain has reacted by allowing the deduction of foreign losses from the British tax base that definitely cannot be relieved in the member state where the loss originated.

126

Ulrich Schreiber

only viable option is to reform the group taxation schemes. Offering current inclusion of foreign profits and losses to a European group might be a superior solution from both a fiscal and an economical point of view compared to solutions that focus only on foreign losses. The case Cadbury Schweppes20 compels member states to adjust their CFC legislation to European law. Current inclusion of foreign profits renders CFC legislation superfluous within the Union. Residence based taxation should not infringe upon European law because it makes neither a distinction between domestic and foreign companies, nor does it depend on the level of taxation in a certain member state. Some member states (France, Denmark and Italy) have already decided to run residence based international group taxation schemes. If more member states bring their group taxation schemes in line with the residence principle, a future European group taxation scheme could build on these structures and on tax rules that member states are already familiar with. Member states that are about to move their group taxation schemes towards the residence principle in order to bring them in line with European law should support the Commissions proposal to introduce common tax accounting rules. With common rules (including loss compensation rules) to determine the taxable profit, residence based taxation can be administered without recalculating the taxable profit of foreign subsidiaries and permanent establishments. Once common rules to determine the corporate tax base are introduced, both compliance costs and administrative costs of RSA will decrease. Under RSA profit allocation within the European group continues to rest on the arm’s length standard. Neither companies nor tax administrations are forced to simultaneously run two different systems of international profit allocation. A residence based European group taxation scheme could address the problem of intragroup transactions. A feasible solution must rest on the realisation principle which is the basic principle of the common tax accounting rules. The group’s profits should be realised and taxed when goods are sold to a third party from the perspective of the group as an economic unit. Thus the taxation of intra-group profits has to be deferred to ultimate realisation. Deferred taxation of gains from intra-group transactions might on the one hand be compatible with the member states’ desire to preserve their tax claim on the respective gains, and on the other hand prevent these gains from being taxed before ultimate realisation. In this step-by step scenario, the last step on the way to a European group taxation scheme would be to choose between the arm’s length standard and formula apportionment. Given that the member states do not proceed to either system before having implemented an alignment of corporate income tax rates, both systems have to be primarily judged on their consequences with respect to compliance and administrative costs and in regard to their tax revenue effects. The total costs of the cor20

European Court of Justice (12 September 2006), Case C-196/04 (Cadbury Schweppes), OJ C 281, 5. Great Britain has focused CFC rules on controlled companies that do not carry on a genuine economic activity in terms of both business establishment and individuals working for the CFC.

D. Consolidation, Allocation and International Aspects

127

porate tax system might be reduced when separate entity accounting is substituted with consolidation and formula apportionment within the European Union. The tax revenue risk due to companies’ tax planning is lower if the corporate tax rate differences between member states are substantially reduced. It is an open question, however, if the advantages of formula apportionment outweigh the difficulties that the member states face when they have to agree on the distribution of the tax revenue via the formula, and when they have to decide on a common international tax policy.

5. Summary European groups are not able to fully exploit the economic advantages of the Common Market because they face the risk of double taxation and they are burdened with high compliance costs. On the other hand, member states have to cope with European groups’ tax planning strategies which threaten their tax revenue. The European Commission’s proposal of a common consolidated corporate tax base (CCCTB) addresses these problems. The CCCTB is a remarkable attempt to come to a group taxation scheme that fits into the Common Market. However, the CCCTB requires far reaching harmonisation of corporate income tax law. Member states have to agree on the group definition and the apportionment formula as well as on rules to protect the common tax base against trans-EU tax planning; moreover, member states have to set up adequate structures to administer the common tax base. It might thus take a long time to implement the CCCTB. On the other hand, there are pressing problems caused by ECJ’s rulings that have to be addressed by the member states. Therefore, it might be advisable to choose a stepby-step strategy instead of a single-stage approach, as it is the case with the CCCTB. The first step should be to introduce common tax accounting rules in order to take advantage of reduced compliance and administrative costs. European law forces member states to deduct losses of EU-subsidiaries from the EU-parents income, and European law sets limits to the inclusion of foreign EU-subsidiaries’ low tax profits into the income of the EU-parent. Residence based group taxation has the potential to overcome both problems. Current inclusion of the subsidiaries profits and losses into the income of the group’s parent also sets limits to international tax planning. Given that the group taxation schemes of the member states move towards residence based taxation, European group taxation could build on these structures of the member states’ tax law and abolish intra-group profit taxation by introducing a common system of deferred taxation. Given this scenario, the replacement of separate entity accounting and the arm’s length standard with consolidation and formula apportionment could be the last step. Member states might, however, not be inclined to accept consolidation and formula apportionment without an agreement on the corporate income tax rate.

128

Jack Mintz

II. Europe Slowly Lurches to a Common Consolidated Corporate Tax Base: Issues at Stake Jack Mintz

1. Introduction The European Union is moving ahead, albeit slowly, towards a proposed system to harmonize corporate income tax bases for companies operating in EU member states. Beginning with several broad proposals,1 discussions have become focused on the adoption of a common consolidated corporate tax base that would be similar to that used in Canada and the United States to integrate corporate tax bases. In the discussion below, I shall review the various issues involved with establishing a common tax base in Europe with reflections and comment on whether some of the discussion is headed in the right or wrong direction, as an outside observer. In my view, a long road remains ahead but the progress has been more substantial than what I would have expected, given the lack unanimity among member states for a harmonized corporate tax base. The rationalization of business activity at the pan-European level and the challenges arising from European Court of Justice decisions, typically centred on the principles of freedom of establishment under the EU treaty, have forced European governments to seek new approaches to corporate tax policy. Thus, economic, compliance and administrative costs arising from separate taxes operating in each EU country is becoming a greater barrier to consolidation of the European corporate sector, ultimately leading to a need to integrate European corporate tax bases. If any advice that needs to be given is that the EU should have a system that is relatively simple – it would not be useful to be too “pure” in the selection of factors, income and other aspect of the common base. Otherwise, administrative and compliance costs become too large in developing a common corporate tax base that would be apportioned to member states participating in the system.

2. What’s it all about? Prior to assessing current European proposals, it is useful to lay out some principles that serve as a guide towards a good corporate tax system in Europe. By their nature, taxes are beneficial to fund public services but they inevitably result in economic, compliance and administrative costs. Each member state in a new system will be concerned about the effect that consolidation would have on its own revenue. Both governments and the taxpayers will be concerned about reducing the costs associated with the collection of taxes. With respect to costs, the best 1

See European Commission, Company Taxation in the Internal Market, Commission staff working paper, Office of Official Publications of the European Communities, Luxembourg, 2002.

D. Consolidation, Allocation and International Aspects

129

corporate tax structure is one that is efficient (minimizes economic distortions), fair (similar burdens on business activities), and does not result in undue administrative and compliance costs. A move from separate member state corporate tax systems to a common consolidated corporate tax base results in a significant change to the calculation of corporate taxes. Instead of calculating corporate profits and taxes by each member state, harmonization would result in the aggregation of profits determined by a common tax base and apportioning amounts by a formula to determine profits earned in each jurisdiction. The apportioned profits would be taxed at a rate selected by the jurisdiction. In Canada, provincial governments typically use tax credits to vary the tax burden by business activity. In the US, state governments make adjustments to the common base, choose apportionment factors or allow for tax credits to change tax burdens across countries. Given the current separate European tax bases, what are the gains that could be achieved from consolidation of corporate tax bases? Any system of formulary apportionment is not perfect and, based on Canadian and US experience, a common consolidated corporate tax regime is bound to result in economic, compliance and administrative costs. However, one would like to see some improvement with a consolidated tax base compared to the existing state of affairs. A consolidated tax base would improve both efficiency and fairness to the extent that differences in tax bases among member states are reduced, tax bases are better coordinated, companies are more able to set off losses from economic profits and a more coherent treatment of outbound and inbound investment from Europe is developed. Administrative and compliance costs could also be reduced under a common consolidated tax base to the extent that transfer prices for transactions between related prices become less critical in determining tax liabilities, businesses have fewer rules to determine their European profits, and governments can reduce their reliance on legislative policy and administrative practices that try to combat tax avoidance and evasion. Some evidence even suggests that the apportionment method for corporate income taxes reduces the scope for profit shifting. Mintz and Smart2 found that corporate profits were less sensitive to corporate rate changes for companies that allocated income to more than one province compared to multi-jurisdictional companies that did not allocate income but operated in more than one jurisdiction with separate subsidiaries (Canada does not permit consolidation for tax purposes). With a reduced incentive to shift profits, governments would also have less incentive to engage in tax competition to counter income-shifting by multinationals from high to low-tax jurisdictions. Policy analysts often focus on the economic gains from moving to a common consolidated corporate tax base by reducing differential tax burdens on businesses. Yet, both Canadian and US experience shows that provincial and state governments have considerable leeway to adjust their tax rates, exclusions and credits so that tax 2

See Mintz/Smart, Journal of Public Economics 2004, pp.1149-1168.

130

Jack Mintz

burdens could vary significantly across business activities.3 The most important gain to taxpayers arising from formulary apportionment is to reduce compliance costs. The gain to governments is improving their ability to collect and administer taxes, especially reducing the scope for transfer price and financial manipulation to shift profits from one jurisdiction to another. To sum up, three important reasons can be given for Europe to adopt a common consolidate corporate income tax base. – European economic integration makes it increasingly difficult to administer corporate tax bases with separate accounting, especially with respect to transfer pricing. – Companies wishing to operate at the pan-European level would be able to reduce compliance costs with the tax system, one of the barriers to economic integration. – The use of an apportionment or allocation method for a common corporate income tax base could counter tax competition in terms of corporate rate cuts, compared to existing regimes. Thus, it is critical that any proposal to move to a common consolidated tax base should improve efficiency, fairness and compliance/administrative cost reduction compared to existing tax systems.

3. What are the Issues? In developing a common consolidated corporate tax base for Europe, a large number of issues need to be resolved. A recent paper provides a review of the current issues left for agreement that provides a good basis for discussion below.4 With EU discussions, corporate tax harmonization is focused on a common consolidated tax base using formulary apportionment. Other options, particularly home state taxation, have been dropped from the agenda, an outcome that I believe is probably best. The specific issues that need to be dealt with in developing the approach is listed in the attached table. I will review various components.

3.1 The Tax Unit and Base As many have noted in the past,5 a difficult issue faced by European negotiators is determining the tax base for a common corporate tax. Unlike Canada and US where the federal corporate income tax is a good starting point for developing a common base, Europe has no central taxing authority. Thus, EU member states have been

3 4

5

See Mintz, International Tax and Public Finance, 2004, pp. 221-234. See European Commission, “An overview of the main issues that emerged during the discussion on the mechanism for sharing the CCCTB,” CCCTB\WP\052\doc\en, Working Document, February 27, Brussels 2007. See McLure, The long shadow of history: sovereignty, tax assignment, legislation and juridical decisions on corporate income taxes in the US and the EU, mimeograph, Hoover Institute, California, 2006. And see Martens-Weiner, Company Tax Reform in the European Union, New York, 2006.

D. Consolidation, Allocation and International Aspects

131

looking at International Accounting Standards (IAS) as a starting point to determine the common base although there will not be a formal link between IAS measures of profits and taxable profits. This is probably the best that one can do since using accounting standards has its limitations. Accounting is at times judgmental and options are sometimes available to companies to determine their profits. Practices can vary across companies and countries. With private companies that have less with concern over reported earnings, businesses can manipulate tax payments by choosing alternative inclusions and deductions. Thus, a fairly common set of rules should be developed and these would differ from accounting practices that vary across jurisdictions. A similar issue arises in determining whether a member state has the right to tax a company. Canada and US use rules to establish whether the sub-national government may receive apportioned income but these rules are established by federal legislation. In the case of Europe, permanent establishment rules can vary across European countries so some harmonization will need to be considered.

3.2 Consolidation In principle, consolidation of income of corporate group members should be implemented – otherwise companies can choose for tax reasons to organize themselves as single entities by state or as a consolidated company subject to apportionment rules. Such choices can increase the scope for tax planning such as transfer pricing and financial restructuring. The US requires consolidation that reduces the scope for tax planning. Canada, as mentioned above, does not permit consolidation so companies have an option to avoid the formulary approach. Many Canadian companies are subject the allocation system rather than setting up separate subsidiaries by province for both tax and non-tax reasons – about 55 percent of corporate taxable is allocated among provinces in Canada.6 The Canadian experience has shown that it can be hard to adopt consolidation if it is not in place. A 1985 federal proposal to adopt a UK-style loss transfer system among members of a corporate group failed as a result of provincial resistance, particularly those provinces that were bound to lose corporate revenues. The EU discussions have focused on whether to adopt or not a consolidation system for those companies opting for apportionment of a common base rather than staying with their own tax bases. While adoption of a consolidation would be similar to the US, the option to adopt a common base puts the system closer to the Canadian model, which effectively allows companies to choose whether to consolidate activities or not across member state boundaries.

6

The small business sector (companies with less than $15 million in assets) comprises about a third of corporate taxable income. That implies a relatively high degree by which large companies allocate income across two provinces or more rather than operate separate corporations by province. See Technical Committee on Business Taxation, Report, Finance Canada, Ottawa, 1998.

132

Jack Mintz

The EU option to adopt the common base would lead some revenue loss7 from tax planning and will require more compliance and administrative resources when both separate accounting and consolidation are both in place. Without consolidation, further revenue losses could be expected. Both the option for businesses to join the common tax base and the absence of consolidation will provide significant scope for tax avoidance. It is appropriate that consolidation should be required for companies choosing to accept a common base. Rules for consolidation, however, can be complex. Membership of a corporate group must be determined by some threshold of ownership or control. Rules must be put in place to deal with loss pools, capital costs and reserves when subsidiaries enter and exit the corporate group. Corporate reorganization rules that vary by member state are also affected. However, these complexities are not an impossible barrier to consolidation since many countries have dealt with these issues. Consolidation by corporate group, however, is a complex issue.

3.3 The Formula A common consolidated tax base will require the use of a formula. Various issues are involved, including whether the formula is specific or uniform across companies, specific to each jurisdiction and upon which factors will be chosen to apportion income to member states. The Canadian and US formulary apportionment uses firm-based factors to divide the common tax base among sub-national governments. In Canada, the general formula is uniform among provinces, which requires corporations to allocate income to each provinces according to the shares of payroll and sales (destination basis), equally weighted. Special formulas apply to certain industries, primarily in finance and transportation. By and large, the provinces have agreed to use the same formula, which has been in place for about 60 years, despite some complaints especially by the resource-rich provinces who feel that the use of sales reduces the allocation of the tax base to them. In the US, most state governments had used the Massachusetts formula – equal weighting of sales, payroll and capital in the past; however, many have chosen other formulas now, with greater weighting of sales to encourage more employment in the state. Special formulas apply to some industries. EU discussions are considering the Canadian-US approach to formulary apportionment but they have also toyed with the notion of a “macro-approach” whereby a formula would be based on economy-wide factors rather than based on the firm’s characteristics. The benefit of a macro-approach is to reduce the scope for tax planning since the alternative encourages companies to shift labour and capital to lowtax rate jurisdictions in order to increase the allocation of income to these states 7

Devereux and Loretz have estimated that an option to join the common base would reduce corporate tax revenues by 1 percent. A requirement to be subject to the common base would increase revenues by 8 percent. See Devereux/Loretz, The Effects of Formula Apportionment on Corporate Tax Revenues,” WP 06/07, Oxford University Centre for Business Taxation, 2007.

D. Consolidation, Allocation and International Aspects

133

(thereby providing a lower cost of capital and labour).8 However, given companies have much different attachment to individual states that would not be reflected in a “macro” formula, the EU discussions are wisely being focused on the Canadian or US approach to formulary apportionment. As for special formulas for certain industries, the EU discussions so far have not paid much attention to these questions except to note the need for a special formula to apply to financial companies. It shall find that other special formulas will be needed since labour, capital and sales are inadequate to split income not only in finance but also transportation and possibly telecommunications. The most interesting question that will need to be addressed is determining the exact formula. The formula will affect how revenues are distributed across member states and the tax paid by a specific corporation. Greater reliance on origin-based factors – labour and capital – provides more revenues to those states with production. The use of sales, measured on a destination basis, provides more revenues to states with large consumer markets. The EU has also considered the use of value-added – labour income and economic rents accruing to production – as a basis for splitting corporate income although the proposal is less likely to be adopted. Given the importance of valueadded tax systems, this option is intriguing as a destination-based method of apportioning corporate income. However, value-added calculations under credit-invoice VAT is quite different than accounting-based calculations so it is not entirely clear how feasible value-added calculations would be to use in a formula without substantial complexity. However, one could consider measuring value-added similar to the Italian IRAP and Hungarian regional taxes, which then would be accounting and origin-based. The use of payroll, capital and sales in the formula raise a number of issues that are being discussed by the EU member states. Payroll is simplest to measure except for two specific considerations. The first is performance-based employment compensation since EU countries do not have the same treatment of stock options and some other forms of deferred compensation. The second is with respect to cross-border arrangements related to temporary employment and contract work. In the Canadian and US systems, no adjustment is made for employment transfers and contracting – I suspect it is not a major issue in terms of the distribution of revenues and tax planning, at least yet. These technical questions will need to be sorted out but it might be easiest to use payroll paid according to the primary place of employment. Some concern has been expressed in EU discussions that payroll should be adjusted for wage levels since low-wage countries might otherwise get a smaller share of the tax base. However, wages reflect productivity of factors, not just the cost of living, so adjustments are bound to hurt countries with higher skilled labour. Neither Canada nor the US adjust for differences in wage rates even though wage levels vary across jurisdictions, perhaps not as widely as Europe but still nevertheless significantly. 8

See Mintz/Weiner “Exploring formula allocation for the European Union,” International Tax and Public Finance 2003, pp. 695-711.

134

Jack Mintz

With respect to capital, the usual issues arise such as whether to include intangible assets and to make adjustments for leased assets. Inventories, intangibles and leased assets contribute to a highly mobile tax base that could influence the factor proportions easily. Probably it would be easiest to exclude intangibles and inventories since the assets are so mobile and focus instead on fixed property and equipment. On the other hand, adjustments to fixed asset calculations for leasing might be appropriate since companies could easily shift their assets into permanent establishments operating in low-tax jurisdictions to be leased to other jurisdictions. Again, much depends on how severe the problem shall be and whether the issue needs to be addressed. A further point of contention is whether to measure capital on a “flow” basis rather than “stock” basis as in the US formula. Gross investment flows reflect replacement of depreciated capital and new investment – flow measures would provide greater allocation of corporate profits to jurisdictions with more depreciable capital. Any asset sales would also reduce the flow measure and lead to potentially negative weights. Stock measures, net of accumulated depreciation, are thus superior conceptually. While it would be desirable to measure assets on a fair market basis, it is highly unlikely that values are easy to develop. Therefore, stock measures would rely mostly on book values for easier compliance. As for sales being included in the formula, the EU discussions focus on whether origin or destination measures of sales should be used in the formula. If property and labour are already included in the formula, origin-based sales, which also includes intermediate factors, would change production shares somewhat when measuring profits on an origin basis. However, sales on a destination basis seem better to consider in the formula since the sales factor could adjust revenues in favour of consuming member states, making it easier to develop a final agreement. Ultimately, the choice of factors will be critical in allocating the tax base that each government receives as well as the tax payments made by companies. Devereux and Loretz9 suggest, however, that the impacts would not be large using different formula. The hard part of reaching an agreement is choosing the formula since it becomes a zero-sum game among EU states in a static sense – any more allocated to one country reduces allocations to the rest. To the extent that a common corporate tax base leads to economic improvement, it is easier to achieve an agreement with selected factors since countries would share a growing tax base.

3.4 Non-Business Income Most income tax systems, including Europe, distinguish between active, investment income and capital gains. Active income – business income more specifically – is arrived at by deducting costs from receipts derived in the course of regular trade or activity. Investment income, such as dividends and interest, and capital gains is taxable after allowing a deduction for costs related to the investment only (such as broker fees). For the purposes of formulary apportionment, Canada does not distinguish 9

Devereux/Loretz, The Effects of Formula Apportionment on Corporate Tax Revenues,” WP 06/ 07, Oxford University Centre for Business Taxation, 2007.

D. Consolidation, Allocation and International Aspects

135

between business and non-business income while US states will allocate non-business income according to location of physical or intangible property or taxpayer domicile. Distinguishing amongst different sources of income creates greater complexity compared to applying a single formula to the common tax base. It could also encourage greater tax planning if different approaches are taken. It seems that little justification can be given for separation of business and non-business income, at least in my view.

3.5 Inbound and Outbound Investment A difficult issue facing EU member states in developing a common tax base relates to third party countries. Given the importance of the federal government in determining international tax policies, Canada and US deal relatively simply with outbound and inbound investments since the federal international tax rules generally apply uniformly to all sub-national governments. Canada levies a federal-only tax on income received from foreign affiliates by applying a federal rate that is 10 points higher than the rate applied to domestic income (more formerly, the federal rate is reduced by an abatement equal to 10 points for income allocated to a province). With respect to foreign portfolio income, the province applies its own tax if the income is earned by a permanent establishment operating there. The province will give a credit for foreign withholding taxes, similar to the federal government. US policy is similar to Canada in this respect as states tend to follow federal rules. However, it can be more complicated if states choose to tax multinationals on a different basis such as the famous California corporate tax levied on a formulary basis. For the EU, however, the starting point is that each country has its own set of bilateral tax treaties and method used to tax foreign-source income earned by multinationals. Unless multilateral tax treaties are considered by the EU to bring harmonization among member states, the EU will be left to develop some other method to handle taxation of foreign source income and withholding taxes on payments to non-residents of third parties. One approach is simply to apply the common tax base to European source income only, leaving foreign-source income to be taxed according to where the permanent establishment resides. This could be an awkward approach, however, since it will require a clear allocation of income and expenses between domestic and foreign sources. Moreover, recent European Court of Justice cases (such as Cadbury Schweppes and Columbus Container Services) may impact on the use of deferral and passive income rules used by individual countries. EU states might look for a common approach to the taxation of foreign-source income if it becomes difficult to apply anything other than the exemption system within the EU.

4. Conclusions Much work is left to do if the EU is to adopt a common consolidated company tax base. Progress has been made on a number of issues but many difficulties remain, especially in choosing the formula for dividing income among member states. Over

136

Jack Mintz

time, however, I believe that governments and the business sector will realize that the integration of European economies leaves little choice but to harmonize company tax bases in some way. COMPARISON OF APPORTIONMENT OR ALLOCATION METHODS FOR CORPORATE INCOME TAXATION United States

Canada

EU Proposals as of March 2007

Name Given

Apportionment

Allocation

Apportionment

Jurisdiction

States

Provinces

Member States

Tax Collection Agreement with Central Government

No

Yes for 7 provinces. Ontario joining system. Quebec and Alberta collect own corporate tax

No

Taxable Unit

Corporation

Corporation (publicly-traded trusts to have special formula to be determined)

Company

Jurisdiction for Taxation

Nexus – substantial connections of activities to the state and 1993 Geoffrey Ltd. Vs South Carolina provided for intangible property to have a connection (royalty income in Delaware)

Permanent establishment – fixed place of business

Permanent Establishment

Tax Base

Federal base with additions and subtractions by state to determine apportionment of base

Generally federal definition of income to determine allocation before application of credits

Common base using International Accounting Standards principles

Consolidation

Yes (unitary combination)

No

Not yet determined – will be optional to adopt a common base

D. Consolidation, Allocation and International Aspects

137

Factors

Chosen by state so factors may not add up to 100% across all states – Massachusetts Common Formula (1/3 capital, payroll, sales) used by most states in 1957 – Most double weight sales since 1977. – Special treatment of finance (financial assets or income), transportation (revenue miles), construction contractors, broadcasting and publishing

Generally common among all provinces based on average payroll and sales. Special treatment of transportation (revenue passenger miles), finance (loans and deposits) and insurance (premiums)

– Use payroll and/or employees, capital and sales Or value-added Or Macro-approach (latter two on standby) – special formula for finance, maybe others

– Sales

– destination basis using retail gross receipts – throwback rule for tangible personal property sales if not taxed in destination state

– destination basis using retail sales

– origin or destination basis to be used

– Labor

– payroll of employees including benefits

– payroll of employees

– payroll with possible adjustment for outsourcing and secondments

– Capital

– accumulated investment in tangible assets including inventory

None

– non-financial assets – tax value of assets, – consideration of lease and intangible assets

Non-business income

Allocated to specific states

Part of business income

Perhaps similar to the Canada

Foreign-source income

States can include worldwide income to determine allocation.

Tax at federal level only on business income. Investment income subject to provincial tax with credit for foreign tax.

Excluded from the common tax base or included under common rules subject to existing DTSs

138 Formula

Jack Mintz

Tax = Sum of tax rate x state weight x national income with weight being shares of factor in state to total in all states where company does business

Tax = Sum of provincial tax rate x provincial weight x national income with weight being shares of factor in province to total in all provinces where company does business

To be determined but likely similar in structure to the US and Canada

Sources: Weiner, Joann M., Company Tax Reform in the European Union, Springer, New York, 2006 and European Commission, “An overview of the main issues that emerged during the discussion on the mechanism for sharing the CCCTB,” CCCTB\WP\052\doc\en, Working Document, February 27, Brussels, Belgium, 2007.

D. Consolidation, Allocation and International Aspects

139

III. Panel Statements 1. Mary Bennett

I am pleased to have the opportunity to say a few words today about the CCCTB project from the perspective of the OECD Secretariat. A significant majority of OECD Member countries are Member States of the EU, and an even higher percentage of EU Member States are OECD Members. So, regardless of one’s views on the merits or demerits of any CCCTB proposal that might emerge from the project, it’s hard to ignore the potential implications of such a proposal on the work of the OECD. When I speak of the work of the OECD, I am thinking in particular of two primary documents: the OECD Model Tax Convention on Income and on Capital and the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations. These two documents have long provided the framework for international tax relations both among EU Member States and between EU Member States and third countries. Even for those of us who are simply observing the CCCTB project from the sidelines, it’s obvious that the project could have significant direct and indirect effects on those two pillars of the international tax structure – effects both on the basic principles they reflect and on the way those principles are developed and applied in practice. Let me start with some very obvious points. The first one relates to transfer pricing. The arm’s length principle that underlies the OECD Transfer Pricing Guidelines has achieved widespread acceptance across the globe as the basis on which transactions between associated enterprises should be priced in order to achieve a fair division of tax revenues between national governments. A core objective of the CCCTB project is to replace that principle for transactions between members of corporate groups within the EU with a yet to be developed and tested sharing mechanism, likely some form of formulary apportionment. A move away from the arm’s length principle within an important and sophisticated region, even if motivated by the special considerations relating to the creation of a single internal market, may have unpredictable consequences on the consensus currently reflected in the widespread use of the Transfer Pricing Guidelines around the world. A second obvious point involves tax treaty relationships. The OECD Model Tax Convention serves as the basis on which over 2500 bilateral tax treaties have been negotiated worldwide, including hundreds both between EU Member States themselves and between EU Member States and third countries. It’s clear that the rules currently applicable under any bilateral tax treaty between two EU Member States would have to be overridden in various respects if those two Member States were to adopt a CCCTB regime. It’s also clear, however, that the adoption of a CCCTB regime by any number of EU Member States would put significant pressure on the bilateral aspect of those States’ treaty relationships with third countries. The work

140

Mary Bennett

done thus far in the CCCTB project has highlighted the potential need for common EU treaty policy on a number of significant points, such as the determination of tax residence, the definition of permanent establishment, the determination of profits attributable to permanent establishments, etc. One senses a possible increase in pressure to develop an EU-specific Model Treaty for relations between EU Member States and third countries, and perhaps even a move towards a multilateral EU Treaty for such relations. Let me turn briefly to a few specific issues and more importantly to the questions they seem to raise. In the area of the territorial scope of the CCCTB, the issue of whether to include foreign income of CCCTB tax residents looms large. Either way, there has been discussion of the need for a common definition of permanent establishment and a common approach to determining the profits attributable to permanent establishments. To what extent might these needs give rise to EU approaches that diverge from the consensus reflected in Articles 5 and 7 of the OECD Model? Similar questions arise with respect to the possibility of including in the CCCTB the income of non-residents attributable to permanent establishments in the EU. There appears to be general agreement that the arm’s length standard reflected in Article 7 of the OECD Model Treaty and in most existing treaties should govern the amount of the non-resident enterprise’s income that would get included in the CCCTB. But if the CCCTB sharing mechanism caused a greater (or lesser) amount to be allocated to the EU jurisdiction where the PE is located, what would be the effect on the tax liability there of the non-resident enterprise (or on the double tax relief obligation of that enterprise’s residence country under its bilateral treaty with the relevant EU country)? What about the sharing mechanism itself – would it be based on factors that could cause business profits to be taxable in a jurisdiction even in the absence of a permanent establishment (e.g. a destination-based sales factor)? What about the treatment of payments by members of a CCCTB consolidated group to other persons – would existing source, withholding, and bilateral treaty relief consequences continue to apply to them, or would they somehow implicate the source taxing jurisdiction of multiple CCCTB jurisdictions at once? How would the continuing need to apply the arm’s length principle to transactions between members of the CCCTB consolidated group and non-member affiliates be handled by CCCTB jurisdictions? Would transfer pricing disputes currently handled under the arm’s length principle of Article 9 and the mutual agreement procedure of Article 25 continue to be dealt with as a bilateral issue between the specific EU Member State and its third country treaty partner? How would any alternative process fit with the Article 26 exchange of information provisions in a bilateral treaty? Would documentation requirements be affected by possible changes in CCCTB taxpayers’ record-keeping practices? If introduction of the CCCTB were to create pressure for widespread re-negotiation of EU Member States’ treaties with third countries, what would be the process for developing streamlined mechanisms to achieve that? I know I’ve added more questions than answers to this debate, but that’s the luxury of being on the sidelines, at least for the time being.

D. Consolidation, Allocation and International Aspects

141

2. Bruno Gibert

Je remarque et salue tout d'abord la présence dans l'assemblée de plusieurs membres du Forum, comme Theo Keijzer, Guy Kersch, Jim Kelly et Jasper Wang-Holm, sans oublier bien sûr les représentants de la Commission Européenne qui ont lancé ce projet et assurent le secrétariat du Forum, en particulier Matthias Mors. Mon propos consistera à faire des commentaires sur l'ACCIS sous l’angle du Forum Conjoint de l’Union Européenne sur les Prix de Transfert, à voir dans quelle mesure l'expérience acquise et les progrès réalisés dans cette enceinte éclairent le projet d'ACCIS (à noter pour une fois que le sigle français, ACCIS, pour Assiette Commune Consolidée pour l'Impôt des Sociétés, est plus aisé à prononcer que l'acronyme anglais de CCCTB, pour Common Consolidated CorporateTax Base !). Pour résumer mon propos par une boutade, je dirais que si l'ACCIS aboutit, le Forum disparaît, perspective peu réjouissante quand on préside un organisme ! Mais peut-être est-ce le Forum qui peut faire disparaître l'ACCIS ?

1. Les travaux du Forum confirment les problèmes de coût de conformité et de double imposition résultant des prix de transfert L'importance des prix de transfert dans l'intérêt de la mise en place de l'ACCIS a été soulignée par les intervenants précédents, notamment Mr Krister Anderson. L'enjeu de ce sujet est pleinement confirmé par les travaux menés par le Forum depuis 2002. L'accroissement des obligations documentaires et du contrôle des prix de transfert est une tendance continue depuis l'adoption des lignes directrices de l'OCDE en 1995 : en attestent par exemple les renforcements des législations des Etats Membres comme celles intervenues en France dès 1996, mais aussi en Allemagne en 2003 et, tout récemment encore, en Espagne à la fin de l'an dernier. Cette tendance est loin d'avoir atteint son terme avec, notamment, les nouveaux Etats membres de l'Union qui sont à leur tour en train de développer leur législation et leurs contrôles dans ce domaine. Sur un plan concret, il n'existe pas beaucoup d’indicateurs statistiques sur ce sujet. Pour la France, le montant des rehaussements d'impôt dans ce domaine sont passés de 200 millions € en 1995 à environ 1 milliard € actuellement, soit cinq fois plus. Le Forum a publié une donnée, certes particulière, mais intéressante et qui corrobore l'ampleur du sujet, c'est le nombre de dossiers ouverts dans le cadre de la procédure de la convention européenne d'arbitrage : – Publication en mai 2005 des données au 31 décembre 2005 : 107 dossiers étaient pendants. – Publication en sept 2006 des données au 31 décembre 2006 : 161 dossiers étaient pendants. Ces chiffres sont élevés, et croissants.

142

Bruno Gibert

L’application des mêmes règles issues des lignes directrices de l’OCDE par les Etats Membres de l’Union Européenne ne permet donc pas d'éviter l'accroissement du coût en termes d’obligations déclaratives, de procédures de contrôle et de risque de double imposition. Au contraire pourrait-on dire, car cela a renforcé l'attention portée par les administrations fiscales nationales, légitimement d'ailleurs, à ces problématiques.

2. La suppression de cette contrainte implique des choix clairs dans la construction de l' ACCIS Il faut en effet qu’il y ait une base consolidée sinon, même si une assiette harmonisée était construite, la question de la fixation des prix de cession intragroupe selon le principe de pleine concurrence demeurerait. C'est d'ailleurs ce qu'indique le document de travail de la Commission "Avancement des travaux et programme futur de l'ACCIS" du 20 novembre 2006 (n° 66). C'est également ce qui ressort du tableau des avantages des trois étapes présenté par le Professeur Spengel ce matin: seule la troisième étape résout les problèmes de prix de transfert. De même, l'inclusion des revenus étrangers est indispensable (cf. n° 24 du document précité). Restera la question du seuil de consolidation : le seuil élevé envisagé (75 %, cf. n° 45 du document de la Commission) laisse des sujets entre ce seuil et le seuil caractérisant des transactions entre entreprises associée qui est généralement de 50 % et peut aussi comprendre des liens de dépendance de fait. Enfin, le sujet des prix de transfert a également un impact sur la méthode de répartition de la base imposable entre les Etats Membres : si on se fonde sur la méthode du micro-facteur fondée sur la valeur ajoutée, cela implique de maintenir la fixation de prix de pleine concurrence pour les cessions intragroupe (cf. document de la Commission, n° 71) Sous ces réserves et dans ces conditions (et seulement dans ces conditions), la question des prix de transfert serait réglée par l'ACCIS (et le Forum pourrait disparaître !)

3. Les résultats du Forum pourraient-ils régler les problèmes de prix de transfert ? On vient de voir l'importance du sujet des prix de transfert- et leur rôle dans l'intérêt de l'ACCIS. On peut se demander si les travaux du Forum, à l'inverse, ne pourraient pas régler les problèmes de prix de transfert et donc rendre l'ACCIS moins nécessaire ou, en tout cas, moins urgent.

3.1 Des progrès importants ont été faits Le Forum a permis des avancées significatives en matière de prix de transfert en Europe dans trois domaines depuis sa création en 2002 : En matière de documentation, avec la mise au point dans le cadre du Forum d'un code de conduite instaurant une documentation désignée sous le vocable d'EUTPD

D. Consolidation, Allocation and International Aspects

143

(pour European Union Transfer Pricing Documentation ; là, on a un sigle anglais, mais c’est le délégué français qui l’a proposé au Forum !). En matière d'élimination de la double imposition, avec le code de bonne conduite sur l'amélioration du fonctionnement de la convention européenne d'arbitrage : on observe d'ailleurs que cette significative avancée communautaire que constitue l'arbitrage en matière fiscale a été reprise ces dernières années par l'OCDE qui vient d'aboutir à l'adoption par le Comité des Affaires fiscales en janvier dernier d'une clause d'arbitrage dans le modèle de convention fiscale ; de même, l'avenant à la convention fiscale germano-américaine a introduit une clause d'arbitrage. En matière d'accords préalables en matière de prix de transfert, le Forum a mis au point des lignes directrices que la Présidence allemande devrait faire adopter lors du Conseil des Ministres ECOFIN du 5 juin prochain.

3.2 Mais il faudrait aller plus loin et tout ne peut être réglé par le Forum Pour que le problème des prix de transfert soit suffisamment allégé par rapport à ce que l'ACCIS pourrait apporter, il faudrait que le Forum aille encore plus loin tant en ce qui concerne les obligations documentaires que sur la convention européenne d'arbitrage : sans entrer dans les détails, ce que le temps imparti ne permet pas, il y a encore, sur ces deux sujets, des questions importantes à traiter. En outre, bien sûr, il existe d'autres problèmes qui ne pourront en tout état de cause pas être réglés par le Forum, même si celui-ci allait au bout de ses possibilités : les coût de mise en conformité existeront toujours tant qu'il y aura 27 législations en présence pour l'assiette de l'impôt sur les sociétés. Au terme de ce bref aperçu de la mise en parallèle du Forum sur les prix de transfert et de l'ACCIS, on voit bien que (i) les prix de transfert sont un sujet majeur, à la base même des travaux de l'ACCIS et (ii) que l'ACCIS pourrait les régler parfaitement, mais seulement si elle conduit à une véritable consolidation. En revanche, les travaux du Forum permettent déjà, et pourraient le faire plus encore, de pallier l'absence d'une telle assiette consolidée en réglant, même imparfaitement, l'essentiel des problèmes posés par les cessions intragroupe. Pour reprendre l'expression du Professeur Schreiber, c'est "une solution alternative". C'était d'ailleurs l'idée de départ de la Commission lorsque, dans sa communication du 21 octobre 2001, elle a lancé des mesures ciblées, dont le Forum et, en parallèle, des approches globales. Ce sera d'autant plus le cas que le Forum pourra aller loin dans les solutions qu'il proposera.

144

Michael Lang

3. Michael Lang

Herr Vorsitzender, meine sehr verehrten Damen und Herren. Sie haben die Frage nach der Zukunft der Doppelbesteuerungsabkommen in einem CCCTB-Regime angesprochen. Ich darf bei den nächsten Überlegungen, die ich in den nächsten Minuten anstellen darf, davon ausgehen, dass ein CCCTB-Regime – wie immer es im Detail aussehen wird – auch Außengrenzen, also einen bestimmten territorialen Anwendungsbereich haben wird. Das bedeutet aber natürlich auch – und dessen muss man sich bewusst sein –, dass Unternehmen, die innerhalb der CCCTB-Länder und außerhalb der CCCTB-Länder tätig sind, sich mindestens zwei verschiedenen Systemen stellen werden müssen. Die CCCTB-Regelungen werden das nicht lösen. Es wird also nach wie vor Schwierigkeiten bei der Rechtsanwendung geben und es wird nach wie vor auch Gestaltungsmöglichkeiten geben, weil man auf diese Weise – unabhängig davon, ob man für das CCCTB Regime optieren wird können oder nicht – jedenfalls faktisch die Möglichkeit hat, aus dem CCCTB-Regime zu flüchten und letztlich herauszuoptieren. Das Thema Außengrenze stellt sich natürlich auch insoweit, als die Mitgliedstaaten der EU völkerrechtliche Verpflichtungen eingegangen sind und es derzeit ein sehr dichtes Netz an Doppelbesteuerungsabkommen mit Drittstaaten gibt, aber genauso innerhalb der Europäischen Union. Das heißt, es gibt auch in Hinblick auf die DBA zumindest zwei Rechtskreise. Herr Vorsitzender, Sie haben auch noch die zusätzliche Möglichkeit angesprochen, dass eben nicht alle EU-Staaten bei einem CCCTB-Regime mitmachen. Das bedeutet aber, dass man zwischen drei Kreisen unterscheiden muss: Die CCCTB-Staaten, die sich dem System anschließen, bilden die erste Gruppe von Staaten. Zur zweiten Gruppe gehören die EU-Mitgliedstaaten, die außerhalb dieses Regimes stehen und die dritte Gruppe sind dann die tatsächlichen Drittstaaten. Die Arbeitsgruppe und die Subarbeitsgruppe, die sich mit internationalen Aspekten beschäftigt hat, hat meines Erachtens hervorragende Arbeit geleistet, weil eine Vielzahl von Problemen in beeindruckender Weise aufgelistet und genannt ist. Die Probleme, die sich aus dem Zusammenspiel einer CCCTB-Regelung mit verschiedenen Doppelbesteuerungsabkommen ergeben, sind intellektuell höchst herausfordernd und äußerst spannend. Dennoch, oder gerade deshalb, möchte ich aber dafür plädieren, bei der Lösung dieser Fragen einen pragmatischen Ansatz zu verfolgen. Ich glaube, man muss hier einfach auch den Mut zur Unvollkommenheit haben, wenn man zu einem Ergebnis gelangen will. Wenn man darauf wartet, bis all die sich stellenden Probleme intellektuell durchdrungen sind, und gleichzeitig bereits auch auf den Abschluss neuer, auf das CCCTB Regime hin adaptierter Doppelbesteuerungsabkommen, wird man die CCCTB-Regelungen wahrscheinlich nie in Kraft setzen. Daher plädiere ich sehr stark für einen step-by-step-approach und fühle mich auch dadurch bestätigt, weil heute ja auch bei etlichen anderen Diskussionsbeiträgen schon deutlich hervorgekommen ist, dass das CCCTB-Projekt, wenn es ein erfolgreiches Projekt sein wird, auch kein Projekt sein wird, das zu einem

D. Consolidation, Allocation and International Aspects

145

bestimmten Tag abgeschlossen sein wird, sondern ein Projekt, das ohnehin laufend weiterentwickelt werden wird. Denken Sie nur daran, dass Herr Schreiber uns darauf aufmerksam gemacht hat, dass es möglicherweise notwendig sein wird, CFCähnliche Regelungen im Verhältnis zu Drittstaaten zu konstruieren. Solche Regelungen tragen es ja immanent in sich, dass sie immer wieder, wenn sie einmal existieren, weiterentwickelt werden müssen. Wir brauchen also ohnehin Instrumentarien, um das CCCTB-Regime, wenn es einmal existiert, weiter zu entwickeln. Daher kann man es auch umgekehrt angehen und einen gewissen Mut zur Unvollkommenheit dabei haben und einmal mit einem step-by-step-approach beginnen. Allerdings glaube ich, dass sich die Probleme, die sich dabei stellen, unterschiedliches Gewicht haben, sowohl rechtlich als auch politisch. Was einmal den ersten Kreis an Staaten betrifft, den Kreis der Mitgliedstaaten der EU, die sich an einem CCCTB-Regime beteiligen, sehe ich nicht das große Problem, weil hier, zumindest im Anwendungsbereich der CCCTB-Regelungen, die Doppelbesteuerungsabkommen weitgehend obsolet für Zwecke der CCCTB werden können. Dies ist weder völkerrechtlich noch gemeinschaftsrechtlich ein Problem, weil die CCCTB-Regelungen Bestandteil des Gemeinschaftsrechts wären und Gemeinschaftsrecht auch gegenüber den Doppelbesteuerungsabkommen, die zwischen diesen Mitgliedstaaten geschlossen wurden, Vorrang hat. In ähnlicher Weise würde ich das auch für die zweite Gruppe von Staaten sehen, die zwar zur EU gehören, die sich aber an den CCCTB-Regelungen nicht beteiligen. In Hinblick auf diese Staaten besteht das Bedürfnis, dass Einkünfte, die Gesellschaften, die den CCCTB-Regelungen unterliegen, aus diesen Staaten erhalten oder in diese Staaten zahlen, einheitlich behandelt werden. Die bestehenden Doppelbesteuerungsabkommen divergieren diesbezüglich nach wie vor. Durch gemeinschaftsrechtliche Regelungen könnten aber die DBA insoweit „overruled“ werden. Dies würde allerdings voraussetzen, dass auch die EU-Staaten, die sich nicht den CCCTB-Regelungen unterwerfen, gewillt sind, an der Erlassung derartiger Regelungen – z.B. in Richtlinien- oder Verordnungsform – mitzuwirken. Inhaltlich könnten sich derartige Regelungen, die für CCCTB-Gesellschaften an die Stelle der DBA zwischen den CCCTB-Staaten und den anderen EU-Staaten treten, an den Musterabkommen der OECD orientieren. Ein weiterer Orientierungspunkt könnten die bestehenden Richtlinien sein, die z.B. im Konzernverhältnis schon jetzt eine Quellensteuerfreiheit für Dividenden, Zinsen und Lizenzgebühren gewährleisten. Diese Regelungen könnten auch für CCCTB-Gesellschaften anwendbar gemacht werden. Komplizierter wird es mit Sicherheit im Verhältnis zu den Drittstaaten, weil hier sind die Mitgliedstaaten völkerrechtliche Verpflichtungen eingegangen, die man sicher nicht über Bord werfen oder negieren sollte, und die man auch nicht von einem Tag auf den anderen oder mit einem Schlag abändern kann. Hier muss man einfach in Kauf nehmen, dass diese Doppelbesteuerungsabkommen im Detail vom Musterabkommen der OECD abweichen. Dies betrifft etwa den Begriff der Betriebsstätte, der mitunter durchaus unterschiedlich in verschiedenen Doppelbesteuerungsabkommen definiert ist und hier für die Verteilung der Besteuerungsrechte natürlich Kernregelungen enthält und der gleichzeitig die Außengrenze eines

146

Michael Lang

CCCTB-Regimes mit definiert. Wenn ich also vorher vom Mut zur Unvollkommenheit gesprochen habe, dann meine ich unter anderem damit, dass man zumindest für einen bestimmten Zeitraum auch akzeptieren wird müssen, dass es unterschiedliche Doppelbesteuerungsabkommen gibt und dass daher die Außengrenze je nach Mitgliedstaat in Teilbereichen unterschiedlich verläuft. Das ist sicher ein Kurzfristszenario. Parallel dazu sollte man aber auch anstreben – und das wäre, wenn man sich einem CCCTB-Regime nähert, für mich der konsequente Schritt –, dass die Europäische Union dann selbst mit Drittstaaten Doppelbesteuerungsabkommen verhandelt, die die Außengrenze des CCCTB-Regimes einheitlich festlegen. Kompetenzrechtlich würde ich hier kein Problem sehen, wenn wir innerhalb der Gemeinschaft einmal CCCTB-Regelungen haben. Dann können wir aufgrund der Rechtsprechung des Europäischen Gerichtshofs davon ausgehen, dass es eine Außenkompetenz gibt und dass dann entweder ausschließlich oder zusätzlich zu den Mitgliedstaaten die Europäische Union auch Doppelbesteuerungsabkommen verhandeln könnte. Dies müssten ja keineswegs umfassende Doppelbesteuerungsabkommen sein. Dies könnten völkerrechtliche Verträge sein, die die zentralen Regelungen beinhalten, die die Außengrenzen der CCCTB regeln und die den unternehmerischen Bereich betreffen und zusätzlich zu den bestehenden Doppelbesteuerungsabkommen der Mitgliedstaaten treten. Ich würde hier mittel- und langfristig also nicht für ein multilaterales Abkommen plädieren, sondern für Abkommen, die von der Europäischen Union abgeschlossen werden. Dies macht auch politisch durchaus Sinn: Schauen wir nur in den historischen Rückspiegel und erinnern wir uns, was Anfang der 90er Jahre passiert ist, als die europäischen Staaten mit den USA die neuen Doppelbesteuerungsabkommen verhandelt haben. Damals war jeder Staat für sich in einer sehr schwachen Position und etliche der Regelungen hätten anders und einheitlicher ausgeschaut, wenn hier die EU als solche den USA gegenüber getreten wäre. Ich würde auch hier zu einem pragmatischen Ansatz raten, und glaube, dass man auch beim Inhalt eines derartigen Abkommens vielleicht differenzieren könnte, was die CCCTB betrifft. Die CCCTB stellt ja eine Harmonisierung der Bemessungsgrundlage dar, weshalb Regelungen, die die Bemessungsgrundlage betreffen, auch von oberster Priorität sind. Hier geht es einerseits um die Definition der Betriebsstätte, die wohl enthalten sein müsste, und andererseits um Regelungen für die Unternehmensgewinne. Die weitere Entwicklung hängt vermutlich stark davon ab, wie weit die OECD hier ihre Grundsätze verändert. Ich sehe ein bisschen die Sorge, dass es hier zu Unterschieden zwischen den Mitgliedstaaten kommen wird, wenn die OECD weiter den separate entity approach betont, da nicht alle Mitgliedstaaten diesem approach dann möglicherweise in der Praxis bei ihren bilateralen Abkommen folgen werden. Dieses rechtliche Problem wird sich aber wahrscheinlich in jedem Fall stellen: Wenn die OECD ihre Grundsätze in Richtung separate entity approach weiterentwickelt, stehen wir in der Europäischen Union ja auch vor dem Problem, dass wir materielles Verrechnungspreis- und Gewinnabgrenzungsrecht in der Schiedskonvention haben und sich daher dann ohnehin die Frage stellt, was für die Betriebsstättengewinnermittlung und die Verrechnungspreisgewinnermittlung gilt, wenn die OECD das Musterabkommen weiterentwickelt. Geänderte Regelungen des OECD-

D. Consolidation, Allocation and International Aspects

147

Musterabkommens oder gar ein geänderter OECD-Kommentar können jedenfalls keine unmittelbaren Wirkungen für die Anwendung der materiell-rechtlichen Regelungen der Schiedskonvention entfalten. Diese Fragen stellen sich uns in der Europäischen Union also vermutlich ohnehin, mit oder ohne CCCTB. Nicht alle DBA-Regelungen sind von erster Priorität und müssten gleich in einem ersten Schritt vereinheitlicht werden. Die CCCTB-Regelungen sollen ja nicht den Steuersatz betreffen. Der Steuersatz ist aber für die Auswirkungen der Methoden zur Vermeidung der Doppelbesteuerung von Bedeutung. Daher würde ich – wiederum sehr pragmatisch – zumindest in der Anfangszeit – akzeptieren, dass es – je nach Mitgliedsstaat – unterschiedliche Regelungen im Verhältnis zu Drittstaaten gibt. Man muss sich aber bewusst machen, dass – was die Administration betrifft – für alle Beteiligten natürlich die Freistellungsmethode erhebliche Vorteile gegenüber der Anrechnungsmethode hat.

148

Ivar Nordland

4. Ivar Nordland

I am going to speak about the Danish development and our view on the common corporate tax base. Denmark is one of the few countries with cross border consolidation and for sure we are the country with the longest history of cross border consolidation. Before the First World War the company law was not very developed. This had effects on the taxation of companies within a group. The question was whether the group should be taxed as one unit or as separate entities. In 1936 the Danish Supreme Court stated, that if the companies acted as an economic unit, they should be taxed as an economic unit by adding the results of the companies. In 1943 the Supreme Court stated that even a Swedish subsidiary could be included in the group. At that time absolutely nobody were allowed to go to neutral Sweden to see what actually happened there. The Danish tax authorities then gave up, and this is how Denmark got a cross border loss relief system without involving the politicians. The Commission could be tempted by this approach. To some extent the European court of Justice has dealt with the issue in the famous Marks- and Spencer decision. The Danish system where the parent company had the opportunity to chose which foreign companies it would like to include under Danish taxation had the effect that Danish multinationals through appropriate planning decided what amount of Danish tax they would like to pay. Loss making subsidiaries where included in the consolidation while profit making where not. When the loss making companies began to be profitable, they could leave the consolidation without any consequences – no recapture of deducted losses, no exit taxes, no nothing. Of course in the long run this was not a sustainable situation, and 1992 a recapture mechanism was introduced. Since then deducted losses were added back when a foreign company left the consolidation. The provisions for recapture have been full of loopholes, and have been amended several times. In 2005 Denmark abolished taxation of foreign PE ´s as well as foreign subsidiaries unless the ultimate Parent of the group chose to include all group companies and activities under Danish taxation. This is what we call the principle of territoriality and the all in/all out principle. The group is defined in accordance with the international standards for accounting. The abolishment of the cherry-picking element gave revenue to lower the company tax rate from 30% to 28%. And right now we are dealing with a bill in the Danish parliament limiting the deduction of interest for multinationals. The ceiling for deduction will be a standard interest rate calculated on a maximum debt corresponding to the tax value of assets involved in creating taxable income in Denmark unless the group has chosen international consolidation. Shares and self-generated goodwill is disregarded when the maximum interest deduction is calculated. This counteracts the tax planning of especially the Private Equity funds and allows a further reduction of the company tax rate to 25%.

D. Consolidation, Allocation and International Aspects

149

Now how does this relate to the Common consolidated corporate tax base? It is hard to imagine, that a small country like Denmark can have two parallel company tax systems with two different methods of cross border consolidation. As you have heard Denmark has lowered the tax rate and broadened the tax base with elements that are relevant the design for the future CCCTB, for example the all in/ all out principle and the principle of territoriality. If these elements are not found in the future CCCTB what should Denmark do? Raise the corporate income tax rate to 30% again? Another question is to what extent the CCCTB group will be treated as one single company instead of separate entities? The abolishment of transfer pricing between group members would lead to the first conclusion. One thing is sure; it should not be possible for the EU multinationals to choose when they would like to be treated as one unit and when they would like to be treated as separate entities. They cannot have the cake and eat it too. For the time being we don’t have the answers to all these questions and we are looking forward to see the proposal from the Commission.

150

Walter Hellerstein

IV. Lessons of US Subnational Experience for EU CCCTB Iniative Walter Hellerstein

I am extremely grateful for the opportunity to share with this most distinguished audience my thoughts on the lessons that the US subnational experience may have for the EU CCCTB initiative. My remarks are based on 35 years of writing about – and, more importantly, working in the trenches of – state tax practice in the US. Let me begin with the obligatory “Lost in Translation” caveat, namely, that there are significant differences between the US and EU contexts that should make one wary about overstating the “lessons” (whether negative or positive) that the US experience may hold for the EU CCCTB initiative. It goes without saying that the EU Member States are sovereign nations and enjoy true tax sovereignty with respect to direct taxation (subject to the broad restraints that the EC Treaty imposes on freedom of establishment and freedom of movement of goods, services, workers and capital). This sovereignty is manifested in the requirement of unanimous approval of Member State decisions affecting taxation as well as the subsidiarity principle that no decision should be made at the EU level that can be better made by the Member States. By contrast, the US subnational states are ultimately subject to the will of the US Congress on matters that substantially affect interstate or international commerce, including the exercise of their tax power, although it is worth observing that Congress has rarely exercised this power and the states have generally been permitted to go their own way on matters of taxation. Other noteworthy contextual differences between the US subnational and EU corporate tax regimes include the existence in the US of the federal (national) corporate income tax that virtually all states follow in defining the corporate tax base and the fact that state corporate tax rates are modest (ranging from 6%-10%), and are deductible from the federal corporate tax base, whereas EU rates are considerably higher (ranging from 10% to 40%). Notwithstanding these contextual differences, there are, in my judgment, three categories of lessons that the US experience holds for the EU CCCTB initiative: historical lessons; political lessons; and technical lessons. As my ensuing remarks will reveal, these categories are hardly airtight, but they are useful organizing principles.

1. Historical lessons I turn first to historical lessons. One of America’s most distinguished jurists, Oliver Wendell Holmes, once said: “A page of history is worth a volume of logic.” I believe this insight is particularly pertinent to the question as to whether the EU is on the right track in seeking to replace the arm’s-length/separate accounting (AL/SA) method for attributing income to EU Member States with formulary apportionment (FA). Whatever one may say as a matter of theory about the relative merits of AL/ SA vis-à-vis FA as a method of income attribution, history strongly suggests that in

D. Consolidation, Allocation and International Aspects

151

a truly integrated economy encompassing independent taxing jurisdictions, transfer pricing between taxpayers engaged in cross-border transactions – the backbone of an AL/SA regime – is difficult if not impossible to administer in a cost effective manner and, as a consequence, the adoption of some form of FA is virtually inevitable. This may be the single most important lesson from the US experience. Because of the highly integrated nature of the US subnational “common market,” the US states recognized early on that an AL/SA approach to income attribution was not an administratively feasible mechanism for dividing income earned by multistate corporations among the states. They therefore adopted FA, not because of its theoretical superiority to AL/SA – a point of continuing controversy that I do not address here – but rather because of practical necessity. Although my remarks are directed to the US experience, I believe my conclusion is reinforced by the historical experience of other highly integrated economies with independent taxing units where the same point emerges. Both the Canadian provinces and the Swiss cantons employ FA for attributing income to the appropriate subnational taxing jurisdiction. I would argue that this reflects the same practical administrative considerations that resulted in the US states’ adoption of FA. In short, the key historical lesson from the US and similar common markets is that true economic integration of independent taxing units creates enormous (and perhaps inexorable) pressure for the adoption of a formulary methodology for assigning income among taxing units within the relevant common market. If the EU Member States economies are in the process of achieving a level of economic integration approaching that of the internal US market, the replacement of AL/SA with FA may well be inevitable as a practical matter.

2. Political lessons I turn next to political lessons that the US experience may hold for the EU. At the highest level of analysis, the principal lesson – which is more the statement of the obvious than a “lesson” – is that the key decisions regarding the base, the group, and the apportionment mechanism for a CCCTB are substantially influenced by political as distinguished from technical concerns. In other words, there are a range of technically feasible choices for the base, the group, and the apportionment mechanism, and the final choice among them will be driven to a substantial extent by political considerations. Two examples from the US experience – one positive (the choice of base) and one negative (the choice of apportionment mechanism) – illustrate this point. In the US, political realities plainly determined the choice of a subnational corporate tax base. Indeed, as a practical matter, there never really was a choice. The existence of a federal (national) corporate income tax base created overwhelming political pressure on the individual states generally to conform their bases to the federal model. To be sure, the states in the absence of a federally mandated base are free to deviate from the federal norm and have occasionally done so with regard to particular items (e.g., accelerated depreciation) when the revenue consequences of adopting the federal rule are unacceptable. Nevertheless, the political pressure from

152

Walter Hellerstein

taxpayers for ease of compliance and the real benefits to the states that conformity offered from an administrative and information-sharing standpoint have largely foreclosed any widespread deviation from the federal template. With respect to the choice of the apportionment mechanism, however, the political lesson is more of a cautionary tale. During the first half of the twentieth century the states used a wide variety of divergent apportionment formulas, before settling on a three-factor formula of property, payroll, and receipts (sales). The consensus was embodied in the model (nonbinding) Uniform Division of Income for Tax Purposes Act, which many states adopted legislatively. The consensus – in particular the inclusion of the receipts factor – reflected a political compromise that assigned income both to the states where labor and capital was located and to the states where the market for goods existed. Indeed, by 1978 the consensus was so strong that 44 of the 45 states with income taxes employed the three-factor formula, and there was a widely shared perception that any deviation from the three-factor formula was constitutionally impermissible. But then a funny thing happened. In 1978, in a constitutional challenge to Iowa’s single-factor receipts formula, which many thought the U.S. Supreme Court would condemn as a “rogue” provision inconsistent with the requirements of a national common market, the Court sustained the formula in Moorman Manufacturing Co. v. Bair [437 U.S. 267 (1978)]. The Court rested its decision largely on the ground that expressing a constitutional preference for the three-factor formula would involve extensive judicial lawmaking and that Congress, not the Court, was the appropriate body to fashion such rules. Whatever one may say about the merits of the Court’s decision in Moorman, the consequences were dramatic. Freed from the perceived constitutional requirement that they adhere to the equally weighted three-factor formula, since 1978 the states have shown little restraint in jettisoning that formula in favor of formulas weighted heavily, if not exclusively, on receipts. Today roughly three-quarters of the states with corporate income taxes place at least half the weight on receipts and close to a dozen have moved or are in the process of moving towards a formula based entirely on receipts. The explanation for this headlong rush towards a more heavily weighted receipts factor lies in concerns about “economic development,” not on concerns about an accurate determination of the source of income. Reducing the importance of capital and labor in the formula and emphasizing the importance of receipts creates a more attractive environment for business location and expansion in a state. The political lesson of the US states experience with the apportionment mechanism is therefore clear. If left to their own devices, without a real or perceived limitation on their power to deviate from a uniform formula, the political pressure to adopt a formula that serves a state’s individual economic interests in attracting business investment is likely to overwhelm any countervailing concerns regarding uniformity or adoption of a formula reflecting accepted norms of tax policy. In short, the lesson for the EU, which it already appears to have embraced as one of its operating assumptions, is that whatever apportionment mechanism the CCCTB adopts will be uniform among Member States and, consistent with technical con-

D. Consolidation, Allocation and International Aspects

153

cerns, will incorporate factors that fairly and accurately reflect income-producing activity within Member States.

3. Technical lessons This brings me to the technical lessons that the US experience may hold for the EU. I will simply state these briefly, because this is not the time or the place for a detailed technical discussion, which I have undertaken elsewhere. First, with respect to the determination of the appropriate group for purposes of consolidation, the US experience with the “unitary business” principle, though sound from an economic standpoint and an excellent source of income for tax lawyers, is administratively complex. A standard based on ownership, perhaps with some mechanism to avoid abusive tax planning, is likely to strike a better balance between theoretical and practical considerations. Second, with respect to the choice of a formula, two of the factors – property and payroll – have given rise to relatively little controversy over the years in the US and generally seem appropriate measures of economic activity, as they reflect capital and labor. The principal issue with respect to the property factor is whether to include intangible property in the factor and, if so, how. This is a difficult question, which once again pits theoretical against practical concerns. As a matter of principle, intangible property should be included in the factor because of the important contribution (at least in many instances) that intangible property makes to the production of a taxpayer’s income. As a practical matter, however, there is a serious problem with including intangible property in the factor. After all, the purpose of an apportionment formula is to identify geographically determinate factors (e.g., the location of tangible property or employees) for purposes of assigning a geographically indeterminate item, namely, income. But intangible property, which by definition has no meaningful physical location, is itself geographically indeterminate. It makes little sense to use one geographically indeterminate factor for purposes of determining the geographic location of another geographically indeterminate factor. Indeed, including intangible property in the property factor simply raises the question of where intangible property is located. Based on these considerations, it may well be that the best solution is the pragmatic one reflected in the US experience, to wit, exclude intangible property from the factor and let the “tangible” factors be the basis for income attribution. This approach becomes more acceptable if one has adopted special apportionment regimes, which I address in a moment, for certain industries where intangibles are particularly important (e.g., financial institutions) and which do reflect intangible property in the factors. By contrast to the property and payroll factors, the receipts factor has given rise to considerable controversy in US state tax practice. The theoretical case for the receipts factor generally rests on the desire to reflect the market for the taxpayer’s goods. The practical problems raised by the receipts factor include the difficulties in implementing it with respect to services and intangibles as well as issues raised by determining the destination of sales. Perhaps the principal lesson from the US regarding the receipts factor if the CCCTB includes such a factor is the desirability

154

Walter Hellerstein

of addressing these known problems legislatively at the outset rather than through years of litigation. Third, the US experience suggests that it would be wise to consider special apportionment regimes for specific industries. The US experience reveals that a generalized formula like the Uniform Division of Income for Tax Purposes Act works awkwardly at best for many industries. When an industry’s factors do not fit neatly into the generalized formula (as is the case with financial institutions, relying heavily on intangibles, or transportation, involving receipts without an obvious location and mobile property and employees), a specialized formula reflecting these specialized factors makes good theoretical and practical sense.

4. Conclusion In conclusion, I would like to stress what may be the overriding lesson from the US experience – and it is a quintessentially American theme – namely, that assignment of income in the end is a pragmatic exercise. In my view, the debate over the theoretical superiority of AL/SA versus FA is inconclusive. What is not in doubt, however, regardless of which side of the debate one favors, is that there is no scientifically precise methodology for assigning income among taxing jurisdictions in a highly integrated economy. Even the OECD, the bastion of AL/SA, appears to recognize this in its approach to global trading operations and its apparent endorsement of profit-split methodologies at least in some circumstances. Once one recognizes that the use of transfer pricing to attribute the income earned by multijurisdictional corporate affiliates to taxing jurisdictions in a highly integrated economic environment enjoys at best the illusion of scientific precision, we can focus on what in my judgment should be the key question in the debate: Whether, from a practical standpoint, AL/SA or FA does a better job at attributing income and at what cost. The principal lesson from the US on this fundamental question is that FA simply works better and at a lower cost, although it by no means works perfectly. Perhaps the other lessons from the US experience are that FA could work a lot better, and the EU has an opportunity to do just that in designing the CCCTB.

E. Administrative Aspects

155

E. Administrative Aspects I. Administrative Aspekte einer Gemeinsamen Konsolidierten Körperschaftsteuerlichen Bemessungsgrundlage in der EU Ernst Czakert

1. Einleitung Die Europäische Union hat sich mit dem Projekt zur Einführung einer Gemeinsamen Konsolidierten Körperschaftsteuerlichen Bemessungsgrundlage (GKKB) ein ehrgeiziges Ziel gesetzt. Die Kommission hat sich selbst die Verpflichtung auferlegt, im Jahr 2008 einen entsprechenden Richtlinienvorschlag vorzulegen. In diesem Richtlinienvorschlag werden die Regeln für die Gewinnermittlung von körperschaftlich organisierten Unternehmen enthalten sein; die Vorschriften zur Definition einer Unternehmensgruppe werden beschreiben, welche Unternehmen einen konsolidierten Gewinn zu versteuern haben und schließlich wird der Mechanismus definiert, wie die konsolidierte Bemessungsgrundlage auf die beteiligten EU-Mitgliedstaaten zu verteilen ist. Ein wesentlicher Teil der Unternehmensbesteuerung in der EU wird damit einander angeglichen. Es werden die Regeln definiert, was genau eine Unternehmensgruppe mit Töchtern bzw. Betriebsstätten in verschiedenen Mitgliedstaaten innerhalb der EU in jedem beteiligten Mitgliedstaat zu versteuern hat. Diese Regeln müssen von jedem beteiligten EU-Mitgliedstaat auf gleiche Art und Weise angewendet werden. Da es um grenzüberschreitende Sachverhalte geht, sind Entscheidungen nur möglich wenn die notwendigen Informationen, auf denen die Entscheidung beruht, aus den beteiligten Mitgliedstaaten zusammengeführt werden. Die GKKB kann demnach nur gemeinsam von den EU-Staaten umgesetzt werden. Die EU-Mitgliedstaaten müssen sich deshalb auf gemeinsame administrative Regeln verständigen, denn die steuerpflichtigen Unternehmensgruppen brauchen klare Vorgaben, welche Behörde für das Besteuerungsverfahren verantwortlich ist, und es muss eindeutig bestimmt sein, auf welche Art mögliche Einsprüche oder gerichtliche Verfahren durchzuführen sind. Schließlich müssen sich die beteiligten EU-Mitgliedstaaten darauf verlassen können, dass der ihnen gebührende Anteil des gemeinsamen Gewinns richtig ermittelt und festgesetzt wird. Die administrativen Aspekte einer GKKB sind deshalb ein elementarer Bestandteil für ein funktionierendes System. In diesem Beitrag wird versucht darzustellen, wie ein solches System aufgebaut sein könnte.

156

Ernst Czakert

2. Derzeitige Situation Zur Zeit werden in der EU 27 verschiedene Systeme bei der steuerlichen Gewinnermittlung verwendet. Diese Systeme beschränken sich in ihrem Anwendungsbereich auf das Gebiet des jeweiligen Mitgliedstaats und sind trotz der prinzipiellen Ähnlichkeiten in ihren Ergebnissen unterschiedlich. Die Steuerverwaltungen, die diese unterschiedlichen Systeme anwenden, haben sehr verschiedene Traditionen und Arbeitsweisen, sodass auch in den Bereichen, in denen die Regeln gleich sind, aufgrund dieser Differenzen die Ergebnisse oft nicht miteinander harmonieren. Gruppenbesteuerungssysteme sind nicht in allen EU-Mitgliedstaaten vorhanden und nur wenige Mitgliedstaaten erlauben eine grenzüberschreitende Gruppenbildung.1 Rechtsbehelfsverfahren sind ebenfalls durch nationale Regeln und Traditionen geprägt. Ein in allen Mitgliedstaaten der EU tätiger Unternehmensverbund muss sich im Extremfall deshalb mit 27 verschiedenen Rechtssystemen auseinandersetzen, was besonders bei kleinen und mittleren Unternehmen eine nicht zu vernachlässigende Kostenbelastung bedeutet.

3. Angestrebte Situation Gemeinsame europäische Regeln für die Ermittlung des steuerlichen Gewinns eines Unternehmensverbunds in der EU bedingen eine intensive und geregelte Zusammenarbeit der Verwaltungen der EU-Mitgliedstaaten. Von der Definition eines Unternehmensverbundes, über das Besteuerungsverfahren und ein eventuell sich anschließendes Rechtsbehelfsverfahren oder eine Betriebsprüfung, bis zu Entscheidungen über den Ein- oder Austritt eines Unternehmens in einen oder aus dem Konsolidierungskreis müssen Entscheidungen auf Seiten der Verwaltung einheitlich und verbindlich für alle beteiligten Mitgliedstaaten getroffen werden. Entscheidungsgrundlagen sind dabei in allen beteiligten Mitgliedstaaten vorhanden und deshalb können Entscheidungen auch nur im Zusammenwirken aller beteiligten Verwaltungen getroffen werden. Einheitliche Regeln müssen von den beteiligten Verwaltungen auch einheitlich ausgelegt und angewendet werden und es ist erforderlich bei Differenzen in der Auslegung oder Anwendung von Regeln eine Instanz zu bestimmen, die verbindlich für alle Beteiligten eine einheitliche Entscheidung zu treffen hat. Bei der Durchführung von Rechtsbehelfsverfahren muss dafür gesorgt werden, dass zum einen die letztgültige Entscheidung für alle Beteiligten verbindlich ist und zum anderen die Steuerpflichtigen, egal in welchem Mitgliedstaat sie ein Rechtsbehelfsverfahren initiieren, überall die gleichen Möglichkeiten und Rechte haben.

4. Verwaltungsstrukturen und Abläufe Die derzeitigen Verwaltungsabläufe und organisatorischen Strukturen der Steuerverwaltungen in den Mitgliedstaaten werden diesen Zielen nicht gerecht. Dies ist nicht verwunderlich, da die Tätigkeit der Verwaltungen im Wesentlichen auf natio1

Wesentlicher Grund ist die damit verbundene Akzeptanz einer grenzüberschreitenden Verlustberücksichtigung.

E. Administrative Aspects

157

nalen Regelwerken und deren Umsetzung beruht. Auch die organisatorischen Strukturen sind oftmals sehr unterschiedlich, was sich aus der unterschiedlichen Geschichte und den verschiedenen Traditionen der Steuerverwaltungen erklärt. Die deutsche Steuerverwaltung ist föderal organisiert; die Steuerverwaltung ist Aufgabe der Länder, weshalb, um eine einheitliche Anwendung des Steuerrechts zu gewährleisten, eine Fülle von steuerlichen Fachgruppen unter dem Vorsitz des Bundes unter Beteiligung aller Länder dafür sorgt, dass das deutsche Steuerecht in den Ländern einheitlich interpretiert und einheitlich umgesetzt wird. Im Gegensatz dazu ist die Steuerverwaltung in z.B. Frankreich traditionell zentral organisiert. Hier ist das zuständige Ministerium in Paris für die Gesetzgebung und für die Verwaltung im Bereich der Besteuerung zuständig. Dies sind nur zwei Beispiele, denen man in der EU noch 25 weitere befügen kann. Innerhalb der EU besteht aber bereits jetzt der Zwang zur Zusammenarbeit und die EU hat für die Zusammenarbeit der Mitgliedstaaten untereinander auch Instrumente geschaffen. Es ergibt sich die Frage, ob auf Basis der vorhanden Instrumente ein System wie die GKKB administrierbar sein könnte.

5. Bestehende Verfahren 5.1 Amtshilferichtlinie2 Die Amtshilferichtlinie regelt den Auskunftsverkehr der Mitgliedstaaten der EU untereinander in Steuersachen. Insbesondere sollen damit die internationale Gewinnverlagerung von Konzernen und Steuerumgehungen verhindert werden. Für eine wirksame Zusammenarbeit der Steuerverwaltungen untereinander müsste es die Richtlinie vom Grundsatz ermöglichen, dass ein Austausch von Informationen zwischen einem Finanzamt in Ungarn und einem Finanzamt in Deutschland genauso schnell und effizient erfolgen würde, wie der Austausch von Informationen zwischen Ämtern in München und Hamburg. Dies ist aber nicht der Fall. Auskunftsersuchen werden nicht direkt zwischen den Beteiligten Behörden abgewickelt sondern über die jeweiligen obersten Dienststellen. Auskunftsersuchen sind aufwendig und langsam, im Durchschnitt dauert die Beantwortung einer Anfrage länger als ein Jahr. Die Amtshilferichtlinie müsste deshalb dringend verbessert werden. So sollte zum Beispiel die automatische und spontane Auskunftserteilung so gestaltet werden, dass alle Mitgliedstaaten bei der Steueraufsicht besser miteinander kooperieren könnten und Betriebsprüfungen müssten auch auf dem Territorium eines anderen EU-Staates möglich sein. Auf Basis der bestehenden Amtshilferichtlinie lässt sich ein System wie die GKKB nur schwerlich administrativ durchführen.

2

Richtlinie 77/799/EWG des Rates vom 19. Dezember 1977 über die gegenseitige Amtshilfe zwischen den zuständigen Behörden der Mitgliedstaaten im Bereich der direkten Steuern und der Steuern auf Versicherungsprämien und zuletzt Richtlinie 2004/56/EG des Rates vom 21. April 2004 zur Änderung der Richtlinie 77/799/EWG über die gegenseitige Amtshilfe zwischen den zuständigen Behörden der Mitgliedstaaten im Bereich der direkten Steuern, bestimmter Verbrauchsteuern und der Steuern auf Versicherungsprämien.

158

Ernst Czakert

5.2 Beitreibungsrichtlinie3 Die Beitreibungsrichtlinie erlaubt es, deutsche Steuerforderungen in anderen Mitgliedstaaten der EU zu vollstrecken. Die den Forderungen zugrunde liegenden Steuerbescheide dürfen dabei nicht älter als 5 Jahre sein. Ein Problem ist, dass für eine Vollstreckung in einem anderen Mitgliedstaat der Aufenthalt des Vollstreckungsschuldners und der Ort seines Vermögens bekannt sein müssen. Auch werden in einzelnen Mitgliedstaaten behördliche Vollstreckungstitel nicht anerkannt. Insgesamt ist die Erfolgsquote bei der Beitreibung von Steuerforderungen im EU-Ausland mit lediglich 1% der beizutreibenden Forderungen außerordentlich gering.4

5.3 Entscheidungsmechanismen Ein gemeinsames System wie die GKKB beruht auf gemeinsamen EU-rechtlichen Vorgaben die in den EU-Mitgliedstaaten einheitlich interpretiert und umgesetzt werden müssen. Bei 27 EU-Staaten und deren unterschiedlichen Verwaltungspraktiken, die sich aus der jeweils eigenen Tradition und Geschichte ergeben, muss sichergestellt sein, dass gemeinsame Regeln nicht unterschiedlich umgesetzt werden. Derzeit gibt es im Bereich der direkten Steuern auf Ebene der Kommission die Arbeitsgruppe IV, in der Probleme besprochen werden können. Diese Gruppe tagt unter dem Vorsitz der Kommission unter Beteiligung von Delegierten aller Mitgliedstaaten. Verbindliche Entscheidungen werden dort nicht getroffen. Die Arbeitsgruppe dient der Kommission zur Diskussion und Abklärung von Lösungsmöglichkeiten bei Problemen bei den direkten Steuern. Soll ein Beschluss verbindlich sein, muss er vom Rat gebilligt werden. Auch Richtlinien können nur vom Rat auf Basis eines Vorschlags der Kommission verabschiedet werden. Im Rat gibt es dazu die Gruppe Steuerfragen, bei der eine Untergruppe sich mit Fragen zu den direkten Steuern beschäftigt. Ist sich diese Gruppe in Bezug auf eine Frage einig, dann entscheidet endgültig der Rat der Wirtschafts- und Finanzminister, ECOFIN, über die Annahme eines Beschlusses oder einer legislativen Regelung. In steuerlichen Fragen entscheidet der Rat einstimmig. Es genügt das Nein eines Mitgliedstaates, um eine Maßnahme zu verhindern. Diese Art der Entscheidungsfindung basiert auf Grundentscheidungen, die bereits in den Römischen Verträgen von 1957 getroffen wurden. 3

4

Richtlinie 2001/44/EG des Rates vom 15. Juni 2001 zur Änderung der Richtlinie 76/308/EWG über die gegenseitige Unterstützung bei der Beitreibung von Forderungen im Zusammenhang mit Maßnahmen, die Bestandteil des Finanzierungssystems des Europäischen Ausrichtungsund Garantiefonds für die Landwirtschaft sind, sowie von Abschöpfungen und Zöllen und bezüglich der Mehrwertsteuer und bestimmter Verbrauchsteuern in Verbindung mit der Richtlinie 2002/94/EG der Kommission vom 9. Dezember 2002 zur Festlegung ausführlicher Durchführungsbestimmungen zu bestimmten Artikeln der Richtlinie 76/308/EWG über die gegenseitige Unterstützung bei der Beitreibung von Forderungen im Zusammenhang mit bestimmten Abgaben, Zöllen, Steuern und sonstigen Maßnahmen. Bericht der Kommission an den Rat und das Europäische Parlament über die Anwendung der Bestimmungen über die gegenseitige Unterstützung bei der Beitreibung von Forderungen in Bezug auf bestimmte Abgaben, Zölle, Steuern und sonstige Maßnahmen vom 8.02.2006, KOM(2006)43 endgültig.

E. Administrative Aspects

159

Ein System wie die GKKB, in der für eine Vielzahl von Sachverhalten eine schnelle Entscheidung getroffen werden muss, kann auf derartig schwerfälligen Entscheidungsstrukturen nicht aufbauen.

6. Anforderungen an ein Verwaltungssystem für die GKKB Das Besondere an dem System der GKKB ist, dass die Mitgliedstaaten der EU eine konsolidierte Unternehmensgruppe so behandeln wie ein einheitliches Unternehmen, das in einem einzelnen Nationalstaat angesiedelt ist. Eine Gruppe mit Unternehmen in allen EU-Mitgliedstaaten ermittelt auf der Basis identischer Regeln eine gemeinsame Bemessungsgrundlage, die dann aufgrund einer Formel auf die einzelnen Mitgliedstaaten verteilt wird. Diese wenden auf ihren Anteil an der Bemessungsgrundlage dann lediglich noch ihren nationalen Körperschaftsteuersatz an. Ein solches System kann nur funktionieren, wenn die beteiligten Steuerverwaltungen eng miteinander kooperieren. So wie die konsolidierte Unternehmensgruppe als eine Einheit behandelt wird, müssen die beteiligten Verwaltungen als Einheit handeln und ein einheitliches Verwaltungsverfahren durchführen. Die gemeinsamen Regeln müssen einheitlich umgesetzt werden und bei Meinungsverschiedenheiten müssen klare und schnelle Streitbeilegungsverfahren eine für alle Mitgliedstaaten verbindliche Entscheidung definieren. Unternehmen müssen die Möglichkeit haben, wenn sie mit der Entscheidung der Verwaltung nicht zufrieden sind, in einem regelbasierten einheitlichen Rechtsbehelfsverfahren, bei dem die letzte Entscheidung wieder für alle Beteiligten verbindlich sein muss, Akte der Verwaltung überprüfen zu lassen. Da die vorhandenen Möglichkeiten dies nicht gewährleisten, muss bei der Einführung der GKKB eine adäquate Verwaltungsstruktur definiert, ein klarer Ablauf des Verwaltungsverfahrens vorgegeben, eine gleichmäßige Auslegung und Aktualisierung der Regeln sichergestellt und ein rechtsstaatlich einwandfreies außergerichtliches und gerichtliches Rechtsbehelfsverfahren ermöglicht werden.

7. Grundsätzliche Möglichkeiten für eine Verwaltungsstruktur zur GKKB 7.1 Nationale Ebene Grundsätzlich wäre es möglich, die derzeit existierenden Strukturen und Verfahrensregeln in den Mitgliedstaaten der EU beizubehalten und die Mitgliedstaaten nur zu verpflichten, dieselben Rechtsgrundlagen bei der Besteuerung von konsolidierten Unternehmensgruppen anzuwenden. Jeder Staat würde die Besteuerungsgrundlagen, die Bemessungsgrundlage für die auf seinem Territorium liegenden Unternehmen ermitteln und die Ergebnisse aus den anderen Mitgliedstaaten anerkennen. Da eine gemeinsame Bemessungsgrundlage ermittelt werden muss, ist es aber erforderlich, die in den verschieden Staaten festgestellten Ergebnisse zusammen zu führen. Ohne eine Regelung, an welcher Stelle die Informationen zusammengeführt werden sollen, könnte ein nur auf nationalen Verwaltungsregeln aufbauendes System aber nicht funktionieren. Schon dies zeigt, dass eine GKKB auch verwaltungstechnische Fragen klären muss. Darüber hinaus ist es, wie bereits erwähnt, erforder-

160

Ernst Czakert

lich, Mechanismen zu schaffen, die eine schnelle Entscheidungsfindung erlauben, wenn es in den verschiedenen Mitgliedstaaten zu unterschiedliche Interpretationen und Anwendungen der gemeinsamen Regeln kommt. Eine GKKB kann demnach nur funktionieren, wenn es die Mitgliedstaaten bindende gemeinsame Regel für die Administration des Systems gibt.

7.2 Zentrale Ebene Eine weitere Möglichkeit bestünde darin, für die Durchführung der GKKB eine gemeinsame Verwaltung auf EU-Ebene zu schaffen. Bei der GKKB geht es um die Besteuerung von bestimmten Unternehmen und Unternehmensgruppen in der EU. Dabei werden basierend auf gemeinsamen Regeln die Bemessungsgrundlage für den zu besteuernden Gewinn ermittelt, die Zugehörigkeit von Unternehmen zu eine konsolidierten Gruppe definiert und die Faktoren für die Aufteilung der Bemessungsgrundlage auf die beteiligten Mitgliedstaaten festgelegt. Zentral würde also der Steuerpflichtige definiert, die Besteuerungsgrundlagen ermittelt, das Besteuerungsverfahren durchgeführt und die Bemessungsgrundlage aufgrund der festgesetzten Faktoren für die Aufteilungsformel auf die beteiligten Staaten verteilt. Bei Zweifelsfragen müsste eine ebenfalls auf EU-Ebene angesiedelte Instanz eine Klärung herbeiführen. Die beschrieben Voraussetzungen skizzieren die wesentlichen Elemente einer EU-Steuerverwaltung. Bei dem derzeitigen Stand der Harmonisierung bzw. Nicht-Harmonisierung im Bereich der direkten Steuern in der EU ist es aber sehr unwahrscheinlich, dass die EU-Staaten für die Durchführung der GKKB die Schaffung einer EU Steuerverwaltung überhaupt als realistische Option in Erwägung ziehen würden.

7.3 Kombination Denkbar erscheint lediglich eine Kombination der vorgestellten Möglichkeiten. Basis müssten die jeweiligen nationalen Strukturen und Abläufe sein. Nur dort, wo tatsächlich erforderlich, müssten zusätzliche, gemeinsame Strukturen und Abläufe geschaffen werden. Grundsätzlich wären die nationalen Verwaltungen für die Durchführung der GKKB verantwortlich. Da der Steuerpflichtige, die konsolidierte Gruppe, in mehreren, im Extrem in allen EU-Mitgliedstaaten ansässig wäre, müsste klar definiert sein, wer für das gesamte Verfahren der GKKB von Seiten der Verwaltung verantwortlich ist und wie die Zusammenarbeit aller beteiligten Verwaltungen gestaltet werden soll.

8. Idealtypische Verwaltungsstruktur Definiert werden muss die verantwortliche Behörde für die Durchführung eines Steuerverfahrens und der verantwortliche Steuerpflichtige, der gegenüber der Behörde als Pflichtiger auftritt. Eine Unternehmensgruppe besteht aus der Muttergesellschaft, aus Unternehmenstöchtern und rechtlich unselbständigen Betriebsstätten. Der verantwortliche Steuerpflichtige kann nur das Unternehmen sein, welches als Muttergesellschaft die übrigen Unternehmen, Betriebsstätten und sonstigen

E. Administrative Aspects

161

Unternehmensteile beherrscht. Als beherrschendes Unternehmen liegen dort alle notwendigen Informationen für die Durchführung eines Besteuerungsverfahrens vor bzw. es besteht für das Unternehmen die Möglichkeit, alle relevanten Informationen zusammenzuführen. Verantwortliche Steuerbehörde sollte dann die Behörde sein, in deren Staat die Muttergesellschaft ansässig ist. So wie die Töchter und Betriebsstätten gegenüber der Muttergesellschaft bei dem Besteuerungsverfahren mitwirken, sind auch die Steuerbehörden der Staaten, in denen Töchter und Betriebsstätten der konsolidierten Unternehmensgruppe angesiedelt sind, an dem Steuerverfahren zu beteiligen.

9. Idealtypisches Verwaltungsverfahren Erfassung

Steuererklärung

Betriebsprüfung

Neufestsetzung

9.1 Erfassung/Registrierung Erster Schritt im Verfahrensablauf ist die Definition und Erfassung einer konsolidierten Unternehmensgruppe. Entsprechend den Regeln zur Konsolidierung muss der verantwortliche Steuerpflichtige bei der verantwortlichen Steuerbehörde einen Antrag stellen, als Konsolidierungsgruppe behandelt zu werden. Dabei ist gegenüber der Behörde nachzuweisen, welche Unternehmen in welchen Staaten zu der Unternehmensgruppe gehören. Sämtliche für eine Entscheidung erforderlichen Informationen sind dazu mit dem Antrag der verantwortlichen Steuerbehörde zu übermitteln. Der verantwortliche Steuerpflichtige informiert darüber gleichzeitig mit der Stellung des Antrags bei der verantwortlichen Steuerbehörde die für die anderen beteiligten Unternehmen (Töchter, Betriebsstätten) zuständigen Steuerbehörden in den anderen EU-Mitgliedstaaten. Die Steuerbehörden der anderen beteiligten EU-Mitgliedstaaten werden dadurch in die Lage versetzt, der verantwortlichen Steuerbehörde im Sitzstaat der Muttergesellschaft eventuelle Tatsachen zu übermitteln, die einer Einbeziehung der Tochter oder Betriebsstätte in ihrem jeweiligen Staat in die konsolidierte Gruppe entgegenstehen. Die verantwortliche Steuerbehörde trifft auf Basis des gestellten Antrags und möglicher weiterer angeforderter Informationen eine Entscheidung und übermittelt diese an den verantwortlichen Steuerpflichtigen und die übrigen beteiligten Mitgliedstaaten.

162

Ernst Czakert

9.2 Steuererklärung In einem zweiten Schritt übermittelt der verantwortliche Steuerpflichtige der verantwortlichen Steuerbehörde eine Steuererklärung über den zu versteuernden Gewinn der konsolidierten Gruppe entsprechend dem zugrunde liegenden Steuerjahr. Dabei fasst der verantwortliche Steuerpflichtige auf Basis der gemeinsamen Regeln die Daten der beteiligten Töchter und Betriebsstätten zu einer einzigen Bemessungsgrundlage zusammen und ermittelt die relevanten Faktoren für die Aufteilungsformel.

9.3 Veranlagungsverfahren In einem dritten Schritt überprüft die verantwortliche Steuerbehörde die übermittelte Steuererklärung auf Richtigkeit und Vollständigkeit. Fehlen Informationen so ist der verantwortliche Steuerpflichtige verpflichtet, diese auf Anfrage nachzuliefern; wenn erforderlich, kann die verantwortliche Steuerbehörde notwendige Informationen auch von den Steuerverwaltungen der anderen beteiligten EU-Mitgliedstaaten anfordern. Auf Basis dieser Informationen setzt die verantwortliche Steuerbehörde die Bemessungsgrundlage fest und teilt diese auf die beteiligten Mitgliedstaaten entsprechend der Aufteilungsformel auf. Die beteiligten Staaten werden dann von der verantwortlichen Steuerbehörde über den zu erlassenden Steuerbescheid und ihren Anteil an der Bemessungsgrundlage informiert. Gegenüber dem verantwortlichen Steuerpflichtigen und den übrigen Mitgliedern der konsolidierten Gruppe ergeht ein entsprechender Steuerbescheid, der die jeweilige Zahlungsverpflichtung der Mitglieder der konsolidierten Gruppe nach Ländern aufgeteilt beinhaltet. Der verantwortliche Steuerpflichtige sowie die Töchter bzw. Betriebsstätten in den anderen beteiligten EU-Mitgliedstaaten begleichen dann gegenüber der jeweils verantwortlichen Steuerverwaltung entsprechend den ihnen zugewiesenen Anteilen an der Bemessungsgrundlage und den jeweiligen nationalen Steuersätzen ihre Steuerschuld.

9.4 Betriebsprüfung Eine Betriebsprüfung wird von der verantwortlichen Steuerbehörde initiiert. Da im Rahmen der Prüfung einer konsolidierten Gruppe auch die Töchter und Betriebsstätten, die in anderen EU-Mitgliedstaaten ansässig sind, einbezogen werden, sind, wenn erforderlich, die Steuerverwaltungen in diesen anderen Mitgliedstaaten auf Ersuchen der verantwortlichen Steuerbehörde zur Mithilfe bei den Prüfungsmaßnahmen verpflichtet. Jede beteiligte Steuerverwaltung in den Staaten, in denen Töchter oder Betriebsstätten ansässig sind, kann wenn entsprechende Gründe vorliegen bei der verantwortlichen Steuerbehörde eine Betriebsprüfung anregen. Die verantwortliche Steuerbehörde ist verpflichtet, einem solchen Ersuchen nachzukommen. Sie informiert die ersuchende Verwaltung über das Ergebnis der Prüfung und über die sich aus der Prüfung ergebenden Schlussfolgerungen.

E. Administrative Aspects

163

9.5 Neufestsetzung Die verantwortliche Steuerbehörde wird das Ergebnis einer Betriebsprüfung feststellen und gegebenenfalls eine Neufestsetzung der Bemessungsgrundlage vornehmen. Der ursprüngliche Steuerbescheid wird entsprechend geändert und an den verantwortlichen Steuerpflichtigen übersandt. Gleichzeitig werden die anderen Steuerverwaltungen über die Änderung des Steuerbescheids informiert. Die verantwortliche Steuerbehörde wird gegenüber dem verantwortlichen Steuerpflichtigen und den übrigen Mitgliedern der konsolidierten Gruppe aufgeteilt nach Staatenzugehörigkeit die Steuerschuld endgültig festsetzen und jeweils entweder eine zusätzliche Zahlung oder eine Erstattung durch die zuständige Steuerverwaltung in Gang setzen.

10. Sicherstellung der gleichmäßigen Auslegung Die Regeln für die GKKB basieren auf EU-Recht. Bei 27 Mitgliedstaaten in der EU ist es zu erwarten, dass die Regeln im Einzelfall unterschiedlich ausgelegt und angewandt werden. Das Ziel, gemeinsame Regeln für die Besteuerung von Unternehmensgruppen in der EU festzulegen würde, wenn es keinen Mechanismus gäbe, unterschiedliche Auslegungen des gemeinsamen Rechts verbindlich für alle durch eine einheitliche Auslegung zu ersetzen, konterkariert. Ohne ein Verfahren zur schnellen Beilegung von Interpretationskonflikten kann eine GKKB deshalb nicht eingeführt werden. Wie bereits aufgezeigt sind die bestehenden Entscheidungsstrukturen auf EU-Ebene im Bereich der direkten Steuern für die Anforderungen der GKKB unzureichend. Bei einer Einführung der GKKB müssen deshalb zugleich mit den materiellrechtlichen Regeln zur Gewinnermittlung, Konsolidierung und Aufteilung Regeln für das gemeinsame Verwaltungsverfahren, zu denen auch schnelle Entscheidungsstrukturen gehören, definiert werden.

10.1 Ständiger Ausschuss Für Entscheidungen auf EU-Ebene im Bereich der GKKB ist es deshalb erforderlich einen Ständigen Ausschuss einzurichten, der verbindlich bei bekannt werden von Auslegungsdifferenzen eine verbindliche Entscheidung für alle EU-Staaten trifft. Die Ausgestaltung eines solchen Ausschusses könnte an das Komitologieverfahren der EU angelehnt werden.5 In einem solchen Ausschuss unter dem Vorsitz der Kommission wären alle Mitgliedstaaten vertreten.

5

Der Begriff „Komitologie“ bezeichnet die Mitwirkung von Ausschüssen bei der Durchführung des Gemeinschaftsrechts durch die Kommission. Der Rat ist ermächtigt, der Kommission Durchführungs-befugnisse zu übertragen (siehe Art. 202 dritter Spiegelstrich EGV). In der Richtlinie zur GKKB müssen dazu Inhalt und Ausmaß der Durchführungsbefugnisse festgelegt sowie das entsprechende Ausschussverfahren festgelegt werden.

164

Ernst Czakert

10.2 Verfahren und Entscheidungsfindung Im Bereich der Steuern gilt das Einstimmigkeitsprinzip. Für eine schnelle Entscheidungsfindung ist dieses Prinzip nicht geeignet. Eine GKKB muss aber schnelle Entscheidungen zulassen, da ansonsten der Konflikt einer unterschiedlichen Interpretation der Regeln zur GKKB auf dem Rücken der betroffenen Unternehmen ausgetragen würde. Es ist weiter ebenfalls möglich, dass sich die Mitgliedstaaten über die Anwendung der Aufteilungsregeln uneins sind, was einen direkten Einfluss auf das einem Staat zustehende Steueraufkommen hätte; auch in einem solchen Fall muss ein formal festgelegter Entscheidungsweg zu einer schnellen Lösung des Konflikts führen. Eine Entscheidung darf aus diesen Gründen nicht auf dem Einstimmigkeitsprinzip aufbauen. Entscheidungen im Rahmen eines ständigen Ausschusses für die GKKB müssen mit dem Mehrheitsprinzip getroffen werden können. Da in der Richtlinie zur GKKB Inhalt und Ausmaß der Regelungsbefugnisse eines ständigen Ausschusses festgelegt werden müssen und bei der Verabschiedung der Basisrichtlinie zur GKKB das Einstimmigkeitsprinzip gilt, wird vom Grundsatz nicht in das Prinzip der Einstimmigkeit eingegriffen. Je nach Bedeutung einer dem ständigen Ausschuss zugewiesenen Entscheidungsbefugnis, könnte die erforderliche Mehrheit auf einer Skala von einer einfachen bis zu einer qualifizierten Mehrheit (z.B. 75% der Stimmen) gestuft werden.

10.3 Mechanismus zur Anpassung der Regeln Damit die Regeln der GKKB schneller sich ändernden Gegebenheit angepasst werden können, könnte das Ausschussverfahren in speziell definierten Bereichen auch auf die Basisregeln zur GKKB angewandt werden. Abschreibungszeiträume, Bewertungsvorschriften, die steuerliche Behandlung neuer Sachverhalte sowie die Reaktion auf Missbrauch könnten zu diesen speziell definierten Bereichen gehören.

11. Rechtsbehelfsverfahren Im Rahmen des Besteuerungsverfahrens kann es an vielen Stellen zu unterschiedlichen Auffassungen zwischen dem verantwortlichen Steuerpflichtigen und der verantwortlichen Steuerbehörde kommen. Ist der Steuerpflichtige mit dem Steuerbescheid nicht einverstanden, dann muss er die Möglichkeit haben, seine Sicht der Dinge in einem rechtsstaatlichen Verfahren vorbringen zu können. Es handelt sich dabei um einen rechtlichen Konflikt über die Anwendung von EU-Recht, welches in allen EU-Staaten angewandt wird. Da es sich bei der GKKB um ein Verfahren handelt, das von einer großen Anzahl von Unternehmen innerhalb der EU genutzt werden wird, sollte das Rechtsbehelfsverfahren in ein außergerichtliches und gerichtliches Verfahren aufgeteilt werden.

11.1 Außergerichtliches Verfahren Die Verwaltung hat in einem außergerichtlichen Verfahren (Einspruchsverfahren) die Möglichkeit, die eigene Entscheidung noch einmal zu überprüfen und bei Bedarf

E. Administrative Aspects

165

zu korrigieren. Der verantwortliche Steuerpflichtige sollte dazu das Recht haben, innerhalb einer definieren Frist einen Antrag auf Überprüfung des im zugegangenen Steuerbescheids bei der verantwortlichen Steuerbehörde zu stellen. In dem Antrag sollte der verantwortliche Steuerpflichtige die Gründe darlegen, warum aus seiner Sicht der Bescheid zu korrigieren ist. Kommt die verantwortliche Steuerbehörde zu der Entscheidung, dass der Antrag gerechtfertigt ist, wird sie den Steuerbescheid entsprechend abändern und an den Steuerpflichtigen übermitteln. Das Verfahren wäre damit beendet. Stellt die verantwortliche Steuerbehörde fest, dass es bei der Anwendung des einschlägigen Rechts in anderen EU-Staaten differierende Auffassungen gibt, dann ist auch im Rahmen eines außergerichtlichen Rechtsbehelfsverfahrens eine Entscheidung des ständigen Ausschusses bei der Kommission einzuholen. Entscheidet sich die Steuerbehörde, den ursprünglichen Steuerbescheid aufrechtzuerhalten oder bleibt auch bei einer Abänderung des Bescheids eine Beschwer für den Steuerpflichtigen erhalten, dann hat der verantwortliche Steuerpflichtige das Recht, ein gerichtliches Verfahren anzustrengen.

11.2 Gerichtliches Verfahren Bei einem Verfahren wie der GKKB dürften sehr viele gerichtliche Verfahren eingeleitet werden. Vom Grundsatz her sollten diese Verfahren von den zuständigen Gerichten der EU-Mitgliedstaaten entschieden werden. Es handelt sich aber bei der GKKB um europäisches Recht und ist die Frage der Auslegung oder Gültigkeit von europäischem Recht für den Ausgang eines Rechtsstreits entscheidungsrelevant, dann ist im Grundsatz eine Vorabentscheidung des Europäischen Gerichtshofes zu der maßgeblichen Rechtsfrage vor einer endgültigen Entscheidung einzuholen.6 Es stellt sich dann aber die Frage, ob der Europäische Gerichtshof, dessen Funktion für das europäische Recht mit der des Bundesverfassungsgerichts für deutsches Recht in Deutschland vergleichbar ist, für rechtliche Fragen zur GKKB der richtige Spruchkörper ist. Bei Entscheidungen zur GKKB geht es um die Anwendung von EU-Recht bei steuerrechtlichen Spezialfragen. Eine Vielzahl von Verfahren, in denen es zum Beispiel um Zurechnungsvorschriften, Abschreibungsregeln, die Zulässigkeit von Rückstellungen oder die Abzugsfähigkeit bestimmter Aufwendungen als Betriebsausgabe gehen kann, sind von einem Fachgericht mit spezialisierten Richtern sicher schneller und effizienter zu erledigen als von einem Gericht für Grundfragen des EU-Rechts wie dem Europäischen Gerichtshof. Sollte es deshalb zu einer Einführung der GKKB kommen, dann ist die Einrichtung eines auf europäischer Ebene angesiedelten Fachgerichts für steuerliche Fragen nahe liegend. Der EuGH könnte sich dann darauf beschränken darüber zu wachen, dass dieses steuerliche Fachgericht die elementaren Grundsätze europäischen Rechts in seiner Entscheidungspraxis beachtet.

6

Siehe Art. 234 EG.

166

Ernst Czakert

12. Fazit Die administrativen Aspekte einer Gemeinsamen Konsolidierten Körperschaftsteuerlichen Bemessungsgrundlage sind für den Erfolg eines solchen Systems von elementarer Bedeutung. Sollten sich die EU-Staaten für die Einführung einer GKKB entscheiden, wird dies nur möglich sein, wenn die Fragen des Verwaltungsablaufs, einer geregelten und effizienten Entscheidungsfindung auf europäischer Ebene und die Durchführung von Rechtsbehelfsverfahren in einer für alle Beteiligten an der GKKB zufriedenstellenden Art gelöst sind. Für die Kommission bleibt bis zur Vorlage ihres angekündigten Richtlinienvorschlags zur GKKB noch viel zu tun. Auf Basis dieses Vorschlags müssen die EU-Mitgliedstaaten dann eine Entscheidung treffen, ob sie das Wagnis einer GKKB eingehen können.

E. Administrative Aspects

167

II. Administrative Aspects of the CCCTB – a Business Perpective Theo Keijzer

Meine Damen und Herren, Kiries kai Kiriy, Dames en Heren, Ladies and Gentlemen: What I have just done is use four languages to express the problems we face in the tax environment in the European Union today. I cannot read a Swedish tax return because I do not read Swedish. I cannot interpret the Portuguese tax return because I do not know what the rules are in Portugal. Nor do I know any tax professional in Europe who speaks all languages and knows all the tax rules in all the 27 Member States. Not able to read and review the corporate tax returns in all Member States is the metaphor for the fragmented and expensive corporate tax landscape we have in Europe1. During this presentation, I will first outline the administrative problems we have with the current arrangements, then describe the solutions and finish by describing the new landscape. The new landscape as I see it. What I say today is not based on Shell specifically but on my 30 years experience as a tax advisor in a company. Today there are many borderlines within Europe meaning many different tax systems and different cultures. All this makes Europe so unique. It’s a great place to do business but it’s also a difficult place to do business if you work on the assumption that we have one Internal Market. After all, well over 10 years ago we were told we had one Internal Market. The problem we have within the EU is not only that we have many countries with different tax rules but we have also many different authorities supervising and monitoring these different rules. We lack a single tax law in Europe to support the one Internal Market, instead we have today 27 different countries with 27 different styles of operating by 27 different Revenue authorities based on 27 very different tax laws. This comes at a cost to us all. Prof. Spengel yesterday demonstrated the differences between the various tax systems. The business community basically sees the EU as one market. Let me give you an example. Assume a group of companies has a plant in Southern France and assume this plant can deliver products to an affiliated factory in Barcelona. Assume also that the same group of companies has a plant in Southern Spain. What would be more efficient, to travel all the way through Spain to Barcelona or just a few kilometres from Southern France to Barcelona? The decision to provide the factory in Barcelona from the plant in France means a cross border transaction and in Europe 1

The terms Europe and EU are both used in this presentation and refer to the same area unless context is clear otherwise.

168

Theo Keijzer

today this results immediately in a transfer pricing and a permanent establishment issue. That is the reality of today’s Internal Market. That is why all of us need to work hard to create the true one Internal Market; with a single tax return for a panEuropean group of companies. What are the main problems? I’ve listed them and I will go through them one by one. The first problem is compliance costs. There are at least 3 authoritative studies that indicate significant costs to all of us in Europe for not having one set of corporate tax rules but 27 tax laws. The Brussels based CEPS indicated in 2002 that the cost would be about 2 to 4 % of total corporate tax revenue. Devereux in 2004 indicated it would be somewhere between 2.7 and 4 % of tax revenues. A recent study in the Netherlands of the Central Plan Bureau indicates a significantly higher number of 10 % of corporate tax revenue. Whatever the correct number is, it is a significant number. Many billions of euro per annum are lost in Europe for not having a single set of corporate tax rules. That represents only the costs of not having a simple compliance system. Can you imagine what the amount would be if studies would take into account new dynamic ways of operating in the EU when we have a single set of rules? The second issue is transfer pricing. What we have seen over the last 6 to 8 years is the sudden rise of transfer pricing as an important tax issue. In 2000, transfer pricing compliance was hardly an issue. Some external firms were starting their transfer pricing departments; a few years later companies followed with internal TP teams. The US was ahead of us in the EU because they had strict TP compliance rules for decades. Already today in Europe it is typical for a group of companies to have for each company or country a TP report of hundreds of pages. 300 pages are not unusual. Why are we doing this to each other? We have the unique opportunity to get rid of this documentation burden but only if we have one true Internal Market. My last remark on this topic is something I heard the other day. There is at least one cost conscious company that has said: “Enough is enough!” The costs to prepare all these TP reports in Europe are too high and it started to offshore the preparation. Think about it, offshoring TP documentation requirements to India. They hired Indian professionals to do the analyses and to produce the reports because it was too expensive to continue doing this in Europe. Have we forgotten that tax compliance is a cost for a company and that the administrative burden should be as low as possible? What are we doing to ourselves in Europe? The third issue is suboptimal business structures. Borders cause companies to behave suboptimal. Commercial structures tend to be based on various parameters; country borders need to be taken into account in the structures but frequently these revised structures create a less optimal structure to avoid the tax costs. A suboptimal business model. It is an issue, it is a cost issue. Recent OECD projects like those dealing with intermediaries, business restructuring and services permanent establishments indicate the impact of borders on commercial activities. These studies would not be as relevant within Europe if we had one true internal market. With the commercial reality of an increasingly inte-

E. Administrative Aspects

169

grated internal market we’re bound to see more cross border disputes. The European Joint Transfer Pricing Forum has done a great job in the last 4 years, but it could not reduce the number of disputes. There is simply more cross border business today than in the past. Where there is cross border business we have seen simplification and rationalization in business models like toll processing arrangements and commissionaire structures. All of these activities aim to do what companies in general want to do and that is to do business at lowest costs. These structures to allow optimal commercial flows are challenged because of their tax consequences. Again, this would not happen if we had one true internal market as expressed in the CCCTB. The fourth problem is non-consolidation. I’ve mentioned so far compliance costs, TP aspects and suboptimal business structures, but not yet the lack of consolidation. The absence of consolidation in the EU causes potential friction when deciding where to locate new business. In the US, start-up losses from new business go against profits from mature business activities, wherever located in the US. Not so in Europe. Tax is a relevant factor and could disturb the commercial drivers. That is not good. The last aspect is assurance. I don’t know how many of you are familiar with Sarbanes-Oxley and FIN 48 etc.? These are new developments for quoted companies. It turns about basically having full insight and control over all that happens – also for tax – in your company. Given that we have 27 different rules in Europe we have an assurance issue. 27 different tax laws and returns with different payment days, different filing dates and different audit procedures do create assurance issues. The question today is how can you control all this doing cross border business. How can you control all this at a reasonable cost? The CCCTB resolves a large number of the topics mentioned. I will clarify in some detail later, but let me already say now that two basic conditions must be fulfilled. First, there must be a “one step” and not a two step approach, so no harmonization first meaning same tax laws in all 27 countries and only a number of years later consolidation. Second, not 27 Revenues to file the return and not 27 Member States in charge of the tax audit. There must be a “one stop shop”. Let us consider a lesson from the past before I address details of the administrative tax aspects seen from the business side where Mr. Czakert has done so in detail from the Revenue side. Let us consider what has happened in the past. Are there any lessons to be learned? It is sometimes good to look back and see what have we done in Europe. We’ve had the EEIG and the SE. These were not a success by any standard. Why? They did not address business issues and concerns. I ask you, let the saying “history repeats itself” not be true this time. Let us all reach out to come to a landing with a simple, optional consolidated tax base across the EU that fuels the economies of all Member States and makes Europe the most competitive place in the world. Let me now turn to the administrative aspects of the CCCTB. I’ve outlined earlier some elements of the framework of a corporate tax system for the Internal Mar-

170

Theo Keijzer

ket that allows us to compete with the single law and single tax authority countries. Many have addressed the single corporation tax code for the EU. There are rumours that certain business activities may be excluded from the CCCTB. Doing so would taint in the Internal Market and be negative for Europe. I’ve mentioned earlier that the one step approach is necessary. A consolidated single tax framework from the start. The two step approach with 27 similar tax systems on day one means 27 different tax systems on day two given the different cultures, environments, legal systems etc. It is not realistic to assume that the systems will stay the same for a number of years. Second, we need one consolidated tax return. I argue for one Principal Tax Authority to receive the single consolidated return. The role of this authority as I see it is to be the leader or the co-ordinator of all other authorities allowing the tax payer to have one single contact, one touch down point. There may be a reason to have a second contact point when it comes to tax payment, because the countries represented here would most likely prefer to see the tax payments flow directly into their own Treasury rather then via the Principal Tax Authority. Such second contact for payment purposes is only procedural; we can handle this. Third, the tax payment. When I thought about the issue, I asked myself: how do you get an efficient mechanism of paying an instalment during a year. Some call it an advance payment on a preliminary assessment. First, you have to estimate the base, the total base in Europe and then allocate it to each country. That’s prone to error. What are we doing to ourselves if we pay first and then ask it back mid year after recalculating our estimated taxable profit. What do we do when we file the return? Later we have the final assessment. This is an issue that should be worked. At the end of the day it should be an efficient mechanism that allows for a reasonable determination what the payments are to each country. One solution might be payment at a date of filing. On the allocation key: the way it ought to work is simple, simplicity first and last. Second, it should not be dispute prone. Let’s have a clear understanding what the allocation key is. The key must not be subject to different interpretations. I suggest that the validation of the allocation keys and the calculation is done by the Principal Tax Authority, not by all countries together. This avoids a prolonged argument of how it works and how it should be allocated. The audit procedure has the same process angles as filing a single tax return. I believe in one country being assigned the role of the Principal Tax Authority. This country has the lead or the coordination of the audit effort. The process should be efficient and not result in long audit periods with much delay when it comes to a final assessment. In resolving this issue again many variations are possible. Details do matter but efficiency is crucial. Think about this question: Why should an audit take more than one year and why should an audit not start shortly after filing the tax return? Audit questions are for the business community very relevant. Today’s environment shows external auditors of a company demanding precise information regarding the tax position of a company. Long delays in an audit procedure create uncertainty. This is not good for the business community. The resolution of the audit aspects of the CCCTB could well determine whether the business embraces the

E. Administrative Aspects

171

CCCTB or not. We know what we have today and the proposed CCCTB audit framework will be taken into account as one of the major deciding factors. An appeal procedure should be efficient and yield the result quickly. The fact that there may be 27 countries whose tax proceeds are determined by the outcome of the appeal should not influence the appeal procedure. There are a number of alternatives and I’m in favour of the following sequence. First, an appeal with the Principal Tax Authority. The next step is not to continue with a domestic procedure in that same country. The next step is a body that can ensure consistency in decisions across Europe. I can for now only think of one body that can do that and that is a European body. Not a Court of First Instance. We may consider introducing a body below the Court of First Instance where a single judge will decide. There will be no doubt many cases and a Court of First Instance cannot handle that work. Whether that Court is based in Luxemburg or whether it has branches in each of the countries, does not matter for now. More important is to have cross EU consistency. On the common interpretation of the tax rules we can distinguish a number of aspects. 1. Same rules in all countries. How do we achieve this? A directive may help but a regulation looks better. 2. All income should be included (non-EU and passive income). Otherwise we have multiple tax returns. The income definition should not be limited only to intra-EU income but have global income included. 3. We need to have a mechanism that causes interpretation between East and West and between North and South Europe to be the same. As Mr. Czakert said we may well look for a Standing Committee in Brussels. Can you imagine that a tax examiner in New Orleans, interprets the US tax code differently than the examiner in Boston? We will see this happen in the EU if we do not ensure that the interpretation is aligned. Lastly queries: Where do we go to when there is a question? I suggest that we go to the Principal Tax Authority where I file my return. The taxpayer should have access to a single contact within the EU who can take decisions and respond with authority to the questions raised. I mentioned Sarbanes-Oxley and FIN 482. Especially FIN 48 forces companies to identify uncertainties and a complex CCCTB system is probably worse than the fragmented system we have today. When companies need to decide whether to go for a CCCTB or not, a complex CCCTB system has an impact on FIN 48 and may well be a relevant decision factor. As explained before, the existing Transfer Pricing (TP) burden is high and we can create a major win for Europe if we can scrap the TP (documentation) burden for intra-EU transactions. One administrative aspect might be “culture”.I am hesitant to mention the issue, but let me try and introduce at least the topic. It refers to the relationship between the 2

It is acknowledged that the situation regarding FIN 48 for EU companies using IFRS has changed since May 2007.

172

Theo Keijzer

taxpayer and the Revenue authority. There are countries where the relationship is based on the “them and us” thinking and there are those where the relationship is based on ”how can we enforce the laws together” realising the different tasks each party has. This topic is addressed in the intermediaries project at the OECD. The question before you is: can we have a good CCCTB system without a common approach on how to enforce the rules and the allocation keys? Allocation keys. The allocation keys have the power to cause the CCCTB to be less efficient in its operation than we might hope for. One example looks to be the sales key in the formulary apportionment. The sales key should not create the back door introduction of the TP burden. We should consider not using the sales key. Let me also mention some start-up issues. Countries appear to have genuine concerns that there may be undesirable planning opportunities in the first years of the CCCTB, that profits will be shifted from high tax to low tax States. If countries are concerned I suggest discussing this with the business community; we may be able to help and resolve the issue. For instance we may propose that if an asset is transferred within two years after the introduction of the CCCTB to another country it needs to be tracked. And if it’s sold within two years thereafter it will be tracked back to the original country. There are solutions to address start-up issues. On optionality, we discussed it at length yesterday. It is relevant to have optionality because otherwise I would not be standing here. A number of years ago I decided that CCCTB was a good project. A true Internal Market is a dream come true for all of us in Europe, not only for the business community but especially for also for the citizens. It increases the welfare of Europe. However, when accepting CCCTB many years ago as a good project I was uncertain about the outcome. I know what I have today and I don’t know what the Member States will decide; they have the power to decide not the business community. That means that I must have optionality. Otherwise I’m signing a blank check. The future landscape for me is, for corporate tax purposes, a borderless Internal Market. In such a Market, entrepreneurs, and fortunately there are many in Europe, can travel across Europe to undertake their business activities where they deem best and not hindered by corporate tax aspects. We can save money by lower compliance costs and scrapping the excessive TP documentation inside Europe. These benefits to Europe will fuel the dynamics of the economy and increase the welfare of all citizens of Europe. I see CCCTB as a step forward, for the European business, in the global competition. According to the European Commission the dream of the early nineties about a borderless Europe can come true in 2010; let us make it happen.

E. Administrative Aspects

173

III. Panel Statements 1. María Teresa Soler Roch

As it has been remarked, the current provisions on exchange of information (EI) are not adequate for CCCTB purposes. It is important to remember that information is not a goal in itself, but an instrument for the correct assessment and moreover, for the correct audit of the tax base. At this point, EI is necessary because the power of the Tax Administration does not have a transnational scope, thus it cannot be applied in cross-border situations. For this reason, the current EI system provides help from other Tax Administrations, but it deals with the tax base of a single entity calculated at the domestic level and as a result of applying national rules. In a CCCTB context the situation is quite different, because it deals with the consolidated tax base of a group of companies and as a result of applying common provisions. So the playing field has changed: it is not any more the state of each single entity, but the different EU Member States that are involved in the application of this system, even if this application is centralised in the MS Tax Administration of the controlling entity. Moreover, due to the complexity of the CCCTB system, its implementation needs to be supported by information on different aspects such as: information required for the assessment and audit of the tax base, both at the group level and at that of each individual entity, information about the relevant data for applying the CCCTB sharing mechanism (and thus for the calculation of the tax debt in each Member State involved) and iInformation about each entity’s assets located in other Member States involved, in so far as mutual assistance in collection might be required. Therefore, the role of EI in a CCCTB system shows a different perspective. The information, including all the relevant data mentioned above, could be obtained during the CCCTB application process, in which at least three different steps could be considered. I will refer to these steps in relation to the hypothesis of the one-stopshop regime: A first step, at the level of the tax return. At this level, the Tax Administration of the controlling entity should automatically obtain the information concerning the tax bases (both CCCTB and single tax bases), as well as the relevant data (for the sharing mechanism and collection) already mentioned. In my view, the easiest way to obtain this information would be through the controlling entity, which should provide it attached to the CCCTB tax return. A second step, when all this information should automatically be provided by the Tax Administration of the controlling entity to the rest of Tax Administrations involved in the process. A third step concerns the level where EI has traditionally played its relevant role: when further information is required for the audit procedure. At this stage, the con-

174

María Teresa Soler Roch

tent of EI is linked to other main issues, such as which Tax Administration will audit the group, or the extent of the information required (i.e.: access to the records). It is precisely at this point where the current EI system is far from being adequate. But the solution is not only a question of reinforcing this system. In this case, I would rather propose another concept, based on the idea of coordination in auditing. The scope of this concept within the CCCTB system would require deciding on different proposals (some of which have already been expressed by the CCCTB WG),1 such as to set up specific units, acting on a coordinated and multinational basis. This proposal also implies accepting the idea of a transnational power, thus overriding the limited national power. These units should act under common procedural rules, specially those related to disclosure of information, statute of limitation, deadlines, and taxpayers’ rights. The audit of the group should be carried out at the same time for all the entities involved. This would avoid the problems related to further adjustments deriving from single audits. In this context, the way of getting the information would be different: the idea is that an automatic and fluent EI should be provided within the unit on a coordination basis, (i.e.: by means of reports and periodical meetings) and the results pooled.

1

See: CCCTB/WP/036 doc.

E. Administrative Aspects

175

2. Miklós Kok

As we are all aware, the Lisbon strategic goal of the European Union of becoming “the most competitive and dynamic knowledge-based economy in the world” has a determining impact on the taxation of companies in that it aims at building up a supportive general framework for economic activity in the EU. Nonetheless, little has happened so far in the area of company taxation as Member States keep on operating the same company tax systems as they did before the set-up of the Internal Market. It seems inevitable that more progress has to be foreseen in this field to achieve the said strategic goal by also setting the objective of abolishing internal tax obstacles, which hinder cross-border trade, establishment and investment. After many years of attempts to establish some degree of a harmonised corporate tax system at EU level by the introduction of some targeted measures such as a guidance on important ECJ rulings, the Merger Directive and the Parent-Subsidiary Directive, etc., a – in many of its elements – renewed and compromised initiative of the European Commission is taking shape. The initiative of a Common Consolidated Corporate Tax Base is – no doubt – still a controversial issue. This is most of all due to the complexity of the issue. As has already been discussed in details by the speakers in the previous two panels, the CCCTB – on the one hand – is seen as the only means, which can do away, within the EU, with transfer pricing problems; allow for the offsetting and comprehensive consolidation of profits and losses on an EU basis; simplify many international restructuring operations; reduce, without extending into the personal tax field, some of the complexities arising from the co-existence of the classical and exemption approaches to international taxation; avoid many situations of double taxation and remove many discriminatory situations and restrictions. On the other hand, however, the need to fit together the different structural elements of the tax base, consolidation and the sharing mechanism in the foreseen CCCTB system suggests that – in addition to the fact that Member States have to accept differences from their existing tax bases – a massive reconciliation procedure is foreseen. Besides the aspects having already been dealt with by speakers of the conference and also by the EC CCCTB Working Group on the taxable income, related international issues, personal scope, consolidation/group taxation, treatment of financial assets, a relatively new additional element is to be taken into account, and i.e., the administrative and legal framework that is to cover practical aspects of the CCCTB system. From this latter perspective, the presentations of the main panel speakers on the overview of the current situation as is seen by both tax administrations and business actors have highlighted a number of existing obstacles that – in their view – need to be remedied. The major difficulty lies in the differences in tax legislation of the Member States. As a consequence corporate taxes are administered and related laws are interpreted in diverse ways. Special focus has been put on the controversies in

176

Miklós Kok

mutual agreement procedures and the anomalies of the slow decision making process at EU level. A clear message has been put forward in that the current legal framework, such the mutual assistance directive and the tax recovery directive and the existing administrative tools (no common tools) are not adequate. However, also from the legal and administrative point of view, there are necessary consequences of a new CCCTB system. For me a key question arises. That is the degree of harmonisation necessarily required by the CCCTB system. It is essential that we identify to what extent the existing national legal framework (procedural laws of the EU Member States) should or could be harmonised at EU level. If so, a consecutive question is which procedural elements of the CCCTB system should be harmonised and at what level of EU legislation. As regards the latter one, a new directive seems to suffice the CCCTB initiative. From a reverse angel though, we may think about the extent to which the existing national procedural rules could be applied for the purposes of the CCCTB system initiative. It is of great importance, and has also been outlined by the Commission, that tax procedures are mostly covered by separate procedural laws in the different Member States, which are often interrelated with other branches of their national public laws (tax audit procedures, tax dispute resolution systems). On this basis, we should bear in mind that harmonising the rules for calculating the corporate tax base does not require an overall harmonisation of the tax administration and procedural laws. One of the goals of the CCCTB concept is to simplify and to reduce compliance costs for companies. Thus, another principle dilemma is whether eligible companies can handle their tax affairs in a on-stop-shop or they should continue to deal with up to 27 different tax systems (multiple-shop). The choice would affect the whole range of tax procedure from registration to appeal. Irrespective of the choice, a number of related practical issues, and as such, rules are to be dealt with and harmonised under the CCCTB concept. It seems to be obvious that harmonised tax return forms would lower administrative costs and would allow companies to fill them in in any of the EU languages (tax authorities could examine a tax return filled in in a foreign language, because it would follow a standardised model). A benefit of harmonisation of forms would be that companies could complete a tax return in an EU language which is not a national language of the MS where the entity is located. The standardised form would also facilitate the processing of returns. The period of time to be covered by a tax return would ideally be a single tax year in all jurisdictions. Should this be adjusted to the calendar year or to a different 12 month period as is current allowed in some Member States is to be clarified. Anyway, different definitions could cause problems. Much would depend on what the primary source of information is (financial accounts, etc.). Similarly, the deadline for submission of the tax return is also an issue. Should this be due several months after the end of the tax year, or should the dates be fixed? As to which entity of a group company should complete and submit a tax return and to which tax administration is much dependant on the applied mechanism (as

E. Administrative Aspects

177

referred to above under one-stop-shop or multiple-shop). This should look like (assuming consolidation of EU-wide income from the beginning) as follows: In a one-stop-shop-approach a principal taxpayer (i.e. controlling EU parent company) files one consolidated tax return for all entities of the entire group to a principal tax authority in its country of residence. That requires information from each entity before consolidation. The result is the consolidated tax base (net 3rd party income) of the group. It follows the apportionment of the consolidated tax base. The principal tax administration would be responsible for the assessment of the consolidated base and the distribution of apportionment information to the tax administrations of Member States concerned – for them to assess the tax liability of the respective entity in those Member States and also for audit purposes. In a mutiple-shop-approach each entity of the CCCTB-group would continue filing its own tax return to the tax administration of its residence (as was already mentioned, permanent establishments of entities in different Member States should have special treatment). In this case the parent company may also have to file a specific tax return for the calculation of the consolidated taxable income to the tax administrations of its residence. The tax administrations of Member States, where entities of a group company are resident for tax purposes, should forward the information to the principal tax administration for consolidation and apportionment purposes. Of course, any of the above options would have its necessary consequences for the legal and administrative environment, and ultimately for the CCCTB concept. The one-stop shop mechanism would allow a centralised management of the common consolidated tax base, through single compliance at a single location with possibly single assessment, audit, reassessment of the consolidation and the apportionment by a principal tax administration. Companies in question thus need to interact with only one single tax administration. If so, the administrative framework would extensively follow the CCCTB concept of simplification, reduction of compliance costs and of the administrative burden on companies and the common application of the CCCTB system. However, the tax administrations in Member States, where the entities of the group are located could have no direct control and access to the tax documentation of these entities. The treatment of group re-structuring under domestic registration rules would also become difficult, if any entity of the group quits the system. As regards tax assessment, determining the consolidated base and the apportionment by the principal tax administration would mean that tax administrations in other Member States could foresee difficult and timely procedures if they do not agree, unless a centralised, coordinated approach is developed at EU level. Also the question arises, which service should be competent to initiate and perform an audit at group companies (parent and sub-entites) and how the tax administrations of Member States should cooperate in this regard and under what procedure, as access to companies’ legal seats and inspection of books, premises and employees can only be performed by officials empowered to do so by law. Differences in audit periods covered, the duration and modalities of audits applied under national legislations in different Member States should also be taken into account

178

Miklós Kok

when talking about harmonisation of related rules to avoid future difficulties under the CCCTB concept. The consequences of an audit are also to be considered as to how to adjust the initial tax base by systematically circulating the official and final outcome of audits amongst the tax administrations concerned. Reassessment is also an issue, if the taxpayer has to initiate it for any reason. Clear rules are to be defined as to how a non-principal tax administrations can initiate the reassessment procedure, which will have an impact on the tax base and apportionment of the overall group. Thoughts should be given to the interpretation of CCCTB rules by the taxpayers and the tax administrations as well. How to ensure harmonised interpretation of the CCCTB rules? A well-designed mechanism should ensure that tax administrations issue uniform interpretations (explanatory circulars, etc.). There is no case law to rely on. Also mutual recognition of such interpretations should be ensured. Similar questions arise when talking about rulings. Common rules are to be set also for CCCTB specific dispute resolution in the pre-litigation phase. Which authority is competent for any case, is a key question, if the system is to smoothly operate. In either way, the currently inadequate tools of tax cooperation and exchange of information should be significantly reinforced to manage the CCCTB. In a multiple-shop approach the need for companies to proceed with several tax systems will remain an issue. In addition, compliance costs and burden will not be reduced. The approach could especially give rise to concerns (both for companies and tax adminstrations) about transfer pricing issues and will not have an immediate solution for current double taxation problems. As for the tax administrations, the multiple-shop approach more resembles the current situation without radical improvement of the administrative procedures with relatively less harmonisation of the procedural rules at community level, leaving them more in the national domains of Member States. Obviously, though strongly needed for the smooth operation of the system, less procedural adaptation would be effected by the individual tax administrations. A consolidated tax base in a multipleshop would – in my view – be even more complex to administer, and therefore, would not make really sense. On this basis, I tend to believe that the one-shop-stop system is a more realistic option. Inevitably, there should be a balance between the need of tax authorities to have an effective control over the group entities in their own jurisdictions and the overall CCCTB concept including the one-stop-shop approach. The need for strengthened coordination at EU level – in particular in those areas where it is indispensable due to the characteristics of the CCCTB – is inevitable for establishing common rules, while leaving some rules to be defined by the existing domestic systems.

E. Administrative Aspects

179

3. Matthias Mors

One of the key challenges in designing the administrative framework for the CCCTB is to reap the significant potential gains in terms of reduced administrative and compliance costs compared to the existing situation while at the same time allowing Member States to preserve a significant amount of national sovereignty. The most desirable administrative solution would be to have a single company (e.g. the parent company), acting on behalf of a multi-national group, dealing with one competent tax authority (e.g. the one where the parent company is tax resident), acting as principal tax authority on behalf of all national tax authorities concerned (one-stop-shop). With respect to the content of the tax declaration, in the context of a self-assessment framework, the consolidated tax return submitted by the principal tax payer to the principal tax administration would calculate the tax base and, by application of the individual statutory tax rates, the tax liability of all the group companies in the various CCCTB-jurisdictions in which the group is liable to tax and include all the necessary information for calculating this liability. Thus, the consolidated tax return would comprise the consolidated tax base, the factors for sharing the consolidated base, assuming the tax base would be shared between Member States on the basis of company-specific profit generating factors, the share of individual companies in the consolidated tax base, and the tax liability of each tax payer in the group. The tax declaration should be made, ideally electronically, in a common format and accompanied by a number of annexes with more detailed information. A common, standardised format would greatly facilitate matters as it would practically eliminate the need for translation and would allow immediate processing by all tax administrations concerned. Upon receipt of the tax declaration, the principal (lead) tax administration would transmit all the information to all other tax administrations concerned. The existing Mutual Assistance Directive (77/799/EEC) provides already for a basic framework for an automatic exchange of information. The most effective way of sharing the information would be a common database with restricted access rights. In this case, a simple message to the tax administrations concerned that new information has been put into the system would seem to be enough. It should be recalled that the Community is already financing the informatics infrastructure required for the functioning of the VAT or excise control system and a similar action can be envisaged for the operation of the CCCTB. As not all the details required for a smooth functioning of the CCCTB can be expected to be laid down in primary legislation, i.e. the basic CCCTB-Directive, the Directive will need to provide for more detailed implementation measures to be introduced. An example of such an implementation measure would be the development of a standardised tax declaration format.

180

Matthias Mors

The adoption of such implementing provisions could be delegated to the European Commission, assisted by a committee of Member State representatives as provided for by the so-called committee procedure (Council Decision 1999/468/EC). Following a regulatory procedure, the Commission would only be able to adopt implementing measures if they obtain a favourable opinion from the committee, given by a qualified majority of Member States. At the same time, the committee could also be a place for establishing a common interpretation of the provisions of the Directive. In order to be effective, such interpretation or corresponding guidelines or codes of conduct should reflect the views of as many Member States as possible and, ideally, should be respected by all Member States. In the case of a tax audit, for obvious reasons of practicality the audit should be conducted by the tax administration of the Member State where the respective company is tax resident (or where a PE is situated). Such an audit could be on the initiative of the 'local' tax administration, or at the request of another tax administration (e.g. the principal or lead tax authority). However, any decision to undertake an audit should be taken jointly by all tax administrations concerned and any audit should be undertaken under the overall responsibility of the principal tax authority. The results of the audit will have to be communicated to all other tax administrations concerned – again use of a common database would make sense. For any adjustments above a certain de minimis threshold, the principal tax authority would have to issue a notice of reassessment as adjustments would eventually require a recalculation of the 'shares' of all the group companies. Adjustments would therefore have to be agreed by all tax administrations and once agreed the resulting reassessment by the principal tax administration would be binding for all tax administrations concerned. To keep the process simple, reassessments could be limited to, for example, one per year, rather than one each time an administration identified the need for an adjustment. In certain circumstances, it might be desirable to provide for simultaneous or even joint audits. In this respect, one could draw on the experience in the context of the Nordic Council or in the US. Again, the existing Mutual Assistance Directive already provides a legal base for such joint audits. In case of disputes, an effective mechanism for resolving disputes between Member States is required. While, in a first stage, this is likely to only involve the Member States concerned by a particular dispute, the second stage could entail a cross-cutting arbitration body.

E. Administrative Aspects

181

4. Philip Baker

I’ve been asked to consider dispute resolution, and particularly to address the question whether the current court system in the European Union meets the requirements of a CCCTB. I have a very clear view that the current court system does not meet the requirements of the CCCTB, but I am also of the view that the current court system does not adequately meet the problems of cross-border tax disputes in any event. So we would be here discussing this issue in any event even if the CCCTB had never been invented. I would like to look at three elements. First, what are the features of the current court system for tax disputes? Secondly, I would like to discuss briefly the types of dispute that may arise under CCCTB. Thirdly, what solutions may be available, and which is the optimal solution. Taking first the current court system, there are I think four aspects we can identify. First, it is dominated by the national courts where most tax disputes are resolved. I will say more about this in a moment. As a second element, we have the sporadic intervention of the European Court of Justice, generally by preliminary rulings on references from national courts but also on infringement proceedings. If you look at the statistics in the last couple of years about a sixth to a quarter of all judgements from the ECJ deal with some form of tax matter (both direct and indirect). At the present moment, though, the Court of First Instance has virtually no role in tax cases. The third element of the current scenario is the limited use of ad hoc arbitration panels under the Arbitration Convention, which is presently available only for transfer pricing disputes. The fourth element that must not be forgotten is that most cross-border disputes are still resolved under MAP, that is under the mutual agreement procedure under bilateral double taxation conventions. At present, this involves the competent authorities in negotiations, but with the possibility in the future (particularly based on the recent work in the OECD) of more use of binding arbitration to resolve otherwise unresolved disputes under MAP. If we take these four elements and focus on the national courts where most disputes arise, there is absolutely no harmonization in this area at all. There are different types of courts: in some countries tax disputes go before specialist tax tribunals, in others the administrative courts, and in others the general courts. There is a huge variation in the number of levels of appeal: one, two, three, and occasionally four level of appealing tax cases. There are no common procedures. In some countries it is a largely written procedure; in others – like my own – it is largely oral. There are no common rules on evidence or even on the burden of proof. On such a basic point as the issue whether the revenue authority or the taxpayer bears the burden of proof, we have examples at both ends of the spectrum. There are very different cultures in countries concerning tax litigation. In some countries going to court over a tax mat-

182

Philip Baker

ter is very rare, in others it is far more common and it is a regular way of resolving differences between the tax payer and the tax authorities. There are also cultural differences on referring questions to the European Court of Justice. Some countries have never referred any tax issues to the European Court. There is also a huge variation in the time taken to resolve tax cases: but I think tax payers generally would agree on one point, and that is in Europe too long is taken in resolving tax disputes through the national courts. Above all, national courts can only take account of one tax system. They cannot act particularly well with cross-border tax problems when more than one country is involved. This brings me to the second element of my comments: the types of disputes that will arise under CCCTB. By definition, these are all issues where more than one state is concerned, and where the resolution of the dispute will impact on more than one jurisdiction. There are likely to be a variety of different types of dispute. Disputes on the interpretation of the Directive or the Regulation that introduces the CCCTB are the first type of relatively simple dispute: they will have to go to the ECJ for a common interpretation. Far more common will be the second type of dispute: disputes involving purely factual questions or mixed questions of fact and law, involving perhaps issues of accountancy evidence. These will include disputes over: was a particular payment made? What is the value of a particular asset at a particular time? What is the correct accounting treatment of this particular item in a balance sheet? There will also be issues that will arise during the conduct of an audit. For example, is the taxpayer required to supply particular information? Then issues will arise after the audit is completed. For example: when does the taxpayer have to pay the taxes? Are there penalties due? What if the taxpayer doesn’t pay the taxes? You should not necessarily assume that all taxpayers who join the CCCTB system will be good taxpayers, and penalties will arise or non-payment will occur. So what are the possible approaches that might be adopted? I think there are four of them, and I have a very clear preference for one of these options. The least attractive option is the status quo. That is, disputes are referred to the national courts of each Member States. That could easily mean that a company that has elected for the CCCTB finds itself in litigation in more than one Member State. If that happens it would destroy much of the efficiency advantages that CCCTB would otherwise introduce. That would not be an optimal solution in any sense. A second solution might arise from a one-stop shop approach, where the courts of the Member State that is the principal Member State for the group deal with all disputes. Personally, I think that is an unlikely solution. I follow Matthias Mors here, where I think audit procedures are likely to have to be carried out by the different revenue authorities of the country concerned, and to try to bring all disputes back to one country would be unworkable. Suppose that State A is the principal State but the revenue authorities of State B are carrying out an audit of the subsidiary or the branch in that State, and a dispute arises. Is the revenue authority of State B really to come to the courts of State A to try to resolve any disputes over the audit? I don’t think that will be possible.

E. Administrative Aspects

183

A third possibility is an extension of binding arbitration. However, it seems to me that the type of dispute we are considering are not ones that lend themselves to arbitration. They are disputes where a rapid, definitive, traditional form of determination is required. That leads to the fourth solution, for which in my mind the arguments are overwhelming. We need a European Tax Court: a single venue for all disputes to be resolved, so a single jurisdiction can resolve all of the issues. Let me add some brief comments on this. While the European Tax Court may be a single court, it should be able to sit in any Member States. If there are witnesses, or documentary evidence to be considered, it will be much more efficient for the court to sit in the Member State concerned. Secondly, and perhaps as a consequence of this, it will require a panel of tax judges. Probably at least one from each Member State involved in the CCCTB, plus a panel of experts to sit with them. Thirdly – and that is easier said than done – the European Tax Court will have to be extremely efficient. It needs to be capable of resolving issues rapidly. If it is slow, and the apportionment of profits between the Member States waits for it to resolve the issues, it will make the CCCTB unworkable. It will need to be adequately staffed, and be proactive to resolve issues within its jurisdiction rapidly. It is, without doubt, my preferred solution.